DIY Electric Car Forums banner

101 - 120 of 195 Posts

·
Registered
Joined
·
952 Posts
Kenny - yeah, I know. When I marked those bus bars it was because I was figuring their length for the installation as I did it. But you are right, and this is my first real experience with such high power devices. I will be re-building it more to the way it is described in Paul's design. Also though, Paul's design utilizes the center post of the IGBT, connecting it to the CAP Neg - although it's hard for me to understand why. But my setup didn't work at all until I made that connection. So he's actually using all three Posts.

But I got Paul's board, and it's set up with three IGBT's so that's what I have to work with. It's a pricey lesson, no doubt.

I'm still excited about it though. Right now I just want to know if I can use these:
https://www.ebay.com/itm/igbt-trans...h=item441d927f72:g:nlMAAOSwjqVZE4nD:rk:2:pf:0

Because it's a pretty good deal and it would cost about the same to replace the single IGBT I burned.
-Ocean
Yes gate driver is the first to go after IGBT since it is shorted out.
Diodes you are talking about are zener diodes. The reason why you would get +8V are two diodes after the filter transformer. They are there to cut the voltage from 24V to +15V / -8V.
Well if they are burned divider would behave differently now. I suggest you replace them as well.

You should just replace every semiconductor with this gate. Usually caps and resistors are ok, but diodes and transistors are risky.

Again i would recommend you ask Paul how to tweak HW overcurrent protection. Also you need to be sure what the current sensor setting is.
Usually that means changing current-sensor-amps-per-volt setting in serial interface. If it doesnt want to accept new value than this is out of scope in firmware. You will have to correct in firmware and upload it with Pickit3.
Allways backup previous version somewhere!

A
 

·
Registered
Joined
·
152 Posts
Thanks Arber for the insight about the +8v shift. I will look on the board for those voltage splitting diodes. Any chance you have a list of the exact components on the Gate side of the board? I'm color blind so that does not help.... I was figuring on using a magnifying glass to figure out the numbers...

I've asked Paul a lot of questions lately so I'm going to let him be for a while... maybe once I get it put back together I will ask him about tweaking it as you suggested.

Meanwhile, I need to replace my burned out IGBT. $122 from Mouser. And it's a 300A Littlefuse which I can get... BUT I saw these on Ebay for only $40 each:

https://www.ebay.com/itm/igbt-trans...h=item441d927f72:g:nlMAAOSwjqVZE4nD:rk:2:pf:0

What do you think??? If they would be compatible, I'd get 3 and have a new set, and save the others for the next build.

Do you think they would work with Paul's board? I think they should but am I missing something?

Yes gate driver is the first to go after IGBT since it is shorted out.
Diodes you are talking about are zener diodes. The reason why you would get +8V are two diodes after the filter transformer. They are there to cut the voltage from 24V to +15V / -8V.
Well if they are burned divider would behave differently now. I suggest you replace them as well.

You should just replace every semiconductor with this gate. Usually caps and resistors are ok, but diodes and transistors are risky.

Again i would recommend you ask Paul how to tweak HW overcurrent protection. Also you need to be sure what the current sensor setting is.
Usually that means changing current-sensor-amps-per-volt setting in serial interface. If it doesnt want to accept new value than this is out of scope in firmware. You will have to correct in firmware and upload it with Pickit3.
Allways backup previous version somewhere!

A
 

·
Registered
Joined
·
952 Posts
Thanks Arber for the insight about the +8v shift. I will look on the board for those voltage splitting diodes. Any chance you have a list of the exact components on the Gate side of the board? I'm color blind so that does not help.... I was figuring on using a magnifying glass to figure out the numbers...

I've asked Paul a lot of questions lately so I'm going to let him be for a while... maybe once I get it put back together I will ask him about tweaking it as you suggested.

Meanwhile, I need to replace my burned out IGBT. $122 from Mouser. And it's a 300A Littlefuse which I can get... BUT I saw these on Ebay for only $40 each:

https://www.ebay.com/itm/igbt-trans...h=item441d927f72:g:nlMAAOSwjqVZE4nD:rk:2:pf:0

What do you think??? If they would be compatible, I'd get 3 and have a new set, and save the others for the next build.

Do you think they would work with Paul's board? I think they should but am I missing something?
I used 400A IGBTs in my built of Pauls board. So they should be good. It was DIP version. I have the BOM for that, but not for SMD variant.
I am still convinced that you somehow did not use correct limit for current sensors.

