Home |
Search |
Today's Posts |
|
Electronic Schematics (alt.binaries.schematics.electronic) A place to show and share your electronics schematic drawings. |
Reply |
|
|
LinkBack | Thread Tools | Display Modes |
#1
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
Here is a circuit that I scribbled on an odd scrap of paper that I would
like to experiment with, but unfortunately I don't have an old valve radio to use as a testbed. The component values that are pencilled in will doubtless need revising and the rest by trial and error. The plan is of course to make a MOSFET substitute for the O/P pentode in an old valve radio, it has 3 stages to maintain its invertingness in case the audio amplifier has a nfb loop. Anyone care to cast a critical eye over it and predict how loud a bang it will make? |
#2
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
On Thu, 29 Apr 2010 16:22:59 +0100, "ian field"
wrote: Here is a circuit that I scribbled on an odd scrap of paper that I would like to experiment with, but unfortunately I don't have an old valve radio to use as a testbed. The component values that are pencilled in will doubtless need revising and the rest by trial and error. The plan is of course to make a MOSFET substitute for the O/P pentode in an old valve radio, it has 3 stages to maintain its invertingness in case the audio amplifier has a nfb loop. Anyone care to cast a critical eye over it and predict how loud a bang it will make? You're making it too hard. Just slip a 2N7000 under your 900V device (making a cascode), tie the 900V device gate to the +30V, and bias the 2N7000 as desired. Personally I've done this where the 900V device was a toob, and, in the cathode circuit was an NPN with an OpAmp wrapped around it ;-) ...Jim Thompson -- | James E.Thompson, CTO | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: Contacts Only | | | Voice480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 | The only thing bipartisan in this country is hypocrisy |
#3
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
"Jim Thompson" wrote in message ... On Thu, 29 Apr 2010 16:22:59 +0100, "ian field" wrote: Here is a circuit that I scribbled on an odd scrap of paper that I would like to experiment with, but unfortunately I don't have an old valve radio to use as a testbed. The component values that are pencilled in will doubtless need revising and the rest by trial and error. The plan is of course to make a MOSFET substitute for the O/P pentode in an old valve radio, it has 3 stages to maintain its invertingness in case the audio amplifier has a nfb loop. Anyone care to cast a critical eye over it and predict how loud a bang it will make? You're making it too hard. Just slip a 2N7000 under your 900V device (making a cascode), tie the 900V device gate to the +30V, and bias the 2N7000 as desired. Personally I've done this where the 900V device was a toob, and, in the cathode circuit was an NPN with an OpAmp wrapped around it ;-) Alternatively I could just use a depletion mode power MOSFET - but none of the suppliers I use have ever heard of the few that are on the market. And that would spoil all the fun of figuring it out with easy to get parts. The main problem I was trying to solve is that designing a self biasing enhancement MOSFET stage for this sort of application can be a bit tricky. |
#4
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
"flipper" wrote in message ... On Thu, 29 Apr 2010 16:22:59 +0100, "ian field" wrote: Here is a circuit that I scribbled on an odd scrap of paper that I would like to experiment with, but unfortunately I don't have an old valve radio to use as a testbed. The component values that are pencilled in will doubtless need revising and the rest by trial and error. The plan is of course to make a MOSFET substitute for the O/P pentode in an old valve radio, it has 3 stages to maintain its invertingness in case the audio amplifier has a nfb loop. Anyone care to cast a critical eye over it and predict how loud a bang it will make? What's the goal? To 'fix' a broken radio? Why not just buy the proper valve and plug it in? To some extent its just a bit of fun. One significant inconvenience is that most UK radios had series heater chains, so a dropper resistor would have to be wired in to replace the valve's heater. |
#5
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
Have ian field wrote:
Here is a circuit that I scribbled on an odd scrap of paper that I would like to experiment with, but unfortunately I don't have an old valve radio to use as a testbed. The component values that are pencilled in will doubtless need revising and the rest by trial and error. The plan is of course to make a MOSFET substitute for the O/P pentode in an old valve radio, it has 3 stages to maintain its invertingness in case the audio amplifier has a nfb loop. Anyone care to cast a critical eye over it and predict how loud a bang it will make? Have those tube replacements been on the (military) market?? Or does my memory play tricks on me? Google: fet tube replacements ..... 1000.000 hits. Rescued by google. |
#6
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
On Thu, 29 Apr 2010 17:04:28 +0100, "ian field"
wrote: "Jim Thompson" wrote in message ... On Thu, 29 Apr 2010 16:22:59 +0100, "ian field" wrote: Here is a circuit that I scribbled on an odd scrap of paper that I would like to experiment with, but unfortunately I don't have an old valve radio to use as a testbed. The component values that are pencilled in will doubtless need revising and the rest by trial and error. The plan is of course to make a MOSFET substitute for the O/P pentode in an old valve radio, it has 3 stages to maintain its invertingness in case the audio amplifier has a nfb loop. Anyone care to cast a critical eye over it and predict how loud a bang it will make? You're making it too hard. Just slip a 2N7000 under your 900V device (making a cascode), tie the 900V device gate to the +30V, and bias the 2N7000 as desired. Personally I've done this where the 900V device was a toob, and, in the cathode circuit was an NPN with an OpAmp wrapped around it ;-) Alternatively I could just use a depletion mode power MOSFET - but none of the suppliers I use have ever heard of the few that are on the market. And that would spoil all the fun of figuring it out with easy to get parts. The main problem I was trying to solve is that designing a self biasing enhancement MOSFET stage for this sort of application can be a bit tricky. You could bias the gate up, to +8 maybe, and put a source resistor to ground, bypassed maybe. 1 volt p-p of AC coupled gate drive would be in the ballpark. Mosfets have higher transconductances than toobs ever dreamed of, so you'll have lots of gain. John |
#7
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
"flipper" wrote in message ... On Thu, 29 Apr 2010 18:50:04 +0100, "ian field" wrote: "flipper" wrote in message . .. On Thu, 29 Apr 2010 16:22:59 +0100, "ian field" wrote: Here is a circuit that I scribbled on an odd scrap of paper that I would like to experiment with, but unfortunately I don't have an old valve radio to use as a testbed. The component values that are pencilled in will doubtless need revising and the rest by trial and error. The plan is of course to make a MOSFET substitute for the O/P pentode in an old valve radio, it has 3 stages to maintain its invertingness in case the audio amplifier has a nfb loop. Anyone care to cast a critical eye over it and predict how loud a bang it will make? What's the goal? To 'fix' a broken radio? Why not just buy the proper valve and plug it in? To some extent its just a bit of fun. One significant inconvenience is that most UK radios had series heater chains, so a dropper resistor would have to be wired in to replace the valve's heater. Here's a quick throw together idea. What's the typical anode current on an average table top set? |
#8
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
"flipper" wrote in message ... On Fri, 30 Apr 2010 22:22:03 +0100, "ian field" wrote: "flipper" wrote in message . .. On Thu, 29 Apr 2010 18:50:04 +0100, "ian field" wrote: "flipper" wrote in message m... On Thu, 29 Apr 2010 16:22:59 +0100, "ian field" wrote: Here is a circuit that I scribbled on an odd scrap of paper that I would like to experiment with, but unfortunately I don't have an old valve radio to use as a testbed. The component values that are pencilled in will doubtless need revising and the rest by trial and error. The plan is of course to make a MOSFET substitute for the O/P pentode in an old valve radio, it has 3 stages to maintain its invertingness in case the audio amplifier has a nfb loop. Anyone care to cast a critical eye over it and predict how loud a bang it will make? What's the goal? To 'fix' a broken radio? Why not just buy the proper valve and plug it in? To some extent its just a bit of fun. One significant inconvenience is that most UK radios had series heater chains, so a dropper resistor would have to be wired in to replace the valve's heater. Here's a quick throw together idea. What's the typical anode current on an average table top set? I don't work on them so I'm not sure but I would imagine it's close to the 'typical operation' given in the respective tube datasheet. Like, for the 35L6 (octal AA5 output, 150mA heaters, circa 1939), it's given as 40mA (idle) at 110V (tube rectified, typically a 35Z4, U.S. AC line Last night I had a flick through my old Mazda valve data book and the popular audio output valves range from Ia about 30 to 70mA, the BF246B would be good enough for all but the most powerful types but the 246C would cover almost any eventuality. The best I have in the parts drawer is BF245C, I'd have to select for highest Idss and still gang at least 2 devices - and tweak the individual decoupled source resistors to balance the dissipation. |
