Anybody hear ever try to build a 3 phase vfd ?

I am in the mood to build something that i could buy ready made cheaper.

Best Regards
Tom.

azotic wrote:
Anybody hear ever try to build a 3 phase vfd ?

I am in the mood to build something that i could buy ready made cheaper.

Best Regards
Tom.

I think that most people wanting to build 3 phase sources build the
rotary converters.

If you were going to build a 3 phase VFD I would suggest taking a look
at the Freescale DSP56F807 microcontroller. It has a great PWM generator
designed for stuff like this. I have used (underused?) the PWM generator
for DC motor drives, and it is really capable.

Good Luck,
Bob

"BobH" wrote in message
...
azotic wrote:
Anybody hear ever try to build a 3 phase vfd ?

I am in the mood to build something that i could buy ready made cheaper.

Best Regards
Tom.

I think that most people wanting to build 3 phase sources build the rotary
converters.

If you were going to build a 3 phase VFD I would suggest taking a look at
the Freescale DSP56F807 microcontroller. It has a great PWM generator
designed for stuff like this. I have used (underused?) the PWM generator
for DC motor drives, and it is really capable.

Good Luck,
Bob


Thanks for the tip, the rotary phase converters don't have the flexability i
want
and are to noisy for me. I was thinking about using one of my old note book
computers interfaced to some power semiconductors and hopefully end up with
a user
programable vfd.

Best Regards
Tom.

On Sat, 4 Aug 2007 09:11:30 -0700, azotic wrote:

Thanks for the tip, the rotary phase converters don't have the
flexability i want and are to noisy for me. I was thinking about
using one of my old note book computers interfaced to some power
semiconductors and hopefully end up with a user programable vfd.


Do not underestimate the cost of this project, and you will end up
with something rather substandard. But it is fun.

For example, single phase input rated, 1 horsepower VFDs can be bought
on eBay for about $47 plus $12 shipping. It is hard to beat that.

i

Ignoramus12500 wrote:
On Sat, 4 Aug 2007 09:11:30 -0700, azotic wrote:


Thanks for the tip, the rotary phase converters don't have the
flexability i want and are to noisy for me. I was thinking about
using one of my old note book computers interfaced to some power
semiconductors and hopefully end up with a user programable vfd.



Do not underestimate the cost of this project, and you will end up
with something rather substandard. But it is fun.

For example, single phase input rated, 1 horsepower VFDs can be bought
on eBay for about $47 plus $12 shipping. It is hard to beat that.

i

Got an example of that price point, Igor?

Can he put several of them in series to up the HP?

Bob (should know better) Swinney
"Ignoramus12500" wrote in message
...
On Sat, 4 Aug 2007 09:11:30 -0700, azotic wrote:

Thanks for the tip, the rotary phase converters don't have the
flexability i want and are to noisy for me. I was thinking about
using one of my old note book computers interfaced to some power
semiconductors and hopefully end up with a user programable vfd.


Do not underestimate the cost of this project, and you will end up
with something rather substandard. But it is fun.

For example, single phase input rated, 1 horsepower VFDs can be bought
on eBay for about $47 plus $12 shipping. It is hard to beat that.

i

Back when I was working as an electronics engineer in San Jose, California
(heart of Silicon Valley) I came up with a design for this. It was a chip which
produced variable PWM waveforms. I planned to use power Darlington transistors
to go between the CMOS logic chip and the gates of beefy SCRs. I got as far as
simulating the whole design in a mixed signal simulator.

Then I found out what the failure mode of SCRs is. About the same as hand
grenades. Did I want this in my house? No.

I scrapped the design, never looked back.

Start with your output phase, your "power semiconductors". If you can find some
of those that meet your spec at a price point you can live with, that's one
thing. I'm guessing that buying a small handful of them will get you a price
much higher than an entire VFD made by TECO or someone else.

Igor, I'm still waiting for that $47 1hp VFD new on ebay ..

Grant

Grant Erwin wrote:
Back when I was working as an electronics engineer in San Jose,
California (heart of Silicon Valley) I came up with a design for this.
It was a chip which produced variable PWM waveforms. I planned to use
power Darlington transistors to go between the CMOS logic chip and the
gates of beefy SCRs. I got as far as simulating the whole design in a
mixed signal simulator.

Then I found out what the failure mode of SCRs is. About the same as
hand grenades. Did I want this in my house? No.

