I need to replace this TO220 regulator in a power supply I'm working
on. The original is a "CT" suffix however looking through the books I
see that there are several other versions of this part. The original
seems to be a 1.0 amp device while several others appear to be able to
handle 1.50 amps. I would like to buy a few extras to have on hand.
Aside from any appreciable price difference, would it make sense to
buy a bunch of "CT's" that will only handle 1.0 amp when there are1.50
amp parts available that will do the equivalent job? Thanks, Lenny

klem kedidelhopper wrote:

I need to replace this TO220 regulator in a power supply I'm working
on. The original is a "CT" suffix however looking through the books I
see that there are several other versions of this part. The original
seems to be a 1.0 amp device while several others appear to be able to
handle 1.50 amps. I would like to buy a few extras to have on hand.
Aside from any appreciable price difference, would it make sense to
buy a bunch of "CT's" that will only handle 1.0 amp when there are1.50
amp parts available that will do the equivalent job? Thanks, Lenny
It makes sense to get a larger unit if it's not going to put you in the
poor house.

Jamie

In article d06e1195-0aa7-4ab7-813a-
,
says...

I need to replace this TO220 regulator in a power supply I'm working
on. The original is a "CT" suffix however looking through the books I
see that there are several other versions of this part. The original
seems to be a 1.0 amp device while several others appear to be able to
handle 1.50 amps. I would like to buy a few extras to have on hand.
Aside from any appreciable price difference, would it make sense to
buy a bunch of "CT's" that will only handle 1.0 amp when there are1.50
amp parts available that will do the equivalent job? Thanks, Lenny

The 1.5A part may have a somewhat beefier onboard series pass transistor
than the 1.0A part. However a big difference in the two parts is the
point that internal current limiting kicks in.

On the other hand keep in mind that the primary driver on how much
current that a particular 7824 regulator installation can handle is
limited by the power being dissipated in the device. Input to output
voltage differential and type/amount of cooling will be major factors to
consider.

--

Michael Karas
Carousel Design Solutions
http://www.carousel-design.com

klem kedidelhopper wrote:

I need to replace this TO220 regulator in a power supply I'm working
on. The original is a "CT" suffix however looking through the books I
see that there are several other versions of this part. The original
seems to be a 1.0 amp device while several others appear to be able to
handle 1.50 amps. I would like to buy a few extras to have on hand.
Aside from any appreciable price difference, would it make sense to
buy a bunch of "CT's" that will only handle 1.0 amp when there are1.50
amp parts available that will do the equivalent job? Thanks, Lenny

We used to use that LM78xx family in the 80's, a lot of them. The 05, 08,
12, 18 and 24's.

I don't remember a 1.5A version in the TO220 case, but technology marches on
as they say.

One hint, they aren't rocket science. It's a cheap regulator that does the
job simply, but I'm unsure if better ratings would give better results.

The main problem we found with those is the source voltage, I think they are
all rated with Vcc max at 36v, even the 5v ones. There is a min of something
like 3v over the regulation voltage (7805 needs min 8v in, the 7812, 15v and
so forth). The closer you can get to that, the better the life span.

The heat sink is also important, it needs one. I think the data sheets from
National said something about 100ma-250ma "free air", and the full 1A draw
with some minimum postage-stamp sink, but don't beleive it.

They are/were supposed to be over-current and short circuit proof, but they
do pop like firecrackers when they let loose.

One handy trick with them, if you lift the ground pin and insert a pot or
some diodes, you can make it regulate at different voltages, keeping all the
ripple rejection and other specs within specs. So a 7805 with a 30v source,
a few caps and like a 500 ohm pot makes a quick and dirty 5v-24v variable
regulated power supply, 1A max as usual.

I don't know what you found with the 1.5A rating but it would just be cost
considerations, if there is a 10 cent difference, fine, if it's $10, forget
it. This is a 30 year old comment that probably doesn't apply but we did
find differences depending who made them. Both the National and Motorola
versions were fine, the ones by SGS stunk. There was another by Fairchild
but I think they were SGS's with a re-ink job on them.

-bruce

On Dec 1, 7:01*am, Bruce Esquibel wrote:
klem kedidelhopper wrote:
I need to replace this TO220 regulator in a power supply I'm working
on. The original is a "CT" suffix however looking through the books I
see that there are several other versions of this part. The original
seems to be a 1.0 amp device while several others appear to be able to
handle 1.50 amps. I would like to buy a few extras to have on hand.
Aside from any appreciable price difference, would it make sense to
buy a bunch of "CT's" that will only handle 1.0 amp when there are1.50
amp parts available that will do the equivalent job? * Thanks, Lenny

We used to use that LM78xx family in the 80's, a lot of them. The 05, 08,
12, 18 and 24's.