#9
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
"flipper" wrote in message ... On Sat, 1 May 2010 13:53:37 +0100, "ian field" wrote: "flipper" wrote in message . .. On Fri, 30 Apr 2010 22:22:03 +0100, "ian field" wrote: "flipper" wrote in message m... On Thu, 29 Apr 2010 18:50:04 +0100, "ian field" wrote: "flipper" wrote in message news:bedjt5t8ds9vopi85fvebp8us54h90835c@4ax. com... On Thu, 29 Apr 2010 16:22:59 +0100, "ian field" wrote: Here is a circuit that I scribbled on an odd scrap of paper that I would like to experiment with, but unfortunately I don't have an old valve radio to use as a testbed. The component values that are pencilled in will doubtless need revising and the rest by trial and error. The plan is of course to make a MOSFET substitute for the O/P pentode in an old valve radio, it has 3 stages to maintain its invertingness in case the audio amplifier has a nfb loop. Anyone care to cast a critical eye over it and predict how loud a bang it will make? What's the goal? To 'fix' a broken radio? Why not just buy the proper valve and plug it in? To some extent its just a bit of fun. One significant inconvenience is that most UK radios had series heater chains, so a dropper resistor would have to be wired in to replace the valve's heater. Here's a quick throw together idea. What's the typical anode current on an average table top set? I don't work on them so I'm not sure but I would imagine it's close to the 'typical operation' given in the respective tube datasheet. Like, for the 35L6 (octal AA5 output, 150mA heaters, circa 1939), it's given as 40mA (idle) at 110V (tube rectified, typically a 35Z4, U.S. AC line Last night I had a flick through my old Mazda valve data book and the popular audio output valves range from Ia about 30 to 70mA, the BF246B would be good enough for all but the most powerful types but the 246C would cover almost any eventuality. The best I have in the parts drawer is BF245C, I'd have to select for highest Idss and still gang at least 2 devices - and tweak the individual decoupled source resistors to balance the dissipation. Are you trying to just make a test device or a 'universal device' of some sort? I didn't mean to suggest the BF246C was the best thing to use. It's just the best ready made jfet model I had (and did seem 'good enough' for a test shot anyway. Btw, 'good enough' was before I realized the zener needed to be increased.) My opinion is you need one with comparatively high Vgs so the source resistor is large enough for some feedback (as well as a comparable input voltage range, depending on how everything else turns out). A j105 looks interesting with 500mA max but it's TO-92 again. Any in that series; J105 - 107 should do, but from Farnell they're a tad pricey, it seems dissipation may be an issue so ganged BF245C or 246C might be the better solution. |
#10
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
"flipper" wrote in message news On Sat, 1 May 2010 20:38:32 +0100, "ian field" wrote: "flipper" wrote in message . .. On Sat, 1 May 2010 13:53:37 +0100, "ian field" wrote: "flipper" wrote in message m... On Fri, 30 Apr 2010 22:22:03 +0100, "ian field" wrote: "flipper" wrote in message news:q68kt5l5noh0o5pupj3uhcvbdcq7u0bavi@4ax. com... On Thu, 29 Apr 2010 18:50:04 +0100, "ian field" wrote: "flipper" wrote in message news:bedjt5t8ds9vopi85fvebp8us54h90835c@4a x.com... On Thu, 29 Apr 2010 16:22:59 +0100, "ian field" wrote: Here is a circuit that I scribbled on an odd scrap of paper that I would like to experiment with, but unfortunately I don't have an old valve radio to use as a testbed. The component values that are pencilled in will doubtless need revising and the rest by trial and error. The plan is of course to make a MOSFET substitute for the O/P pentode in an old valve radio, it has 3 stages to maintain its invertingness in case the audio amplifier has a nfb loop. Anyone care to cast a critical eye over it and predict how loud a bang it will make? What's the goal? To 'fix' a broken radio? Why not just buy the proper valve and plug it in? To some extent its just a bit of fun. One significant inconvenience is that most UK radios had series heater chains, so a dropper resistor would have to be wired in to replace the valve's heater. Here's a quick throw together idea. What's the typical anode current on an average table top set? I don't work on them so I'm not sure but I would imagine it's close to the 'typical operation' given in the respective tube datasheet. Like, for the 35L6 (octal AA5 output, 150mA heaters, circa 1939), it's given as 40mA (idle) at 110V (tube rectified, typically a 35Z4, U.S. AC line Last night I had a flick through my old Mazda valve data book and the popular audio output valves range from Ia about 30 to 70mA, the BF246B would be good enough for all but the most powerful types but the 246C would cover almost any eventuality. The best I have in the parts drawer is BF245C, I'd have to select for highest Idss and still gang at least 2 devices - and tweak the individual decoupled source resistors to balance the dissipation. Are you trying to just make a test device or a 'universal device' of some sort? I didn't mean to suggest the BF246C was the best thing to use. It's just the best ready made jfet model I had (and did seem 'good enough' for a test shot anyway. Btw, 'good enough' was before I realized the zener needed to be increased.) My opinion is you need one with comparatively high Vgs so the source resistor is large enough for some feedback (as well as a comparable input voltage range, depending on how everything else turns out). A j105 looks interesting with 500mA max but it's TO-92 again. Any in that series; J105 - 107 should do, but from Farnell they're a tad pricey, it seems dissipation may be an issue so ganged BF245C or 246C might be the better solution. Well, we'll just have to fix that power thingie, then. Try this idea on for size. 4mW in the jfet and idle is about 660 mW in the MJE340. Btw, that IRFU410 is there simply by reflex as it's a close spice sub for the 600V STP2NK60Z I like (built in gate protection and [used to be] cheap). Same with the MJE340. Certainly don't need a 20 W (well, with heatsink) 300V transistor around a 25 V jfet but it'll sure as hell survive. It would still need some heatsinking because they're going into a hot environment but so will the MOSFET so you could use one and put 'em both on it. You could maybe build it on an octal base but it isn't a pin for pin straight 'plug in' kind of thing like a fetron because, for one, it needs an extra ground pin for the zener reference (could put the plus end on 'screen'). Also, existing Rk wouldn't work. Btw, adding the power bypass ends any chance of faking a soft clip as follower gain smashes any low Vds 'triode region' in the jfet flat as a pancake. Thanks - that looks about it, but it won't please the critics who commented my original 3 MOSFET contraption was too difficult. |
#11
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
On Thu, 29 Apr 2010 11:33:10 -0500, flipper wrote:
On Thu, 29 Apr 2010 08:41:52 -0700, Jim Thompson wrote: On Thu, 29 Apr 2010 16:22:59 +0100, "ian field" wrote: Here is a circuit that I scribbled on an odd scrap of paper that I would like to experiment with, but unfortunately I don't have an old valve radio to use as a testbed. The component values that are pencilled in will doubtless need revising and the rest by trial and error. The plan is of course to make a MOSFET substitute for the O/P pentode in an old valve radio, it has 3 stages to maintain its invertingness in case the audio amplifier has a nfb loop. Anyone care to cast a critical eye over it and predict how loud a bang it will make? You're making it too hard. Just slip a 2N7000 under your 900V device (making a cascode), tie the 900V device gate to the +30V, and bias the 2N7000 as desired. Personally I've done this where the 900V device was a toob, and, in the cathode circuit was an NPN with an OpAmp wrapped around it ;-) Well, in the toob world there are those who would say the same thing, that "You're making it too hard," since the toob will self bias with a simple resistor (audio bypassed for maximum gain) under the cathode, grid to ground, similar to what is done with a simple jfet current source. Except for when it doesn't work, to bias the final / power output stage. |
#12
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