I scrapped the design, never looked back.

Now the power devices will probably be Insulated Gate Bipolar
Transistors, but the failure mode is the same. 240VAC Rectifies to over
300VDC and a multi-horsepower drive will need many amps of DC supply
capability. I have built some smaller drives and from that my advice is:
A) keep a CO2 fire bottle handy.
B) always wear safety glasses.
C) never leave the room with a prototype energized.
D) buy lots of spare transistors and drivers.
E) don't try to build with point to point wiring, parasitics will kill it.
F) keep a CO2 fire bottle handy.


In the few commercial VFD's I have played with programmability was never
in short supply. Figuring out all the programability they gave you was
tough enough.

That said, doing it is still one of the best ways to learn about a new
area of technology. I have been kicking around the idea of designing and
building an inverter welding machine for a while. It would not replace
my commercial machine for a long time, but it would be a heck of a
project. I wonder if argon puts out fires as well as CO2? (grin)

Bob

I wonder if argon puts out fires as well as CO2?

Probably better, but it sure costs a heck of a lot more. It costs me $13 to
refill a 20 pound CO2 cylinder but it costs me about $42 to refill a 150 cf
argon cylinder, about twice as tall.

Grant

On Sat, 04 Aug 2007 12:24:17 -0700, Grant Erwin wrote:
Ignoramus12500 wrote:
On Sat, 4 Aug 2007 09:11:30 -0700, azotic wrote:


Thanks for the tip, the rotary phase converters don't have the
flexability i want and are to noisy for me. I was thinking about
using one of my old note book computers interfaced to some power
semiconductors and hopefully end up with a user programable vfd.



Do not underestimate the cost of this project, and you will end up
with something rather substandard. But it is fun.

For example, single phase input rated, 1 horsepower VFDs can be bought
on eBay for about $47 plus $12 shipping. It is hard to beat that.

i

Got an example of that price point, Igor?

Well, check out eBay item 180144202452, "ALLEN BRADLEY AC DRIVE".

Allen Bradley VFD
CAT NO. : 160S-AA04NSF1P1

The manual for it is he

http://igor.chudov.com/manuals/Allen...VFD-Manual.pdf

Check page 14 of it to see that it is a Bulletin 160, single phase
input, 240v, 1 HP, open enclosure, analog signal follower, P1
(includes programmer keypad).

It went for $47, the seller charged me $12 for shipping. It would sell
for less, if I had ability to snipe at the last second, which I did
not because I was with my 1.5 year old after he took a bath. So I
placed my last bid 3 minutes before auction end.

Beware that 160-B* drives are 380-460 volts only. You need 160-A*. The
letter S after 160 denotes single phase input.

i

On Sat, 04 Aug 2007 20:50:36 -0500, Ignoramus12500 wrote:
It went for $47, the seller charged me $12 for shipping. It would sell
for less, if I had ability to snipe at the last second, which I did
not because I was with my 1.5 year old after he took a bath. So I
placed my last bid 3 minutes before auction end.

Forgot to say, my snipe program did not work at the time due to ebay
changes.

i

Ignoramus12500 wrote:

On Sat, 04 Aug 2007 12:24:17 -0700, Grant Erwin wrote:

Ignoramus12500 wrote:

On Sat, 4 Aug 2007 09:11:30 -0700, azotic wrote:



Thanks for the tip, the rotary phase converters don't have the
flexability i want and are to noisy for me. I was thinking about
using one of my old note book computers interfaced to some power
semiconductors and hopefully end up with a user programable vfd.



Do not underestimate the cost of this project, and you will end up
with something rather substandard. But it is fun.

For example, single phase input rated, 1 horsepower VFDs can be bought
on eBay for about $47 plus $12 shipping. It is hard to beat that.

i

Got an example of that price point, Igor?


Well, check out eBay item 180144202452, "ALLEN BRADLEY AC DRIVE".

Allen Bradley VFD
CAT NO. : 160S-AA04NSF1P1

The manual for it is he

http://igor.chudov.com/manuals/Allen...VFD-Manual.pdf

Check page 14 of it to see that it is a Bulletin 160, single phase
input, 240v, 1 HP, open enclosure, analog signal follower, P1
(includes programmer keypad).

It went for $47, the seller charged me $12 for shipping. It would sell
for less, if I had ability to snipe at the last second, which I did
not because I was with my 1.5 year old after he took a bath. So I
placed my last bid 3 minutes before auction end.

