Jump to content

Are relays capable of PWM.


rtn1

Recommended Posts

Hey. I am trying to control 200 volts using PWM. I understand that DAQ boards that i know of cannot produce 200 volts so i would need a seperate DC power supply. The only way of turning on and off this 200 volt source using labview that i know of is a relay, but i am wondering if there is a relay fast enough to process the PWM. Is there any other way of varying a voltage as high as 200v using labview? Any information will be helpful. Thanks in advance.

Link to comment

QUOTE(rtn1 @ Feb 8 2008, 09:31 AM)

Hey. I am trying to control 200 volts using PWM. I understand that DAQ boards that i know of cannot produce 200 volts so i would need a seperate DC power supply. The only way of turning on and off this 200 volt source using labview that i know of is a relay, but i am wondering if there is a relay fast enough to process the PWM. Is there any other way of varying a voltage as high as 200v using labview? Any information will be helpful. Thanks in advance.

that depends of the base frequency of your generated PWM signal and the switching time of your relays. If your signal is very slow (e.g. 10 Hz) a solid state relay may be fast enough. If you use e.g. 20 kHz (which is a common base frequency of frequency-converters ...) a relay will not be fast enough. In that case you should think about useing MOSFETs or IGBTs, but that is a science of its own.

Link to comment

QUOTE(i2dx @ Feb 8 2008, 12:47 AM)

you should think about using MOSFETs or IGBTs

Make sure you check the heat dissipation requirements. With that kind of voltage switching through a MOSFET you'll probably need a good heatsink - Still I think this would be a much better choice than a relay.

Link to comment

QUOTE(AnalogKid2DigitalMan @ Feb 8 2008, 02:42 PM)

What type of load or device are you driving with the 200V PWM?

Current requirements?

I think these are really important information to consider. Making a circuit to drive a motor at 200V with a large current may cost you many MOSFETs for testing happy.gif. In this case you should use a heatsink with a dimension correspondent to the current.

Link to comment

QUOTE(rtn1 @ Feb 8 2008, 10:31 AM)

Hey. I am trying to control 200 volts using PWM. I understand that DAQ boards that i know of cannot produce 200 volts so i would need a seperate DC power supply. The only way of turning on and off this 200 volt source using labview that i know of is a relay, but i am wondering if there is a relay fast enough to process the PWM. Is there any other way of varying a voltage as high as 200v using labview? Any information will be helpful. Thanks in advance.

my salt: this is not a LV question, much rather a hardware one. if you have little experience with relaying, better get the help of an electronician. MOSFETS, and especially IGBTs, are expensive and relatively much more complicated equipment than simple relays. but for your voltage and frequency range there is not much choice. for higher currents as already pointed out you have not much choice but a IGBT. be aware tough, that they require separate power supplys, and that they tend to break very easily if not handled with great care in designing your circuit. for example, if the induced magnetic field is high, then you can expect a 'backcurrent' at turn off of the IGBT, which could lead to a surge of voltage to very high values. IGBTs in general cannot handle more than a thousand volts or so, (not to mention health and risks to living organisms...), therefore should be protected with varistors. These in turn have a breakdown voltage proportional to the current, so you should also look into the datasheet of those and choose them carefully. They also induce a longer shut down time, (depending on max voltage, current and induced field), limiting the overall rate of switching.

to your consideration...

Link to comment

QUOTE(Gabi1 @ Feb 8 2008, 10:42 PM)

my salt: this is not a LV question, much rather a hardware one. if you have little experience with relaying, better get the help of an electronician. MOSFETS, and especially IGBTs, are expensive and relatively much more complicated equipment than simple relays. but for your voltage and frequency range there is not much choice. for higher currents as already pointed out you have not much choice but a IGBT. be aware tough, that they require separate power supplys, and that they tend to break very easily if not handled with great care in designing your circuit. for example, if the induced magnetic field is high, then you can expect a 'backcurrent' at turn off of the IGBT, which could lead to a surge of voltage to very high values. IGBTs in general cannot handle more than a thousand volts or so, (not to mention health and risks to living organisms...), therefore should be protected with varistors. These in turn have a breakdown voltage proportional to the current, so you should also look into the datasheet of those and choose them carefully. They also induce a longer shut down time, (depending on max voltage, current and induced field), limiting the overall rate of switching.

to your consideration...

sounds like you have some experience with such circuits?

I'm still searching for someone who could give me a little input about how to build a frequency converter circuit (20 kHz base frequency, 1-35V, 0-150A, 0-240 Hz) which is directly connected to a cRIO 9401 DIO Module (which generates the PWM) ...

Link to comment

First off i like to thank everybody in this forum that has given me help, it is much appreciated. the 200v will be running 16 devices connected in series with a current requirement of around 10-15 amps. I guess it looks like i am going to have to do a lot of research on MOSFETS and IGBT. If anybody has any good sites that has a great explanation of these devices it would be much appreciated. All and all thank yall once again. Yall help is more than expected.

Link to comment

Hi Rtn1:

Coming in late on this topic, but would like to put in my 2 cents...