"flipper" wrote in message ... On Sat, 1 May 2010 22:42:45 +0100, "ian field" wrote: "flipper" wrote in message news On Sat, 1 May 2010 20:38:32 +0100, "ian field" wrote: "flipper" wrote in message m... On Sat, 1 May 2010 13:53:37 +0100, "ian field" wrote: "flipper" wrote in message news:2b8nt593rafnfocp63ceg5g5f8steq8l6r@4ax. com... On Fri, 30 Apr 2010 22:22:03 +0100, "ian field" wrote: "flipper" wrote in message news:q68kt5l5noh0o5pupj3uhcvbdcq7u0bavi@4a x.com... On Thu, 29 Apr 2010 18:50:04 +0100, "ian field" wrote: "flipper" wrote in message news:bedjt5t8ds9vopi85fvebp8us54h90835c@ 4ax.com... On Thu, 29 Apr 2010 16:22:59 +0100, "ian field" wrote: Here is a circuit that I scribbled on an odd scrap of paper that I would like to experiment with, but unfortunately I don't have an old valve radio to use as a testbed. The component values that are pencilled in will doubtless need revising and the rest by trial and error. The plan is of course to make a MOSFET substitute for the O/P pentode in an old valve radio, it has 3 stages to maintain its invertingness in case the audio amplifier has a nfb loop. Anyone care to cast a critical eye over it and predict how loud a bang it will make? What's the goal? To 'fix' a broken radio? Why not just buy the proper valve and plug it in? To some extent its just a bit of fun. One significant inconvenience is that most UK radios had series heater chains, so a dropper resistor would have to be wired in to replace the valve's heater. Here's a quick throw together idea. What's the typical anode current on an average table top set? I don't work on them so I'm not sure but I would imagine it's close to the 'typical operation' given in the respective tube datasheet. Like, for the 35L6 (octal AA5 output, 150mA heaters, circa 1939), it's given as 40mA (idle) at 110V (tube rectified, typically a 35Z4, U.S. AC line Last night I had a flick through my old Mazda valve data book and the popular audio output valves range from Ia about 30 to 70mA, the BF246B would be good enough for all but the most powerful types but the 246C would cover almost any eventuality. The best I have in the parts drawer is BF245C, I'd have to select for highest Idss and still gang at least 2 devices - and tweak the individual decoupled source resistors to balance the dissipation. Are you trying to just make a test device or a 'universal device' of some sort? I didn't mean to suggest the BF246C was the best thing to use. It's just the best ready made jfet model I had (and did seem 'good enough' for a test shot anyway. Btw, 'good enough' was before I realized the zener needed to be increased.) My opinion is you need one with comparatively high Vgs so the source resistor is large enough for some feedback (as well as a comparable input voltage range, depending on how everything else turns out). A j105 looks interesting with 500mA max but it's TO-92 again. Any in that series; J105 - 107 should do, but from Farnell they're a tad pricey, it seems dissipation may be an issue so ganged BF245C or 246C might be the better solution. Well, we'll just have to fix that power thingie, then. Try this idea on for size. 4mW in the jfet and idle is about 660 mW in the MJE340. Btw, that IRFU410 is there simply by reflex as it's a close spice sub for the 600V STP2NK60Z I like (built in gate protection and [used to be] cheap). Same with the MJE340. Certainly don't need a 20 W (well, with heatsink) 300V transistor around a 25 V jfet but it'll sure as hell survive. It would still need some heatsinking because they're going into a hot environment but so will the MOSFET so you could use one and put 'em both on it. You could maybe build it on an octal base but it isn't a pin for pin straight 'plug in' kind of thing like a fetron because, for one, it needs an extra ground pin for the zener reference (could put the plus end on 'screen'). Also, existing Rk wouldn't work. Btw, adding the power bypass ends any chance of faking a soft clip as follower gain smashes any low Vds 'triode region' in the jfet flat as a pancake. Thanks - that looks about it, but it won't please the critics who commented my original 3 MOSFET contraption was too difficult. Well, maybe not. Not knowing your thinking I couldn't quite figure out how you intended yours to work so I simply fell back to what I'd done before. Maybe yours would be a gentler clip. I dunno. The clipping characteristics would be more of a concern if it was intended for use as a guitar amp, it is hoped that the increased gain and current handling will mean it never goes anywhere near clipping. BTW: why does the bipolar need to be an MJE340 ? |
#13
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
On Sun, 02 May 2010 06:56:01 -0700,
wrote: On Thu, 29 Apr 2010 11:33:10 -0500, flipper wrote: On Thu, 29 Apr 2010 08:41:52 -0700, Jim Thompson wrote: On Thu, 29 Apr 2010 16:22:59 +0100, "ian field" wrote: Here is a circuit that I scribbled on an odd scrap of paper that I would like to experiment with, but unfortunately I don't have an old valve radio to use as a testbed. The component values that are pencilled in will doubtless need revising and the rest by trial and error. The plan is of course to make a MOSFET substitute for the O/P pentode in an old valve radio, it has 3 stages to maintain its invertingness in case the audio amplifier has a nfb loop. Anyone care to cast a critical eye over it and predict how loud a bang it will make? You're making it too hard. Just slip a 2N7000 under your 900V device (making a cascode), tie the 900V device gate to the +30V, and bias the 2N7000 as desired. Personally I've done this where the 900V device was a toob, and, in the cathode circuit was an NPN with an OpAmp wrapped around it ;-) Well, in the toob world there are those who would say the same thing, that "You're making it too hard," since the toob will self bias with a simple resistor (audio bypassed for maximum gain) under the cathode, grid to ground, similar to what is done with a simple jfet current source. Except for when it doesn't work, to bias the final / power output stage. My "toob+tranny" was incredibly linear as compared to a toob alone. You can even make a triode behave with the linearity of a pentode... handy in HV conditions. ...Jim Thompson -- | James E.Thompson, CTO | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: Contacts Only | | | Voice480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 | The only thing bipartisan in this country is hypocrisy |
#14
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
Mine saturates to a lower voltage.
Tim -- Deep Friar: a very philosophical monk. Website: http://webpages.charter.net/dawill/tmoranwms "flipper" wrote in message ... On Sun, 2 May 2010 15:20:11 +0100, "ian field" wrote: trim snip |
#15
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
On Sun, 02 May 2010 14:48:19 -0500, flipper wrote:
On Sun, 2 May 2010 15:20:11 +0100, "ian field" wrote: trim snip The clipping characteristics would be more of a concern if it was intended for use as a guitar amp, it is hoped that the increased gain and current handling will mean it never goes anywhere near clipping. I thought you were talking about a radio output stage substitute. If so, then you're constrained by everything else already in the radio: the OPT, B+, and so on. Neither gain nor 'more current' is going to make much of a, if any, difference because you have an already defined B+ into an already defined output impedance. You can't swing more voltage across it than is there and with no more voltage you swing no more current. Increased idle current would likely affect the B+ filter and the preamp stages already expect a certain output gain so I'd think the goal would be to 'match' the designed for tube as close as practical. BTW: why does the bipolar need to be an MJE340 ? I said it was there simply by reflex and probably overkill. I wasn't trying to optimize it, just demonstrate the idea. The MJE340 is a nice 'all around' HV transistor and I do use where the 300V is needed, so that puts it in my parts bin. 300V also makes it robust in 'low voltage' circuits under fault conditions, like say a tube short, but, as I mentioned, that's probably moot in this circuit. Hmm, speaking of robust, it might not be a bad idea to put a reverse protection diode in there. Anyway, it's just a transistor I probably don't have to 'worry much about' so slap it in for testing, since I'll have one in the bin (as well as the STP2NK60Z) to breadboard with, and optimize things later, etc. Feel free to juggle things around to suit your needs. Which reminds me, splitting R4 into two, mentioned in my last post, also widens the range of jfets you can use since Vgs can be catered for by adjusting the 'gate boost' ratio. I mean, the main reason I suggested a large Vgs device was to get some source R under it for feedback (reducing gain & distortion) and boosting Vgate accomplishes the same thing for a low Vgs device: you increase the source resistor (actually the one under the MJE340 in this case) to counter the 'boost'. Should be able to tailor it to almost any desired gain by selecting appropriate jfets and suitable 'boost' ratio/source (emitter) resistor. Also acts as an input impedance partial boostrap. How's that for serendipity? The R4 split might also be the best place to put a 'bias adjust' trimmer. Once set it shouldn't need readjusting so that doesn't need to be user accessible. I kinda like the 'bias boost' idea because it keeps the bias trimmer out of high current and makes for quite a bit of flexibility. (attachment) I didn't look at Vgs variability in picking that adjustment range. Again, it's 'for show'. The BF244 is lower Vgs than the previous BF246 but with 'bias boost' we end up with a 20V zener again, equivalent source resistor feedback, and similar gain (clip point). Also note we don't need high current jfets anymore because the NPN is doing that work. Btw, that could be a (MOS)FET too but it's Vgs would increase the reference voltage needed, cutting into output swing. That's why I went bipolar. Something I'd keep in mind when testing is potential noise from that zener on the gate. I'm gambling it's small enough to not matter on an output stage but if it's a problem then filter it or create a voltage reference by other means. One might be tempted to use a simple resistor divider but, while I haven't done an analysis, I suspect component tolerances and B+ variation are likely too much to ensure things work properly on the low end while staying safe for the jfet on the high end, plus that's a hum injection point, hence the zener for a quickie first shot. Maybe we should patent this thing Why not just do this? No adjustments needed. John |