I don't remember a 1.5A version in the TO220 case, but technology marches on
as they say.

One hint, they aren't rocket science. It's a cheap regulator that does the
job simply, but I'm unsure if better ratings would give better results.

The main problem we found with those is the source voltage, I think they are
all rated with Vcc max at 36v, even the 5v ones. There is a min of something
like 3v over the regulation voltage (7805 needs min 8v in, the 7812, 15v and
so forth). The closer you can get to that, the better the life span.

The heat sink is also important, it needs one. I think the data sheets from
National said something about 100ma-250ma "free air", and the full 1A draw
with some minimum postage-stamp sink, but don't beleive it.

They are/were supposed to be over-current and short circuit proof, but they
do pop like firecrackers when they let loose.

One handy trick with them, if you lift the ground pin and insert a pot or
some diodes, you can make it regulate at different voltages, keeping all the
ripple rejection and other specs within specs. So a 7805 with a 30v source,
a few caps and like a 500 ohm pot makes a quick and dirty 5v-24v variable
regulated power supply, 1A max as usual.

I don't know what you found with the 1.5A rating but it would just be cost
considerations, if there is a 10 cent difference, fine, if it's $10, forget
it. This is a 30 year old comment that probably doesn't apply but we did
find differences depending who made them. Both the National and Motorola
versions were fine, the ones by SGS stunk. There was another by Fairchild
but I think they were SGS's with a re-ink job on them.

-bruce


That's very interesting about the 500 ohm pot. I've never heard of
doing that before. I would have thought that the regulated voltage
would have been dictated by some internal zener diode or something.
But in any case I have a bag of 7805's and I'm going to experiment
with them. As far as the capacitors go are you talking about using any
additional from the usual input and output caps? Lenny.

klem kedidelhopper wrote:

That's very interesting about the 500 ohm pot. I've never heard of
doing that before. I would have thought that the regulated voltage
would have been dictated by some internal zener diode or something.
But in any case I have a bag of 7805's and I'm going to experiment
with them. As far as the capacitors go are you talking about using any
additional from the usual input and output caps? Lenny.


Here is the datasheet on that family of regulators. the first page
shows 'Adjustable Output Regulator' and that you need a fixed resistor
from output to common, along with the pot from common to ground.

http://www.national.com/ds/LM/LM340.pdf

Texas Instruments UA7824CKC is 1.5A

Mouser has several 1.5A types in stock:
http://www.mouser.com/Semiconductors/Integrated-Circuits-ICs/_/N-6j73k?Keyword=7824&FS=True
starting at 47 cents each, which is only seven cents more than the 1A
types. They are $4.56 for ten or $43.70 per hundred. this is for the
511-L7824ACV made by STMicroelectronics. The right part is cheaper than
screwing around with extra resistors & pots that can go open.

http://www.mouser.com/ProductDetail/STMicroelectronics/L7824ACV/?qs=sGAEpiMZZMtOXy69nW9rM5taedogOtNRM5vVsMlyJSo%3d


The description column list the output of each part.


--
You can't have a sense of humor, if you have no sense.

Michael A. Terrell wrote:

Here is the datasheet on that family of regulators. the first page
shows 'Adjustable Output Regulator' and that you need a fixed resistor
from output to common, along with the pot from common to ground.

http://www.national.com/ds/LM/LM340.pdf


Yeah but I think that is specific to the LM340, on the 78xx series you just
need a resistor in the path to ground.

This is more what I was talking about:

http://www.fairchildsemi.com/ds/LM/LM7805.pdf

It's a bit messy with the 741 in circuit (fig 13), but allows you to use
higher value pots. The single resistor method works but may require a
multi-turn pot to dial it in with precision or a need for odd values to hit
the target.

Just play around with it, use a 7805 and Vcc around 24-28v for the best
range and just experiment with different values to ground. Just tie the
wiper to one of the ends, one end to ground and the other to pin 2.

You need that cap on the output to ground to snuff oscillation and something
on the input for ripple but thats about it. That 78xx family is forgiving
but if it's too hot to touch, you are doing something wrong.

-bruce

Bruce Esquibel wrote:

Michael A. Terrell wrote:

Here is the datasheet on that family of regulators. the first page
shows 'Adjustable Output Regulator' and that you need a fixed resistor
from output to common, along with the pot from common to ground.

http://www.national.com/ds/LM/LM340.pdf

Yeah but I think that is specific to the LM340, on the 78xx series you just
need a resistor in the path to ground.

