Best Practice: TVS Diode vs MOV for Inductive Load in MOSFET AC Switch Circuit?

12 6

#1 21682703 11 Jun 2019 23:24

Matthew Wood Matthew Wood

Anonymous
Post #1
21682703 11 Jun 2019 23:24

I have searched for this simple answer and I can't seem to come up with a good clear answer. In the supplied circuit picture, how can this circuit be hardened for an inductive load such as a ceiling fan? Can I put a TVS diode or MOV across the 'Load Hot' and 'Neutral' terminals? I have seen both mentioned, but then both are also said to be wrong. I have seen that TVS diodes is the way to go but might not be UL approved. Doesn't have to be, but wanted to build on best practice. MOV's appear to be the right answer, but have seen people say they are too slow for this application. I have also seen RC snubber designs but I know this can cause other issues with small loads. Not sure. Please help!

Attachments:

forums-5160.jpg Download (21.47 KB)
ADVERTISEMENT
#2 21682704 13 Jun 2019 02:02

Aubrey Kagan Aubrey Kagan

Anonymous

Post #2
21682704 13 Jun 2019 02:02

I believe all 3 are possible with some qualification.You can connect the suppression across the load or across the transistors. Depending on the voltage you may be able to find a non-polarised TVS at above your maximum AC operating voltage. Both a MOV and a TVS have the same problem though- after a number of hits they start to degrade. Their failure mode can be of concern- if they go open circuit, you have no idea- the circuit will continue to operate until the transistors or IC blow up. if they go short circuit- well, BANG.An RC snubber across the transistors will lead to a small leakage current when the transistors are off and this may be enough to energise or at least partially energise your load. This articlemay be of some use.I would reccomend an RC snubber across the load.You may want to look at my blog Back EMF and snubber
ADVERTISEMENT
#3 21682705 13 Jun 2019 03:32

David Ashton David Ashton

Anonymous

Post #3
21682705 13 Jun 2019 03:32

I'd be a fan of putting the MOV/TVS BEFORE the circuit (ie on the line side) so that they catch any spikes before they enter your circuit.Aubrey and I come from Zimbabwe which has mean lighting during the rainy season. I used to use mains distribution power boards for my equipment, and my supplier proudly had in his window an opened up power board which was blackened and the MOVs had split open, but apparently the equipment supplied had all survived. He said another useful trick is to put a couple of turns in the supply wire before the MOV if you can, it gives a bit of inductance and that helps as well. He told me too much - I started making my own boards, but I did buy the MOVs from him :-)You can test MOVs - get an insulation tester that will provide 500V at least, and put the MOVs across the test probes. You should get a deflection inversely proportional to the MOV voltage (ie a 250V MOV should give you around 1/2, a 400V MOV will give you about 1/5 deflection of the needle.). No deflection means they are open circuit. Full deflection means they are shorted.
#4 21682706 14 Jun 2019 01:01

Rick Curl Rick Curl

Anonymous

Post #4
21682706 14 Jun 2019 01:01

Hi Matthew-I'm leaning more towards Aubrey's answer on this one- put an RC snubber across the load.The goal here is not to stop spikes before they get into your circuit- the object is to mitigate the spikes coming from the inductive winding of the motor when the IGBT's (or MOSFETS) turn off.This is one of those cases where lots of trial and error may be involved since you don't know the characteristics of the fan motor. You might start by getting one of the capacitors used in ceiling fans. Many of them have several sections of different values. Put a small series resistance (maybe 100 ohms at 5 watts or so) in series with the capacitor and then place the resistor and capacitor across the motor winding. Try different capacitance values until you get good results.You'll need to use an oscilloscope to find out how big the spikes are. BE VERY CAREFUL since oscilloscopes have the case grounded. You might have to use two high voltage 'scope probes and put the oscilloscope in differential mode.-Rick
ADVERTISEMENT
#5 21682707 16 Jun 2019 01:53

Matthew Wood Matthew Wood

Anonymous

Post #5
21682707 16 Jun 2019 01:53

I had done RC snubbers before on TRIAC based AC switch circuits and at the moment I am drawing a blank as to why I thought I couldn't do that across a MOSFET based one. I have seen that littelfuse document before actually and that was a part of my original confusion. lol If I remember correctly I was using a 100ohm resistor and a .1uf capacitor. I will double check that but I have those TRIAC switches on ceiling fans and never had an issue. Thank you for your time so far! Any other thoughts, I'm all ears!
ADVERTISEMENT
#6 21682708 16 Jun 2019 01:56

Matthew Wood Matthew Wood

Anonymous

Post #6
21682708 16 Jun 2019 01:56

Thanks for your answer but I think you miss understand my problem. You are correct about the MOV's before. I will do that. I'm concerned about the "kickback" from an inductive load like a ceiling fan motor on the MOSFET switch. I tried one of my older designs using MOSFET's on a ceiling fan and as expected, it turned on fine, but first time off...bang. One less MOSFET.
#7 21682709 16 Jun 2019 01:59

Matthew Wood Matthew Wood

Anonymous

Post #7
21682709 16 Jun 2019 01:59

Thanks Rick, yeah you're right with Aubrey. I did 100ohm with .1uf capacitor in a TRIAC based switch but some reason I was thinking this wouldn't work with a MOSFET based switch. I think I might just be overthinking it.
Create an account, log in here. You will receive points by participating in discussions.
Join this discussion.

Install Elektroda application

Didn't find an answer? Ask Artificial Intelligence

*I agree to send the question to OpenAI, Anthropic PBC, Perplexity AI, Inc., Kagi Inc., Google LLC - owners of language models in order to prepare the best response. The companies may monitor and log information entered into the form.

*I agree to publicly display my question and answer. The question and answer will be publicly available to everyone. The process may take a few minutes. Upon completion, you will be redirected to the page with the answer.

Wait...(2min)

Topic summary

The discussion addresses best practices for protecting MOSFET AC switch circuits driving inductive loads such as ceiling fans from voltage spikes caused by load inductance. Both TVS diodes and MOVs are considered for suppression across the load terminals, but each has drawbacks: both degrade after repeated surges and can fail either open or short circuit, posing reliability concerns. MOVs are commonly placed on the line side to clamp incoming surges, sometimes with added inductance from supply wire turns to improve effectiveness. TVS diodes may be faster but are less commonly UL approved. RC snubber circuits across the load are recommended to mitigate back-EMF spikes generated when MOSFETs turn off, with typical values around 100 ohms and 0.1 µF capacitors, similar to those used in TRIAC-based switches. Oscilloscope measurements are advised to optimize snubber parameters. The goal is to protect the MOSFET from destructive voltage spikes without causing leakage current that could partially energize the load. Trial and error is often necessary due to varying motor characteristics.
Summary generated by the language model.

Best Practice: TVS Diode vs MOV for Inductive Load in MOSFET AC Switch Circuit?

Didn't find an answer? Ask Artificial Intelligence

Topic summary