#16
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
John Larkin wrote:
On Sun, 02 May 2010 14:48:19 -0500, flipper wrote: On Sun, 2 May 2010 15:20:11 +0100, "ian field" wrote: trim snip The clipping characteristics would be more of a concern if it was intended for use as a guitar amp, it is hoped that the increased gain and current cut Maybe we should patent this thing Why not just do this? No adjustments needed. John A 400?? v vcr protecting the fet and you have a dandy replacement for a tube!! Removing the speaker while working might kill the fet. |
#17
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
On Sun, 02 May 2010 18:39:11 -0500, flipper wrote:
On Sun, 02 May 2010 14:59:19 -0700, John Larkin wrote: On Sun, 02 May 2010 14:48:19 -0500, flipper wrote: On Sun, 2 May 2010 15:20:11 +0100, "ian field" wrote: trim snip The clipping characteristics would be more of a concern if it was intended for use as a guitar amp, it is hoped that the increased gain and current handling will mean it never goes anywhere near clipping. I thought you were talking about a radio output stage substitute. If so, then you're constrained by everything else already in the radio: the OPT, B+, and so on. Neither gain nor 'more current' is going to make much of a, if any, difference because you have an already defined B+ into an already defined output impedance. You can't swing more voltage across it than is there and with no more voltage you swing no more current. Increased idle current would likely affect the B+ filter and the preamp stages already expect a certain output gain so I'd think the goal would be to 'match' the designed for tube as close as practical. BTW: why does the bipolar need to be an MJE340 ? I said it was there simply by reflex and probably overkill. I wasn't trying to optimize it, just demonstrate the idea. The MJE340 is a nice 'all around' HV transistor and I do use where the 300V is needed, so that puts it in my parts bin. 300V also makes it robust in 'low voltage' circuits under fault conditions, like say a tube short, but, as I mentioned, that's probably moot in this circuit. Hmm, speaking of robust, it might not be a bad idea to put a reverse protection diode in there. Anyway, it's just a transistor I probably don't have to 'worry much about' so slap it in for testing, since I'll have one in the bin (as well as the STP2NK60Z) to breadboard with, and optimize things later, etc. Feel free to juggle things around to suit your needs. Which reminds me, splitting R4 into two, mentioned in my last post, also widens the range of jfets you can use since Vgs can be catered for by adjusting the 'gate boost' ratio. I mean, the main reason I suggested a large Vgs device was to get some source R under it for feedback (reducing gain & distortion) and boosting Vgate accomplishes the same thing for a low Vgs device: you increase the source resistor (actually the one under the MJE340 in this case) to counter the 'boost'. Should be able to tailor it to almost any desired gain by selecting appropriate jfets and suitable 'boost' ratio/source (emitter) resistor. Also acts as an input impedance partial boostrap. How's that for serendipity? The R4 split might also be the best place to put a 'bias adjust' trimmer. Once set it shouldn't need readjusting so that doesn't need to be user accessible. I kinda like the 'bias boost' idea because it keeps the bias trimmer out of high current and makes for quite a bit of flexibility. (attachment) I didn't look at Vgs variability in picking that adjustment range. Again, it's 'for show'. The BF244 is lower Vgs than the previous BF246 but with 'bias boost' we end up with a 20V zener again, equivalent source resistor feedback, and similar gain (clip point). Also note we don't need high current jfets anymore because the NPN is doing that work. Btw, that could be a (MOS)FET too but it's Vgs would increase the reference voltage needed, cutting into output swing. That's why I went bipolar. Something I'd keep in mind when testing is potential noise from that zener on the gate. I'm gambling it's small enough to not matter on an output stage but if it's a problem then filter it or create a voltage reference by other means. One might be tempted to use a simple resistor divider but, while I haven't done an analysis, I suspect component tolerances and B+ variation are likely too much to ensure things work properly on the low end while staying safe for the jfet on the high end, plus that's a hum injection point, hence the zener for a quickie first shot. Maybe we should patent this thing Why not just do this? No adjustments needed. John I dunno. Give me some component values and we'll see. Make the upper resistor whatever it needs to be to get +10 on the gate. Make the source resistor whatever you like to get your preferred idle current. Just off hand, the 100k makes input impedance rather low and if that's a MOSFET then gate capacitance is a problem. Depends on what you drive it from. 100K is not going to load many driver circuits. If you delete the source bypass cap, Cin goes way down. At that point, drive levels will be closer to what a toob would need, and linearity is better. Of course, drain impedance is high, so speaker damping is poor. Some people like that. Local NFB is easy (just connect the upper resistor to the drain) and helps that situation. I was just showing a bias scheme, not a full amplifier. John |
#18
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
On Sun, 02 May 2010 23:43:14 -0500, flipper wrote:
On Sun, 02 May 2010 17:08:30 -0700, John Larkin wrote: On Sun, 02 May 2010 18:39:11 -0500, flipper wrote: On Sun, 02 May 2010 14:59:19 -0700, John Larkin wrote: snip Why not just do this? No adjustments needed. John I dunno. Give me some component values and we'll see. Make the upper resistor whatever it needs to be to get +10 on the gate. Make the source resistor whatever you like to get your preferred idle current. Actually, I was wondering if you had a magic FET but that's now moot. Just off hand, the 100k makes input impedance rather low and if that's a MOSFET then gate capacitance is a problem. Depends on what you drive it from. 100K is not going to load many driver circuits. It's not all that great in the tube world and AA5 radios normally have 470k to 1M grid leaks but that's solvable. The trick here is to make a separate reference, not tied to the gate, and then take gate to it through an equivalent 470k to 1M 'gate leak'. If you delete the source bypass cap, Cin goes way down. At that point, drive levels will be closer to what a toob would need, and linearity is better. First thing I looked at was just like what you drew but bandwidth stunk so I moved on. Source feedback is the trick to it and looks like there's enough to make it work. Of course, drain impedance is high, so speaker damping is poor. Some people like that. Local NFB is easy (just connect the upper resistor to the drain) and helps that situation. I think adding some local drain feedback might be a little more complicated than that because bias would go away with large output swings. Anyway, a 'plain Jane version might look something like the attachment. A zener, rather than the resistor divider, keeps hum off the gate. Or one could filter the divider. [snip] What will Cdg feedback do to an uncascoded circuit such as yours? Limit the high end? ...Jim Thompson -- | James E.Thompson, CTO | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: Contacts Only | | | Voice480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 | The only thing bipartisan in this country is hypocrisy |
#19
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