Beware that 160-B* drives are 380-460 volts only. You need 160-A*. The
letter S after 160 denotes single phase input.






Some of these drives are for only digital control rather than an analog
input. Some come without the programming contol pad on the front. Dont
buy the B type drives, thats the ones I use.


John

On Sat, 04 Aug 2007 22:13:43 -0400, john wrote:


Ignoramus12500 wrote:

On Sat, 04 Aug 2007 12:24:17 -0700, Grant Erwin wrote:

Ignoramus12500 wrote:

On Sat, 4 Aug 2007 09:11:30 -0700, azotic wrote:



Thanks for the tip, the rotary phase converters don't have the
flexability i want and are to noisy for me. I was thinking about
using one of my old note book computers interfaced to some power
semiconductors and hopefully end up with a user programable vfd.



Do not underestimate the cost of this project, and you will end up
with something rather substandard. But it is fun.

For example, single phase input rated, 1 horsepower VFDs can be bought
on eBay for about $47 plus $12 shipping. It is hard to beat that.

i

Got an example of that price point, Igor?


Well, check out eBay item 180144202452, "ALLEN BRADLEY AC DRIVE".

Allen Bradley VFD
CAT NO. : 160S-AA04NSF1P1

The manual for it is he

http://igor.chudov.com/manuals/Allen...VFD-Manual.pdf

Check page 14 of it to see that it is a Bulletin 160, single phase
input, 240v, 1 HP, open enclosure, analog signal follower, P1
(includes programmer keypad).

It went for $47, the seller charged me $12 for shipping. It would sell
for less, if I had ability to snipe at the last second, which I did
not because I was with my 1.5 year old after he took a bath. So I
placed my last bid 3 minutes before auction end.

Beware that 160-B* drives are 380-460 volts only. You need 160-A*. The
letter S after 160 denotes single phase input.






Some of these drives are for only digital control rather than an analog
input. Some come without the programming contol pad on the front. Dont
buy the B type drives, thats the ones I use.

I thought that analog signal means potentiometer? No?

i

"Grant Erwin" wrote in message
...
Start with your output phase, your "power semiconductors". If you can find
some of those that meet your spec at a price point you can live with,
that's one thing. I'm guessing that buying a small handful of them will
get you a price much higher than an entire VFD made by TECO or someone
else.

Igor, I'm still waiting for that $47 1hp VFD new on ebay ..

Grant

Agreed the power stage is the heart of a vfd, i have a variety of mosfets,
igbt and scrs
to play with. This will be a learning experience, been reading all the
available information
from the semiconductor manufactures and online textbooks, now i would like
to experiment
with some of designs i studied and test some of my own ideas.

Best Regards
Tom.

"BobH" wrote in message
...
Grant Erwin wrote:
Back when I was working as an electronics engineer in San Jose,
California (heart of Silicon Valley) I came up with a design for this. It
was a chip which produced variable PWM waveforms. I planned to use power
Darlington transistors to go between the CMOS logic chip and the gates of
beefy SCRs. I got as far as simulating the whole design in a mixed signal
simulator.

Then I found out what the failure mode of SCRs is. About the same as hand
grenades. Did I want this in my house? No.

I scrapped the design, never looked back.

Now the power devices will probably be Insulated Gate Bipolar Transistors,
but the failure mode is the same. 240VAC Rectifies to over 300VDC and a
multi-horsepower drive will need many amps of DC supply capability. I have
built some smaller drives and from that my advice is:
A) keep a CO2 fire bottle handy.
B) always wear safety glasses.
C) never leave the room with a prototype energized.
D) buy lots of spare transistors and drivers.
E) don't try to build with point to point wiring, parasitics will kill it.
F) keep a CO2 fire bottle handy.


In the few commercial VFD's I have played with programmability was never
in short supply. Figuring out all the programability they gave you was
tough enough.

That said, doing it is still one of the best ways to learn about a new
area of technology. I have been kicking around the idea of designing and
building an inverter welding machine for a while. It would not replace my
commercial machine for a long time, but it would be a heck of a project. I
wonder if argon puts out fires as well as CO2? (grin)

Bob


Saftey is a main concern, so i will be expermenting at reduced power levels
and gradualy step up the power. I did download some schematics from
esab for thier inverter welders. Learned a lot from studing thier
schematics.