This is more what I was talking about:

http://www.fairchildsemi.com/ds/LM/LM7805.pdf

It's a bit messy with the 741 in circuit (fig 13), but allows you to use
higher value pots. The single resistor method works but may require a
multi-turn pot to dial it in with precision or a need for odd values to hit
the target.

Just play around with it, use a 7805 and Vcc around 24-28v for the best
range and just experiment with different values to ground. Just tie the
wiper to one of the ends, one end to ground and the other to pin 2.

You need that cap on the output to ground to snuff oscillation and something
on the input for ripple but thats about it. That 78xx family is forgiving
but if it's too hot to touch, you are doing something wrong.


I did it over 30 years ago for a high current bench power supply to
service business radios. It provided 5.5 V to 16 V at up to 90 amps, and
was used once to test a car starter I'd rebuilt for one of my service
trucks.

If you read the datasheet, LM340 is the National Semiconductor name
for the 78** family. It is followed by a dash & the output voltage.

The reason for the second resistor is to make the output voltage more
stable. The current consumed by the regulator varies with temperature,
so the lower value fixed resistor across the output to the common
terminal makes it a lot stiffer.

WTH would you spend money on a precision variable resistor just to
let it drift, as the die temperature changes with the output current? A
pair of resistors is a better choice for a fixed output, since no one
can screw with it and the output won't go high if the rheostat opens.

It's still cheaper to buy the common values 10 or more at a time and
use the right part. You waste time & parts to kluge the wrong part into
the circuit, and both cost you money. In some types of repairs, you are
not allowed to kluge in the wrong part without a huge pipe of paperwork,
and I did a lot of that kind of work, from '72 until I was sidelined by
my health a few years ago.

I keep 5, 8,9,12,15 & 24 positive regulators and 5,12,15 V negative
regulators in stock. Most TO-220 are under 25 cents each, in single
quantities. I still have some TO-3 metal cans in stock, along with some
long obsolete Motorola ten pin regulators that were common in early
external 300 baud modems.

There was an early EPROM programmer project in BYTE magazine that
used separate transistors to ground different resistors to set the
programming voltage on various pins, since there were so many
combinations used in the '70s.


--
You can't have a sense of humor, if you have no sense.

In article ,
Bruce Esquibel wrote:

Michael A. Terrell wrote:

Here is the datasheet on that family of regulators. the first page
shows 'Adjustable Output Regulator' and that you need a fixed resistor
from output to common, along with the pot from common to ground.

http://www.national.com/ds/LM/LM340.pdf


Yeah but I think that is specific to the LM340, on the 78xx series you just
need a resistor in the path to ground.

This is more what I was talking about:

http://www.fairchildsemi.com/ds/LM/LM7805.pdf

It's a bit messy with the 741 in circuit (fig 13), but allows you to use
higher value pots. The single resistor method works but may require a
multi-turn pot to dial it in with precision or a need for odd values to hit
the target.

Just play around with it, use a 7805 and Vcc around 24-28v for the best
range and just experiment with different values to ground. Just tie the
wiper to one of the ends, one end to ground and the other to pin 2.

You need that cap on the output to ground to snuff oscillation and something
on the input for ripple but thats about it. That 78xx family is forgiving
but if it's too hot to touch, you are doing something wrong.

Maybe not. The temperature where most folks say "too hot to touch" (too
hot to keep your finger on) is around 130 F; the maximum junction
temperature rating for those regulators is around 300 F.

Isaac

On Dec 3, 1:08*am, isw wrote:
In article ,
*Bruce Esquibel wrote:



Michael A. Terrell wrote:

* *Here is the datasheet on that family of regulators. the first page
shows 'Adjustable Output Regulator' and that you need a fixed resistor
from output to common, along with the pot from common to ground.

http://www.national.com/ds/LM/LM340.pdf

Yeah but I think that is specific to the LM340, on the 78xx series you just
need a resistor in the path to ground.

This is more what I was talking about:

http://www.fairchildsemi.com/ds/LM/LM7805.pdf

It's a bit messy with the 741 in circuit (fig 13), but allows you to use
higher value pots. The single resistor method works but may require a
multi-turn pot to dial it in with precision or a need for odd values to hit
the target.

Just play around with it, use a 7805 and Vcc around 24-28v for the best
range and just experiment with different values to ground. Just tie the
wiper to one of the ends, one end to ground and the other to pin 2.

You need that cap on the output to ground to snuff oscillation and something
on the input for ripple but thats about it. That 78xx family is forgiving
but if it's too hot to touch, you are doing something wrong.

Maybe not. The temperature where most folks say "too hot to touch" (too
hot to keep your finger on) is around 130 F; the maximum junction
temperature rating for those regulators is around 300 F.