"flipper" wrote in message ... On Sun, 2 May 2010 15:20:11 +0100, "ian field" wrote: trim snip The clipping characteristics would be more of a concern if it was intended for use as a guitar amp, it is hoped that the increased gain and current handling will mean it never goes anywhere near clipping. I thought you were talking about a radio output stage substitute. If so, then you're constrained by everything else already in the radio: the OPT, B+, and so on. Neither gain nor 'more current' is going to make much of a, if any, difference because you have an already defined B+ into an already defined output impedance. You can't swing more voltage across it than is there and with no more voltage you swing no more current. Increased idle current would likely affect the B+ filter and the preamp stages already expect a certain output gain so I'd think the goal would be to 'match' the designed for tube as close as practical. BTW: why does the bipolar need to be an MJE340 ? I said it was there simply by reflex and probably overkill. I wasn't trying to optimize it, just demonstrate the idea. The MJE340 is a nice 'all around' HV transistor and I do use where the 300V is needed, so that puts it in my parts bin. 300V also makes it robust in 'low voltage' circuits under fault conditions, like say a tube short, but, as I mentioned, that's probably moot in this circuit. Hmm, speaking of robust, it might not be a bad idea to put a reverse protection diode in there. Anyway, it's just a transistor I probably don't have to 'worry much about' so slap it in for testing, since I'll have one in the bin (as well as the STP2NK60Z) to breadboard with, and optimize things later, etc. Feel free to juggle things around to suit your needs. Which reminds me, splitting R4 into two, mentioned in my last post, also widens the range of jfets you can use since Vgs can be catered for by adjusting the 'gate boost' ratio. I mean, the main reason I suggested a large Vgs device was to get some source R under it for feedback (reducing gain & distortion) and boosting Vgate accomplishes the same thing for a low Vgs device: you increase the source resistor (actually the one under the MJE340 in this case) to counter the 'boost'. Should be able to tailor it to almost any desired gain by selecting appropriate jfets and suitable 'boost' ratio/source (emitter) resistor. Also acts as an input impedance partial boostrap. How's that for serendipity? The R4 split might also be the best place to put a 'bias adjust' trimmer. Once set it shouldn't need readjusting so that doesn't need to be user accessible. I kinda like the 'bias boost' idea because it keeps the bias trimmer out of high current and makes for quite a bit of flexibility. (attachment) I didn't look at Vgs variability in picking that adjustment range. Again, it's 'for show'. The BF244 is lower Vgs than the previous BF246 but with 'bias boost' we end up with a 20V zener again, equivalent source resistor feedback, and similar gain (clip point). Also note we don't need high current jfets anymore because the NPN is doing that work. Btw, that could be a (MOS)FET too but it's Vgs would increase the reference voltage needed, cutting into output swing. That's why I went bipolar. Something I'd keep in mind when testing is potential noise from that zener on the gate. I'm gambling it's small enough to not matter on an output stage but if it's a problem then filter it or create a voltage reference by other means. One might be tempted to use a simple resistor divider but, while I haven't done an analysis, I suspect component tolerances and B+ variation are likely too much to ensure things work properly on the low end while staying safe for the jfet on the high end, plus that's a hum injection point, hence the zener for a quickie first shot. Maybe we should patent this thing One thing I was wondering, can the JFET have a much lower Idss since its buffered with a bipolar? |
#20
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
"flipper" wrote in message
... Anyone know if this is an 'inherent' relationship: that as Gfs increases so does gate capacitance sort of correspondingly so we're fighting a loosing battle trying to trade one for the other? 1 / (2*pi*R*C) is the cutoff frequency. Put in R = 1/Gfs and you get something like fT. This isn't so much a physically realizable frequency as it is a quality factor (tubes were similarly rated in terms of Gm vs. C). 1 mho is an awful lot of transconductance. At 170pF, that's 936MHz. The equivalent transresistance is quite small (= 1 ohm), which suggests actual resistances (like gate spreading resistance) or inductances (figure 5nH typical source inductance) will dominate. (Ls is 29 ohms at that frequency, way past cutoff, and Rg might be 1 ohm or so. Cg resonates with Ls at 173MHz, which would be series resonant with the input, but tweaked by drain current, giving rise to funny phase shifts as seen in the s-parameters.) Tim -- Deep Friar: a very philosophical monk. Website: http://webpages.charter.net/dawill/tmoranwms |
#21
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
"flipper" wrote in message
... It's source resistance into gate capacitance that's the problem, not Rds on. Source impedance is approximately 1/Gm ;-) Tim -- Deep Friar: a very philosophical monk. Website: http://webpages.charter.net/dawill/tmoranwms |
#22
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
"flipper" wrote in message
... So you really don't think it matters whether one tries to drive the gate through a 10 ohm or 100 megohm resistor? I said cutoff frequency, which means voltage source. Do you have a 0 ohm source at ~1GHz? No? Then you won't get to see fT. ;-) Tim -- Deep Friar: a very philosophical monk. Website: http://webpages.charter.net/dawill/tmoranwms |
#23
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
On Tue, 04 May 2010 01:03:43 -0500, flipper wrote:
On Mon, 3 May 2010 23:02:27 -0500, "Tim Williams" wrote: "flipper" wrote in message . .. So you really don't think it matters whether one tries to drive the gate through a 10 ohm or 100 megohm resistor? I said cutoff frequency, which means voltage source. Do you have a 0 ohm source at ~1GHz? No? Then you won't get to see fT. ;-) Tim I know we won't see it so I have no idea what the heck your tap dance is for. Gm/Co was a popular figure of merit for tubes used in video amps and oscilloscopes. They took a hockey-stick turn upward when frame-grid tubes were invented, just before semiconductors took over. Mosfets have impressive Gm/Co values, at the cost of high input capacitance. John |
#24
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
"John Larkin" wrote in message
... Gm/Co was a popular figure of merit for tubes used in video amps and oscilloscopes. They took a hockey-stick turn upward when frame-grid tubes were invented, just before semiconductors took over. 7KY6 has four times the transconductance of 2N3819, and plenty more voltage and power capacity. Tim -- Deep Friar: a very philosophical monk. Website: http://webpages.charter.net/dawill/tmoranwms |
#25
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
On Tue, 4 May 2010 14:31:11 -0500, "Tim Williams"
wrote: "John Larkin" wrote in message .. . Gm/Co was a popular figure of merit for tubes used in video amps and oscilloscopes. They took a hockey-stick turn upward when frame-grid tubes were invented, just before semiconductors took over. 7KY6 has four times the transconductance of 2N3819, and plenty more voltage and power capacity. Tim Sure, but a 3819 is a dog. Jfets are wimpy, although a BF862 has a bit more Gm than the 7KY6. For serious transconductance, phemts and GaN fets are the winners, with whole digits of siemens to the left of the decimal and absurdly low capacitances. John |
#26
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
"flipper" wrote in message ... On Tue, 04 May 2010 09:11:47 -0700, John Larkin wrote: On Tue, 04 May 2010 01:03:43 -0500, flipper wrote: On Mon, 3 May 2010 23:02:27 -0500, "Tim Williams" wrote: "flipper" wrote in message m... So you really don't think it matters whether one tries to drive the gate through a 10 ohm or 100 megohm resistor? I said cutoff frequency, which means voltage source. Do you have a 0 ohm source at ~1GHz? No? Then you won't get to see fT. ;-) Tim I know we won't see it so I have no idea what the heck your tap dance is for. Gm/Co was a popular figure of merit for tubes used in video amps and oscilloscopes. They took a hockey-stick turn upward when frame-grid tubes were invented, just before semiconductors took over. Mosfets have impressive Gm/Co values, at the cost of high input capacitance. John I understand but that doesn't answer the question at hand because we do not have unlimited gate drive capability. FOM, Rdson x Qg, isn't the issue as the device is 'capable' of more bandwidth if we could drive it but, because the 'driver' already exists, we do not have that choice. At least not without buffering it with something and if we do that we might as well use the jfet under the source version. But, in either case, it's no longer 'plain Jane'. For a given MOSFET technology FOM is roughly constant and I was asking if the same kind of thing applied to our problem of trading off gm, with source feedback, vs Ciss, because that is what the driving impedance works into. Unfortunately, Mouser doesn't let me search by Ciss, or gate charge either, so it's not a simple task to find a low one but I seem to remember that when I originally latched onto the STP2NK60Z it was 'pretty low'... but I wasn't taking gm into account. As I said, it's probably moot for a (AA5) 'table radio', because we don't need much bandwidth to begin with, but I was pondering it's potential suitability as a more general replacement. I suspect we need the more complex version if one wants 'full capability'. Just a thought, but if the common source section of the cascode is boosted by a bipolar, then a really low Idss FET can be used - like a 40673 or a BF991, even with both gates tied together its only a few pF. Being depletion MOSFETs, they should be even better. |
#27
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