I've done some testing for a company that brewed its own IGBT based PWM variable speed motor controllers-- at power ratings well above what you are talking about but still your application is somewhat high power. I'm not totally clear on the finer points of designing these circuits, but I do know a bit about testing them and it is an activity that definitely requires safety glasses, ear protection and lots of time and money... You usually can't think of them as a simple on-off switch, but have to briefly run them in linear mode during the switching transitions. Switch them too fast and they make really nasty broadband electrical noise, which might crash the electronics controlling them, or make voltage spikes that can blow up whatever you are controlling. Switch them too slow, and they overheat and explode.

If you're working towards something that you are going to build by the hundreds, you might consider brewing your own-- find someone with experience in the power electronics field and put them on the payroll. If you only want to build a few of whatever you are building, you are better off buying something ready-made. There are a number of companies out there-- Trace Inverters now owns the rights to the work I was involved with years ago. Copley Controls makes some nice motor controllers which interface nicely with LabVIEW which I've used recently-- and I think they also make other types of PWM power supplies. Lots of other companys, I think even NI markets motor drives, which might be adapted to your needs.

Best luck-- If you can't buy the ready made, at least buy the safety glasses.

Best Regards, Louis

Link to comment

QUOTE(AnalogKid2DigitalMan @ Feb 11 2008, 05:44 PM)

If the drive signal orignates from a PC, optical or galvanic isolation would be wise on the input stage of the PWM drive.

Pah, isolation is for cowards ;) That 9401 Module does not cost that much anyway and NI will love me, if I order them in dozens ;)

seroius again: yepp, sounds like a good idea!

Link to comment

QUOTE(i2dx @ Feb 9 2008, 09:59 AM)

sounds like you have some experience with such circuits?

A little bit. i use magnetic coils and high currents.

QUOTE

I'm still searching for someone who could give me a little input about how to build a frequency converter circuit (20 kHz base frequency, 1-35V, 0-150A, 0-240 Hz) which is directly connected to a cRIO 9401 DIO Module (which generates the PWM) ...

are you offering me a job? :laugh:

, you better call a real electronician if you are afraid, but it is not that difficult, if you follow all the datasheets. but the "directly connected to a cRIO" is a dont do at home stuff.

in general:

the IGBT should be switched using an optocoupler, even tough it necessite an external power supply. on the bright side, you probably can use same power supply to drive the IGBT (plus the one on the cRIO side of the connection). the IGBTs i buy generally are from Mitsuishi, and they do the job nicely. switching on should be digital, the IGBT should be cooled using a water cooled plate glued at the bottom (the IGBT eats up about 3V of voltage, independant on the current. you can see that it is fast reaching high temperatures). for switching off, the IGBT should always be fully protected by a varistor (put several in parallel, as they tend to die fast). for 20KHz, this should definitely not be a problem.

notes:

1) in general, it is much easier to modulate the current trough the current supply rather than the IGBT, and use the IGBT only as a switch. in this way, one avoids lots of troubles. altough limiting the capabilities, complex waveforms at very high frequencies cannot be achieved without completely taking in account the whole circuit ( due to induced fields and so), therefore most of the time the relatively slow response of the power supplies (order of ms) is enough for such modulations. on the other hand, IGBTs used as switched have proved their usefullness for example in hybrid cars (DC power supply to AC motors). the power supplies i like most are from DELTA. expensive but very reliable.

2) cool the IGBT!!!

3)the driving circuit for the IGBT is easy to build, and you should find an example circuit in the datasheet of the IGBT itself.

4)use varistors all around in your circuit. they are cheap and easy to plug everywhere: between the nodes of the power supply (check max voltage it can withhold), between the nodes of the IGBT (use a good margin voltage for breackdown - your circuit will probably not have too high an inductance, therefore you should not be too worried about longer switchoff times - but calculate to be sure).

5) for testing, drive your system at low current. you minimise risks to burning the IGBT to zero, and they are expensive stuff.

6) the whole thing will take you some time (actually quite a bit). if you dont expect to build too much of those, much better is to buy a finished solution!! when one think that a worker costs 6000Euro per month, plus material and long learning curve...

7) good luck! but arent you more of a Labview guy?

Link to comment

QUOTE(Gabi1 @ Feb 12 2008, 09:53 PM)

A little bit. i use magnetic coils and high currents.

are you offering me a job? :laugh:

[...]

7) good luck! but arent you more of a Labview guy?

WHOW, thanks a LOT!

maybe I could offer you a job, but it's a long way until then. First I have to write my business plan and so on and so on ... ;)

--> 7.) yes, I am a LabVIEW guy, and more severly: I know allmost nothing about how to build a circuit. Ok, it's still enough to impress my little sister, but I'm an absolute no0b compared to an "interested amateur".

If there is allready any finished solution which can act as a frequency converter circuit (20 kHz base frequency, 1-35V, 0-150A, 0-240 Hz) and create a 3 Phase rotary field for AC motors, and HAS DIRECT ACCESS to the MOSFETS/IGBTS through digital inputs, please let me know. This solution has not to be able to convert the Voltage, I just want to switch the current. All finished solutions I could find so far are "ready to run" frequency converters with it's own µC, but I want to control the frequency by myself (in my FPGA VI)

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.