Isaac

Thanks for all the replies guys. While I can certainly appreciate not
having to screw around klugeing a circuit it is interesting to know
that I can play with this basic design as it was suggested if I should
ever need to. And many times over the years I've had to do things like
this to get the job done. When you're 75 miles from home on a Friday
afternoon and "Miller time" is fast approaching it's nice to be able
to make something work that otherwise might require another service
call on Monday. Not to mention the added inconvenience this saves the
customer. BTW, I was always taught that good design practice dictates
that if a semiconductor is too hot to keep your finger on, (greater
than around 55.0 degrees C), then it's running too hot. However I
remember some military aircraft power supplies I worked on in the 70's
that ran a lot hotter then that on a regular basis. It always seemed
like poor practice to me but I never heard of an F15 going down
because of one, so I guess it was OK. Lenny

Michael A. Terrell wrote:

If you read the datasheet, LM340 is the National Semiconductor name
for the 78** family. It is followed by a dash & the output voltage.

Yeah, for some reason I had the LM317 stuck in my head which was some kind
of variable regulator. It wasn't very good of a device, but was common.

It's still cheaper to buy the common values 10 or more at a time and
use the right part. You waste time & parts to kluge the wrong part into
the circuit, and both cost you money. In some types of repairs, you are
not allowed to kluge in the wrong part without a huge pipe of paperwork,
and I did a lot of that kind of work, from '72 until I was sidelined by
my health a few years ago.

The above comment depends on the situation. Of course it's better to use the
exact part and do the repair properly, but that doesn't hold true of all
circumstances.

Back in the 80's I was involved with a wholesale guy here in Illinois out
near Rockford, he used to buy semi-trailers of returned merchandise and
closeouts, anything from clock radios, multi-band radios, cable converters
and some stuff I couldn't figure out its use.

At best, and I do mean at best, it was $2 a repair. Most of the time it was
50 cents, but parts were billed separate. On first thought, most people
would just laugh getting paid that little, but when you are looking at 3000,
4000 or 5000 peices with one group exceeding 20,000 items, it makes you
think.

Basically if you spent more than 5 minutes to repair it, you were doing
charity work. So it comes down to meatball surgury, cut out the bad part,
get a new fix in, button it up and move on to the next.

The majority of that crap had power supply problems, I'm talking the old,
linear device type, some kind of transformer, a couple diodes (if you were
lucky) and maybe 1 filter cap. After that it was transistor de'jour to feed
the device. Some of these had PNP where it was supposed to be NPN (I guess
2SB is close enough to 2SD), or on some, they were missing altogether.

Identifying a zener diode, ha, didn't have the luxury. Most of those were
probably some pin diode from the 60's that looked close enough.

Point was, I made a lot of money doing the kludge, paying someone to tack
solder a couple parts to a 7805, pay someone else to clip here, clip there
then solder in the kludge to these points, instant 9v regulation. Plug it
in, if it works, on to the next.

Was easier using the mess soldered onto the 78xx than getting the right one
because you had leads to work with, not jumpers, which attaching them would
of slowed things down.

It wasn't rocket science, wasn't elegant and in some respects, was kind of
embarassing. But it was work, paid well if you could figure out the routine
and overall made everyone happy.

The only major botch job I did where nearly everything came back to be
reworked was these goofy modulators, which I can't even imagine to this day
what they were being used for. They wanted to retune them from a channel 2
output to channel 4, which after playing with a couple, seemed to be a
simply coil retuning, but turned out they weren't 75 ohm "things" but 50
ohm, so they didn't work wherever they ended up.

I hated that, not only because of doing it twice, but the chamber that the
coils were in was sealed via a half dozen sheet metal screws that were like
1/8", tiny ****ers. I had to have custom made hardened steel bits made
because any socket or hand tool would simply wear out after a days use. And
of course, the heads to them mostly didn't like being untighten and tighten
twice.

Anyway, getting back to the subject, I understand what you are saying but I
wasn't employed by NASA here. That guy was trying to make a dollar on
something that probably cost him 25 cents. Here in Chicago there was a lot
of opportunities like him. We used to have these places that sold "hotel
tv's", cheap 19" RCA sets (usually) for half or less what they were going
for retail. In the early 80's, most video stores that did tape rentals also
did machine rentals, they were still too expensive for everyone to own one.

All of the above didn't care about the quality of repair, just the costs.
Even if up front you told them that on the cheap it may only last 3 months,
if it was a lot less than doing it the right way, cheap it was.

Just saying that LM78xx series is a versatile little device that can be used
in a far wider arena than what the spec sheets say.

-bruce