"flipper" wrote in message ... On Tue, 4 May 2010 21:20:41 +0100, "ian field" wrote: "flipper" wrote in message . .. On Tue, 04 May 2010 09:11:47 -0700, John Larkin wrote: On Tue, 04 May 2010 01:03:43 -0500, flipper wrote: On Mon, 3 May 2010 23:02:27 -0500, "Tim Williams" wrote: "flipper" wrote in message news:1k4vt5pctqpqush7g57aappqgt1e4ooubd@4ax. com... So you really don't think it matters whether one tries to drive the gate through a 10 ohm or 100 megohm resistor? I said cutoff frequency, which means voltage source. Do you have a 0 ohm source at ~1GHz? No? Then you won't get to see fT. ;-) Tim I know we won't see it so I have no idea what the heck your tap dance is for. Gm/Co was a popular figure of merit for tubes used in video amps and oscilloscopes. They took a hockey-stick turn upward when frame-grid tubes were invented, just before semiconductors took over. Mosfets have impressive Gm/Co values, at the cost of high input capacitance. John I understand but that doesn't answer the question at hand because we do not have unlimited gate drive capability. FOM, Rdson x Qg, isn't the issue as the device is 'capable' of more bandwidth if we could drive it but, because the 'driver' already exists, we do not have that choice. At least not without buffering it with something and if we do that we might as well use the jfet under the source version. But, in either case, it's no longer 'plain Jane'. For a given MOSFET technology FOM is roughly constant and I was asking if the same kind of thing applied to our problem of trading off gm, with source feedback, vs Ciss, because that is what the driving impedance works into. Unfortunately, Mouser doesn't let me search by Ciss, or gate charge either, so it's not a simple task to find a low one but I seem to remember that when I originally latched onto the STP2NK60Z it was 'pretty low'... but I wasn't taking gm into account. As I said, it's probably moot for a (AA5) 'table radio', because we don't need much bandwidth to begin with, but I was pondering it's potential suitability as a more general replacement. I suspect we need the more complex version if one wants 'full capability'. Just a thought, but if the common source section of the cascode is boosted by a bipolar, then a really low Idss FET can be used - like a 40673 or a BF991, even with both gates tied together its only a few pF. Sure, you can. A 'typical' (if there is such a thing) tube in this power range looks to be about 11 or 12 pF and the BF244 looks to be about 3 pF so I wouldn't think that's a problem. The BF246 looks to be about 15pF so it's not as good. Being depletion MOSFETs, they should be even better. Better than what in what way? More similar in character to a grid. |
#28
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
"flipper" wrote in message news On Wed, 5 May 2010 15:53:19 +0100, "ian field" wrote: "flipper" wrote in message . .. On Tue, 4 May 2010 21:20:41 +0100, "ian field" wrote: "flipper" wrote in message m... On Tue, 04 May 2010 09:11:47 -0700, John Larkin wrote: On Tue, 04 May 2010 01:03:43 -0500, flipper wrote: On Mon, 3 May 2010 23:02:27 -0500, "Tim Williams" wrote: "flipper" wrote in message news:1k4vt5pctqpqush7g57aappqgt1e4ooubd@4a x.com... So you really don't think it matters whether one tries to drive the gate through a 10 ohm or 100 megohm resistor? I said cutoff frequency, which means voltage source. Do you have a 0 ohm source at ~1GHz? No? Then you won't get to see fT. ;-) Tim I know we won't see it so I have no idea what the heck your tap dance is for. Gm/Co was a popular figure of merit for tubes used in video amps and oscilloscopes. They took a hockey-stick turn upward when frame-grid tubes were invented, just before semiconductors took over. Mosfets have impressive Gm/Co values, at the cost of high input capacitance. John I understand but that doesn't answer the question at hand because we do not have unlimited gate drive capability. FOM, Rdson x Qg, isn't the issue as the device is 'capable' of more bandwidth if we could drive it but, because the 'driver' already exists, we do not have that choice. At least not without buffering it with something and if we do that we might as well use the jfet under the source version. But, in either case, it's no longer 'plain Jane'. For a given MOSFET technology FOM is roughly constant and I was asking if the same kind of thing applied to our problem of trading off gm, with source feedback, vs Ciss, because that is what the driving impedance works into. Unfortunately, Mouser doesn't let me search by Ciss, or gate charge either, so it's not a simple task to find a low one but I seem to remember that when I originally latched onto the STP2NK60Z it was 'pretty low'... but I wasn't taking gm into account. As I said, it's probably moot for a (AA5) 'table radio', because we don't need much bandwidth to begin with, but I was pondering it's potential suitability as a more general replacement. I suspect we need the more complex version if one wants 'full capability'. Just a thought, but if the common source section of the cascode is boosted by a bipolar, then a really low Idss FET can be used - like a 40673 or a BF991, even with both gates tied together its only a few pF. Sure, you can. A 'typical' (if there is such a thing) tube in this power range looks to be about 11 or 12 pF and the BF244 looks to be about 3 pF so I wouldn't think that's a problem. The BF246 looks to be about 15pF so it's not as good. Being depletion MOSFETs, they should be even better. Better than what in what way? More similar in character to a grid. If you mean it seems 'more similar' physically then I suggest the SS device, as a whole, is sufficiently different as to make focusing on one bit not very illuminating but if you like them there's nothing to prevent it's use. At least, I don't see one off hand. I think it better to focus on device parameters, like the BF244 has higher breakdown but the BF991 has more gain. And then there's that one is through hole and the other is surface mount. I wanted to buy some of the old BF9xx "pill" devices but couldn't find any, I got hold of a few BF991 SMDs and the only UHF types had Vds 7V abs-max, there's some very old TV tuners in a tea chest in the garage, so I do have a very small selection of "pill" types. I found that if you cut a 4-hole square of Veroboard and carefully V-groove across the 2 tracks with a scalpel, making 4 copper lands with a hole through each, the 4-pin SMD sits between the lands nicely, add 22SWG lead wires through the holes and the device is ready for general experimenting.. There's no reason the same can't be done directly on the underside of a project stripboard, it just requires a bit of carefull planning of the layout. |
#29
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
"flipper" wrote in message ... On Wed, 5 May 2010 20:58:30 +0100, "ian field" wrote: "flipper" wrote in message news On Wed, 5 May 2010 15:53:19 +0100, "ian field" wrote: "flipper" wrote in message m... On Tue, 4 May 2010 21:20:41 +0100, "ian field" wrote: "flipper" wrote in message news:8vo0u5lcehppt1if7msgt9gujo0qe87ag7@4ax. com... On Tue, 04 May 2010 09:11:47 -0700, John Larkin wrote: On Tue, 04 May 2010 01:03:43 -0500, flipper wrote: On Mon, 3 May 2010 23:02:27 -0500, "Tim Williams" wrote: "flipper" wrote in message news:1k4vt5pctqpqush7g57aappqgt1e4ooubd@ 4ax.com... So you really don't think it matters whether one tries to drive the gate through a 10 ohm or 100 megohm resistor? I said cutoff frequency, which means voltage source. Do you have a 0 ohm source at ~1GHz? No? Then you won't get to see fT. ;-) Tim I know we won't see it so I have no idea what the heck your tap dance is for. Gm/Co was a popular figure of merit for tubes used in video amps and oscilloscopes. They took a hockey-stick turn upward when frame-grid tubes were invented, just before semiconductors took over. Mosfets have impressive Gm/Co values, at the cost of high input capacitance. John I understand but that doesn't answer the question at hand because we do not have unlimited gate drive capability. FOM, Rdson x Qg, isn't the issue as the device is 'capable' of more bandwidth if we could drive it but, because the 'driver' already exists, we do not have that choice. At least not without buffering it with something and if we do that we might as well use the jfet under the source version. But, in either case, it's no longer 'plain Jane'. For a given MOSFET technology FOM is roughly constant and I was asking if the same kind of thing applied to our problem of trading off gm, with source feedback, vs Ciss, because that is what the driving impedance works into. Unfortunately, Mouser doesn't let me search by Ciss, or gate charge either, so it's not a simple task to find a low one but I seem to remember that when I originally latched onto the STP2NK60Z it was 'pretty low'... but I wasn't taking gm into account. As I said, it's probably moot for a (AA5) 'table radio', because we don't need much bandwidth to begin with, but I was pondering it's potential suitability as a more general replacement. I suspect we need the more complex version if one wants 'full capability'. Just a thought, but if the common source section of the cascode is boosted by a bipolar, then a really low Idss FET can be used - like a 40673 or a BF991, even with both gates tied together its only a few pF. Sure, you can. A 'typical' (if there is such a thing) tube in this power range looks to be about 11 or 12 pF and the BF244 looks to be about 3 pF so I wouldn't think that's a problem. The BF246 looks to be about 15pF so it's not as good. Being depletion MOSFETs, they should be even better. Better than what in what way? More similar in character to a grid. If you mean it seems 'more similar' physically then I suggest the SS device, as a whole, is sufficiently different as to make focusing on one bit not very illuminating but if you like them there's nothing to prevent it's use. At least, I don't see one off hand. I think it better to focus on device parameters, like the BF244 has higher breakdown but the BF991 has more gain. And then there's that one is through hole and the other is surface mount. I wanted to buy some of the old BF9xx "pill" devices but couldn't find any, I got hold of a few BF991 SMDs and the only UHF types had Vds 7V abs-max, there's some very old TV tuners in a tea chest in the garage, so I do have a very small selection of "pill" types. 7V? Boy, that's not a lot to play with. I don't know what values would be needed to make that one work but, for example, in the BF244 v8 circuit your 7V job wouldn't survive because there's up to 16V across the jfet (a bit over 9V even just at idle). IIRC only the UHF type has that limitation and I wouldn't bother using a UHF dg MOSFET for audio work when I have VHF types with a more useful voltage rating. Some of the tuners have "pill" type UHF MOSFETS with voltage ratings in the general direction of 20V, so I'm not at all cornered. |