Heres a link to esabs document library:

http://www.esabna.com/html/downloads...rc%20Equipment

Best Regards
Tom.

On Sat, 4 Aug 2007 20:06:33 -0700, azotic wrote:

"Grant Erwin" wrote in message
...
Start with your output phase, your "power semiconductors". If you can find
some of those that meet your spec at a price point you can live with,
that's one thing. I'm guessing that buying a small handful of them will
get you a price much higher than an entire VFD made by TECO or someone
else.

Igor, I'm still waiting for that $47 1hp VFD new on ebay ..

Grant

Agreed the power stage is the heart of a vfd, i have a variety of mosfets,
igbt and scrs
to play with. This will be a learning experience, been reading all the
available information
from the semiconductor manufactures and online textbooks, now i would like
to experiment
with some of designs i studied and test some of my own ideas.


They sell special inverter semiconductor blocks for VFDs, with six
transistors on one block packaged exactly for this application. They
show up regularly on ebay.

I think that microcontroller really is the heart of a VFD, it would
not be easy to program.

i

On Sat, 4 Aug 2007 20:06:33 -0700, azotic wrote:

"Grant Erwin" wrote in message
...
Start with your output phase, your "power semiconductors". If you can find
some of those that meet your spec at a price point you can live with,
that's one thing. I'm guessing that buying a small handful of them will
get you a price much higher than an entire VFD made by TECO or someone
else.

Igor, I'm still waiting for that $47 1hp VFD new on ebay ..

Grant

Agreed the power stage is the heart of a vfd, i have a variety of mosfets,
igbt and scrs
to play with. This will be a learning experience, been reading all the
available information
from the semiconductor manufactures and online textbooks, now i would like
to experiment
with some of designs i studied and test some of my own ideas.

Tom, without any doubt, you should buy a VFD, and a three phase motor,
to play with, in order to make your own. Your cost for a very small
combo should be negligible compared to the time wasted making guesses
that could be easily clarified with an existing drive.

i

"Ignoramus12500" wrote in message
news

I think that microcontroller really is the heart of a VFD, it would
not be easy to program.

i

Rather than using a microcontroller for experimenting i intend to use one
of my old notebook computers, interface it to the power block and
write some sofware to controll the power block thru one of computers
I/O ports.

Best Regards
Tom.

"Ignoramus12500" wrote in message
...
On Sat, 4 Aug 2007 20:06:33 -0700, azotic wrote:

Tom, without any doubt, you should buy a VFD, and a three phase motor,
to play with, in order to make your own. Your cost for a very small
combo should be negligible compared to the time wasted making guesses
that could be easily clarified with an existing drive.

i


I do have some comercial VFD"s and a couple of 3-Phase motors, and your
right
comparison is always a good idea. At this time i want to learn thru doing,
time and
cost are not a factor when have fun.

Best Regards
Tom.

azotic wrote:

I do have some comercial VFD"s and a couple of 3-Phase motors, and your
right
comparison is always a good idea. At this time i want to learn thru doing,
time and
cost are not a factor when have fun.
Definitely check out International Rectifier. They have a bunch
of reference designs and kits for making VFDs in a variety of
sizes. These are for both PM and induction motors. The
reference designs can be bought with PC boards, but the
published data sheets on them have not only schematics but PC
board layouts for you to study.

Jon

azotic wrote:
I do have some comercial VFD"s and a couple of 3-Phase motors, and your
right
comparison is always a good idea. At this time i want to learn thru doing,
time and
cost are not a factor when have fun.
Oh, yeah, Don Lancaster has some REALLY interesting stuff he did
a few years ago. he called it "magic sine waves". It is mostly
a scheme for storing families of waves in EPROMS to synthesize
sine waves with the minimum number of transitions (less
transitions = less energy loss in the power transistors).
The scheme stores just a single-bit sequence in the EPROM for
every combination of frequency and voltage.

Jon

"Jon Elson" wrote in message
...
azotic wrote:
I do have some comercial VFD"s and a couple of 3-Phase motors, and your
right
comparison is always a good idea. At this time i want to learn thru
doing, time and
cost are not a factor when have fun.
Oh, yeah, Don Lancaster has some REALLY interesting stuff he did a few
years ago. he called it "magic sine waves". It is mostly a scheme for
storing families of waves in EPROMS to synthesize sine waves with the
minimum number of transitions (less transitions = less energy loss in the
power transistors).
The scheme stores just a single-bit sequence in the EPROM for every
combination of frequency and voltage.