#30
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
"flipper" wrote in message ... On Wed, 05 May 2010 16:01:07 -0500, flipper wrote: On Wed, 5 May 2010 20:58:30 +0100, "ian field" wrote: "flipper" wrote in message news On Wed, 5 May 2010 15:53:19 +0100, "ian field" wrote: "flipper" wrote in message om... On Tue, 4 May 2010 21:20:41 +0100, "ian field" wrote: "flipper" wrote in message news:8vo0u5lcehppt1if7msgt9gujo0qe87ag7@4ax .com... On Tue, 04 May 2010 09:11:47 -0700, John Larkin wrote: On Tue, 04 May 2010 01:03:43 -0500, flipper wrote: On Mon, 3 May 2010 23:02:27 -0500, "Tim Williams" wrote: "flipper" wrote in message news:1k4vt5pctqpqush7g57aappqgt1e4ooubd @4ax.com... So you really don't think it matters whether one tries to drive the gate through a 10 ohm or 100 megohm resistor? I said cutoff frequency, which means voltage source. Do you have a 0 ohm source at ~1GHz? No? Then you won't get to see fT. ;-) Tim I know we won't see it so I have no idea what the heck your tap dance is for. Gm/Co was a popular figure of merit for tubes used in video amps and oscilloscopes. They took a hockey-stick turn upward when frame-grid tubes were invented, just before semiconductors took over. Mosfets have impressive Gm/Co values, at the cost of high input capacitance. John I understand but that doesn't answer the question at hand because we do not have unlimited gate drive capability. FOM, Rdson x Qg, isn't the issue as the device is 'capable' of more bandwidth if we could drive it but, because the 'driver' already exists, we do not have that choice. At least not without buffering it with something and if we do that we might as well use the jfet under the source version. But, in either case, it's no longer 'plain Jane'. For a given MOSFET technology FOM is roughly constant and I was asking if the same kind of thing applied to our problem of trading off gm, with source feedback, vs Ciss, because that is what the driving impedance works into. Unfortunately, Mouser doesn't let me search by Ciss, or gate charge either, so it's not a simple task to find a low one but I seem to remember that when I originally latched onto the STP2NK60Z it was 'pretty low'... but I wasn't taking gm into account. As I said, it's probably moot for a (AA5) 'table radio', because we don't need much bandwidth to begin with, but I was pondering it's potential suitability as a more general replacement. I suspect we need the more complex version if one wants 'full capability'. Just a thought, but if the common source section of the cascode is boosted by a bipolar, then a really low Idss FET can be used - like a 40673 or a BF991, even with both gates tied together its only a few pF. Sure, you can. A 'typical' (if there is such a thing) tube in this power range looks to be about 11 or 12 pF and the BF244 looks to be about 3 pF so I wouldn't think that's a problem. The BF246 looks to be about 15pF so it's not as good. Being depletion MOSFETs, they should be even better. Better than what in what way? More similar in character to a grid. If you mean it seems 'more similar' physically then I suggest the SS device, as a whole, is sufficiently different as to make focusing on one bit not very illuminating but if you like them there's nothing to prevent it's use. At least, I don't see one off hand. I think it better to focus on device parameters, like the BF244 has higher breakdown but the BF991 has more gain. And then there's that one is through hole and the other is surface mount. I wanted to buy some of the old BF9xx "pill" devices but couldn't find any, I got hold of a few BF991 SMDs and the only UHF types had Vds 7V abs-max, there's some very old TV tuners in a tea chest in the garage, so I do have a very small selection of "pill" types. 7V? Boy, that's not a lot to play with. I don't know what values would be needed to make that one work but, for example, in the BF244 v8 circuit your 7V job wouldn't survive because there's up to 16V across the jfet (a bit over 9V even just at idle). Remember, the MOSFET gate is fixed and there has to be enough headroom for the jfet *plus* current swing through it's source resistor (plus the NPN). One might try a swinging gate on the power MOSFET but that creates stability and frequency response problems for the same reason we went cascode to begin with. I found that if you cut a 4-hole square of Veroboard and carefully V-groove across the 2 tracks with a scalpel, making 4 copper lands with a hole through each, the 4-pin SMD sits between the lands nicely, add 22SWG lead wires through the holes and the device is ready for general experimenting.. That's clever but I found TO-92 devices just sort of come "ready for general experimenting." hehe On the other hand it's irritating to pay shipping for a 10 cent device. There's no reason the same can't be done directly on the underside of a project stripboard, it just requires a bit of carefull planning of the layout. P.S. If you want to just try something why not start with the simpler 'plain Jane' version? As explained at the start, I don't have an old radio to experiment on, so its a purely theoretical excercise. Thanks to the suggestions in the replies, I shouldn't have any trouble getting started should the opportunity for practical arise. |
#31
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
On Sun, 02 May 2010 12:28:29 -0500, flipper wrote:
On Sun, 02 May 2010 06:56:01 -0700, wrote: On Thu, 29 Apr 2010 11:33:10 -0500, flipper wrote: On Thu, 29 Apr 2010 08:41:52 -0700, Jim Thompson wrote: On Thu, 29 Apr 2010 16:22:59 +0100, "ian field" wrote: Here is a circuit that I scribbled on an odd scrap of paper that I would like to experiment with, but unfortunately I don't have an old valve radio to use as a testbed. The component values that are pencilled in will doubtless need revising and the rest by trial and error. The plan is of course to make a MOSFET substitute for the O/P pentode in an old valve radio, it has 3 stages to maintain its invertingness in case the audio amplifier has a nfb loop. Anyone care to cast a critical eye over it and predict how loud a bang it will make? You're making it too hard. Just slip a 2N7000 under your 900V device (making a cascode), tie the 900V device gate to the +30V, and bias the 2N7000 as desired. Personally I've done this where the 900V device was a toob, and, in the cathode circuit was an NPN with an OpAmp wrapped around it ;-) Well, in the toob world there are those who would say the same thing, that "You're making it too hard," since the toob will self bias with a simple resistor (audio bypassed for maximum gain) under the cathode, grid to ground, similar to what is done with a simple jfet current source. Except for when it doesn't work, to bias the final / power output stage. You'll have to explain what you mean by "for when it doesn't work" because self biased was the most common means of biasing output stages, with fixed bias next. I wouldn't say it's necessarily the 'best', depending on what the goals are and, so, how 'best' is measured, but it's arguably the 'least hard'. It is mostly a pointer to push-pull output class AB finals. Self bias works very poorly in that case. The issues start with low PSRR and poor linearity. Nor does self bias work well with class A single ended power output stages. I regularly saw combined bias in power output cases, often with negative grid supplies. |
#32
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve). - NFET.jpg
"flipper" wrote in message ... On Wed, 05 May 2010 16:01:07 -0500, flipper wrote: On Wed, 5 May 2010 20:58:30 +0100, "ian field" wrote: "flipper" wrote in message news On Wed, 5 May 2010 15:53:19 +0100, "ian field" wrote: "flipper" wrote in message om... On Tue, 4 May 2010 21:20:41 +0100, "ian field" wrote: "flipper" wrote in message news:8vo0u5lcehppt1if7msgt9gujo0qe87ag7@4ax .com... On Tue, 04 May 2010 09:11:47 -0700, John Larkin wrote: On Tue, 04 May 2010 01:03:43 -0500, flipper wrote: On Mon, 3 May 2010 23:02:27 -0500, "Tim Williams" wrote: "flipper" wrote in message news:1k4vt5pctqpqush7g57aappqgt1e4ooubd @4ax.com... So you really don't think it matters whether one tries to drive the gate through a 10 ohm or 100 megohm resistor? I said cutoff frequency, which means voltage source. Do you have a 0 ohm source at ~1GHz? No? Then you won't get to see fT. ;-) Tim I know we won't see it so I have no idea what the heck your tap dance is for. Gm/Co was a popular figure of merit for tubes used in video amps and oscilloscopes. They took a hockey-stick turn upward when frame-grid tubes were invented, just before semiconductors took over. Mosfets have impressive Gm/Co values, at the cost of high input capacitance. John I understand but that doesn't answer the question at hand because we do not have unlimited gate drive capability. FOM, Rdson x Qg, isn't the issue as the device is 'capable' of more bandwidth if we could drive it but, because the 'driver' already exists, we do not have that choice. At least not without buffering it with something and if we do that we might as well use the jfet under the source version. But, in either case, it's no longer 'plain Jane'. For a given MOSFET technology FOM is roughly constant and I was asking if the same kind of thing applied to our problem of trading off gm, with source feedback, vs Ciss, because that is what the driving impedance works into. Unfortunately, Mouser doesn't let me search by Ciss, or gate charge either, so it's not a simple task to find a low one but I seem to remember that when I originally latched onto the STP2NK60Z it was 'pretty low'... but I wasn't taking gm into account. As I said, it's probably moot for a (AA5) 'table radio', because we don't need much bandwidth to begin with, but I was pondering it's potential suitability as a more general replacement. I suspect we need the more complex version if one wants 'full capability'. Just a thought, but if the common source section of the cascode is boosted by a bipolar, then a really low Idss FET can be used - like a 40673 or a BF991, even with both gates tied together its only a few pF. Sure, you can. A 'typical' (if there is such a thing) tube in this power range looks to be about 11 or 12 pF and the BF244 looks to be about 3 pF so I wouldn't think that's a problem. The BF246 looks to be about 15pF so it's not as good. Being depletion MOSFETs, they should be even better. Better than what in what way? More similar in character to a grid. If you mean it seems 'more similar' physically then I suggest the SS device, as a whole, is sufficiently different as to make focusing on one bit not very illuminating but if you like them there's nothing to prevent it's use. At least, I don't see one off hand. I think it better to focus on device parameters, like the BF244 has higher breakdown but the BF991 has more gain. And then there's that one is through hole and the other is surface mount. I wanted to buy some of the old BF9xx "pill" devices but couldn't find any, I got hold of a few BF991 SMDs and the only UHF types had Vds 7V abs-max, there's some very old TV tuners in a tea chest in the garage, so I do have a very small selection of "pill" types. 7V? Boy, that's not a lot to play with. I don't know what values would be needed to make that one work but, for example, in the BF244 v8 circuit your 7V job wouldn't survive because there's up to 16V across the jfet (a bit over 9V even just at idle). Remember, the MOSFET gate is fixed and there has to be enough headroom for the jfet *plus* current swing through it's source resistor (plus the NPN). One might try a swinging gate on the power MOSFET but that creates stability and frequency response problems for the same reason we went cascode to begin with. I found that if you cut a 4-hole square of Veroboard and carefully V-groove across the 2 tracks with a scalpel, making 4 copper lands with a hole through each, the 4-pin SMD sits between the lands nicely, add 22SWG lead wires through the holes and the device is ready for general experimenting.. That's clever but I found TO-92 devices just sort of come "ready for general experimenting." hehe On the other hand it's irritating to pay shipping for a 10 cent device. There's no reason the same can't be done directly on the underside of a project stripboard, it just requires a bit of carefull planning of the layout. P.S. If you want to just try something why not start with the simpler 'plain Jane' version? Just out of curiosity, regarding the basic JFET/MOSFET cascode with the JFET buffered by a bipolar - would there be any particular advantage from using a single BJT current source in place of the source/emitter resistor? |
#33
Posted to alt.binaries.schematics.electronic
|
|||
|
|||
Experiment (valve).
On Tue, 11 May 2010 00:20:39 -0500, flipper wrote:
On Mon, 10 May 2010 19:10:40 -0700, wrote: On Sun, 02 May 2010 12:28:29 -0500, flipper wrote: On Sun, 02 May 2010 06:56:01 -0700, wrote: On Thu, 29 Apr 2010 11:33:10 -0500, flipper wrote: On Thu, 29 Apr 2010 08:41:52 -0700, Jim Thompson wrote: On Thu, 29 Apr 2010 16:22:59 +0100, "ian field" wrote: Here is a circuit that I scribbled on an odd scrap of paper that I would like to experiment with, but unfortunately I don't have an old valve radio to use as a testbed. The component values that are pencilled in will doubtless need revising and the rest by trial and error. The plan is of course to make a MOSFET substitute for the O/P pentode in an old valve radio, it has 3 stages to maintain its invertingness in case the audio amplifier has a nfb loop. Anyone care to cast a critical eye over it and predict how loud a bang it will make? You're making it too hard. Just slip a 2N7000 under your 900V device (making a cascode), tie the 900V device gate to the +30V, and bias the 2N7000 as desired. Personally I've done this where the 900V device was a toob, and, in the cathode circuit was an NPN with an OpAmp wrapped around it ;-) Well, in the toob world there are those who would say the same thing, that "You're making it too hard," since the toob will self bias with a simple resistor (audio bypassed for maximum gain) under the cathode, grid to ground, similar to what is done with a simple jfet current source. Except for when it doesn't work, to bias the final / power output stage. You'll have to explain what you mean by "for when it doesn't work" because self biased was the most common means of biasing output stages, with fixed bias next. I wouldn't say it's necessarily the 'best', depending on what the goals are and, so, how 'best' is measured, but it's arguably the 'least hard'. It is mostly a pointer to push-pull output class AB finals. Self bias works very poorly in that case. The issues start with low PSRR and poor linearity. I'm not sure where you get the 'low PSRR' and 'poor linearity' ideas. Compared to what? PSRR is primarily a function of hum into the plates (and screens) and is going to be essentially the same whether cathode or fixed bias. And if things are well balanced it'll cancel in the push pull output transformer. The 'linearity' issue is inherent to the A to B transition and is there with fixed bias too. Where self bias has a 'problem' is in sustained 'B' operation as bias voltage will rise due to the higher 'average current'. But, in practice, that isn't a problem because 'music' is not a single mono-tone max volume note and the 'B' region only handles transient peaks. That is, as long as there is enough 'A' in the AB to well cover the average music level. That isn't the case in heavily driven Class AB guitar amps so they are most commonly fixed bias but most commodity, as well as low to medium level 'hi-fi', AB amps used self bias for the (power and) simplicity (cost) and convenience of no user adjustment required. Meaning you didn't have to worry about Blondie buying replacements out of the corner Piggly Wiggly tube tester and setting Dagwood's phonograph on fire because she didn't know to re-bias the amp. Nor does self bias work well with class A single ended power output stages. That's where it works best because, in idealized theory, there is no 'change' in average current. In reality, distortion causes a slight change but virtually every AA5 radio made, and the vast majority of TV sets, employed single ended, self bias, audio outputs. Class A, btw, is also where simple bias servos work best, and for the same reason: current is relatively constant. When you think about it, most bias servos are essentially glorified 'high gain' variations of self bias. Hum is considerably more difficult to deal with SE, because there's no inherent cancellation, but the less overall complexity more than makes up for it (in the context of commodity table radio and TV set cost). I regularly saw combined bias in power output cases, often with negative grid supplies. What would be a "combined bias" without a negative supply? The holy grail of AB (tube) bias has always been to devise a 'fixed bias' that 'self biases' and Lincoln Walsh may have come closest with the Brook Bias. http://www.tubecad.com/2007/12/15/Hi...%20Control.pdf The idea is, if you look at the sum of the dynamic current through both output tubes in an AB amp the 'minimum' observed value represents twice the idle current. (Spice it. It works). His circuit merely detects that minimum and then adjusts grid voltage accordingly. The problem, however, is in accurately catching the minimum of a variable frequency waveform so it tends to fluctuate as well, but not so much as 'plain' self bias. Patrick Turner has a different approach where 'excess' (meaning essentially the 'B' portion of AB) current is shunted around the 'self bias' resistor. http://www.turneraudio.com.au/300w-5...stabilizer.htm Am i getting this straight? You are proposing servoed bias as simple PA bias? That servoed bias is comparable to self bias? Somehow i suspect not. The examples you show argue for my case that self bias (and fixed bias to a lesser extent) does not work well for power output stages. |
Reply |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
Experiment | Woodturning | |||
Brass Ball Valve,Gas Valve,Needle Valve,Angle Valve Sale on good price | Home Repair | |||
Valve,Butterfly valve,Globe valve,Check valve,Ball valve,Plug valve,Marine valve,Gate valve,Flow control valve | UK diy | |||
Valve,butterfly valve,ball valve,check valve,globe valve | Home Repair | |||
Vac Experiment 2 | Woodworking |