Jon


That sounds interesting, push pull power block driven with magic sine waves.
Gotta give it a try down the road.

Best Regards
Tom.

azotic wrote:

Saftey is a main concern, so i will be expermenting at reduced power levels
and gradualy step up the power. I did download some schematics from
esab for thier inverter welders. Learned a lot from studing thier
schematics.

Heres a link to esabs document library:

http://www.esabna.com/html/downloads...rc%20Equipment

Best Regards
Tom.


Thanks and good Luck,
Bob

"Jon Elson" wrote in message
...
azotic wrote:

I do have some comercial VFD"s and a couple of 3-Phase motors, and your
right
comparison is always a good idea. At this time i want to learn thru
doing, time and
cost are not a factor when have fun.
Definitely check out International Rectifier. They have a bunch of
reference designs and kits for making VFDs in a variety of sizes. These
are for both PM and induction motors. The reference designs can be bought
with PC boards, but the published data sheets on them have not only
schematics but PC board layouts for you to study.

Jon


I visit IR on a regular basis, in particular i like thier app notes on
mosfet
push-pull designs. Seems that it may be possible to generate sine waves
of sufficiant power to drive small motors. I will be experimenting with that
concept down the road. Microchip also has some usefull app notes.

Best Regards
Tom.

well, I for one would be astounded if a commercial notebook running windows
or linux could keep up with the required IO for this task- start by deciding
what your refresh rate for the waveform is - the waveform itself will be
somewhere below 400 hz, but you will want to sample current and voltage and
calcuate what to do at maybe 40 khz to 400 khz - how will you do that on a
machine whose basic "fast" interrupt is 1.8 ms?


"azotic" wrote in message
...

"Ignoramus12500" wrote in message
news

I think that microcontroller really is the heart of a VFD, it would
not be easy to program.

i

Rather than using a microcontroller for experimenting i intend to use one
of my old notebook computers, interface it to the power block and
write some sofware to controll the power block thru one of computers
I/O ports.

Best Regards
Tom.







--
Posted via a free Usenet account from http://www.teranews.com

"William Noble" wrote in message
...
well, I for one would be astounded if a commercial notebook running
windows or linux could keep up with the required IO for this task- start
by deciding what your refresh rate for the waveform is - the waveform
itself will be somewhere below 400 hz, but you will want to sample current
and voltage and calcuate what to do at maybe 40 khz to 400 khz - how will
you do that on a machine whose basic "fast" interrupt is 1.8 ms?



The computer will not have to sample voltage or current if i decide to
interface to a IR2130.
Proportional voltage current controll is a function of the IR2130 so i dont
need a fast interrupt.

Best Regards
Tom.

azotic wrote:
Anybody hear ever try to build a 3 phase vfd ?

I am in the mood to build something that i could buy ready made cheaper.

When the amount you've spent on destroyed power
transistors approaches about 2/3 the cost of a
new VFD, it's time to bale. Don't ask how I know....

BobH wrote:
Grant Erwin wrote:
Back when I was working as an electronics engineer in San Jose,
California (heart of Silicon Valley) I came up with a design for this.
It was a chip which produced variable PWM waveforms. I planned to use
power Darlington transistors to go between the CMOS logic chip and the
gates of beefy SCRs. I got as far as simulating the whole design in a
mixed signal simulator.

Then I found out what the failure mode of SCRs is. About the same as
hand grenades. Did I want this in my house? No.

I scrapped the design, never looked back.

Now the power devices will probably be Insulated Gate Bipolar
Transistors, but the failure mode is the same. 240VAC Rectifies to over
300VDC and a multi-horsepower drive will need many amps of DC supply
capability. I have built some smaller drives and from that my advice is:
A) keep a CO2 fire bottle handy.
B) always wear safety glasses.
C) never leave the room with a prototype energized.
D) buy lots of spare transistors and drivers.
E) don't try to build with point to point wiring, parasitics will kill it.
F) keep a CO2 fire bottle handy.

G) Keep the spouse/kids/pets out of the lab when
powering it up.

H) Show the spouse how to cut off the power quickly
if you start screaming "cut off the power"

I) Don't underestimate the danger of a plasma fire.