I think the best would be to connect all C protection wires (+) to a single point as well as all E wires (-). Then you disconnect other IGBTs, you only use one and try this for your DC motor. I am not sure if dsPic will mind if there is only a single current sensor on line.... It definitely has HW desat protection.
 

·
Registered
Joined
·
152 Posts
I used 400A IGBTs in my built of Pauls board. So they should be good. It was DIP version. I have the BOM for that, but not for SMD variant.
I am still convinced that you somehow did not use correct limit for current sensors.

I think the best would be to connect all C protection wires (+) to a single point as well as all E wires (-). Then you disconnect other IGBTs, you only use one and try this for your DC motor. I am not sure if dsPic will mind if there is only a single current sensor on line.... It definitely has HW desat protection.
Arber, I can barely keep up with your language. I'm working with a P&S DC Control Board that was built and populated by Paul. So I don't really know what DIP version is, or what BOM is, or SMD or dsPic. I know these are important terms but I don't really understand them. And I don't have a PICT.

I'd like the boards to work as they are. I assume Paul put the latest software on the board before he sent them to me (I got two at the same time) - about a year ago.

I have not even been able to successfully connect with Serial! And I have tried - really tried.

Paul indicated that he set the maximum current to 330 for each IGBT (if I remember correctly - somewhere back in my email thread). When I explained what happened to Paul, he suggested that the problem was a voltage spike, due to this formula :

V = di/dt * L

He said this:

"Under high current, you have a large di/dt (change in current over time). That can't be helped. The layout caused L to be too big (the stray inductance). L is related to the loop area of the current path. So,you had a large voltage spikes, which caused the insulation to break down inside the igbt, which took some of the driver sections with it on the board."

And when I look back at my layout, I can see what he means... because I had different length cables to different IGBT's both going in and out - my big mistake - and the IGBT that burned out was the one on the shortest path. I think perhaps it was not burned by overcurrent, but by a voltage spike going over 600v (the rating of the IGBT) and finding it's way up into the board taking the Gate driver with it.

ANYWAYS, what I hear you suggest is to try and make it work with 1 IGBT. Connecting all the current sensors wires together??? But if I do this, then I only have 300amps of potential @70degC since that's what I have left - two 300a IGBT's.

Yeah I want to make it work correctly. I understand your idea that it makes sense to have one DC motor driven by 1 big IGBT - but I don't have 1 big IGBT. Also I'm not so sure about hacking Paul's board until I really understand it better. For example, if the three current sensors are put all in parallel all to the same point, would than not confuse the board? Paul indicated that they need to be connected properly or the board see's a hardware fault.

ONE possibility I can see is to put all three current sensors on one single output from one IGBT - lining them all up in a row so they all sense the output of a single IGBT, but they are feeding back to the three sensor inputs on the board. Leaving the other IGBT's disconnected (either removed completely, or still plugged into the board but with no big cables connected). I sill wonder though if the Gates are sensing anything through their connection to the IGBT and if two were missing, would the Gate's feed back a HW fault to the brain of the board?

Thanks for your thoughts Arber - and also for the feedback on that Ebay listing.

-Ocean
 

·
Registered
Joined
·
329 Posts
Before you make any changes, it would be best to build the controller exactly as Paul did in his instructible. It works very well. If you're struggling with the serial, you may need a different cable or to change the bit rate and format in your terminal program. I had a lot of trouble with that initially, and part of it was a crummy cable.

Sent from my SM-N960U using Tapatalk
 

·
Registered
Joined
·
152 Posts
Thanks JBMan I will have another look at my cable setup. Yeah I'm leaning towards those 400a IGBT's I can get 3 of them for the price of 1 Littlefuse (what I burned). Also the 400's are 1200v so maybe a little more resistant to possible voltage spikes.

Before you make any changes, it would be best to build the controller exactly as Paul did in his instructible. It works very well. If you're struggling with the serial, you may need a different cable or to change the bit rate and format in your terminal program. I had a lot of trouble with that initially, and part of it was a crummy cable.

Sent from my SM-N960U using Tapatalk
 

·
Registered
Joined
·
117 Posts
Hi arber,

Have you had any experience with Paul's AC board lately
 

·
Registered
Joined
·
1,478 Posts
The idea with only using 1 device right now instead of 3, is that it is simpler to work the bugs out with less chance of blowing 3 devices. Once you get 1 channel working, then you will know how to set the second and third channels.

If anything, set the current limit to 50 Amps and see if you can get the motor to run, and monitor the voltage, current, temperature readings, current sharing, etc. to make sure that it all looks okay. Then increase the limit to 75A and repeat, etc. work your way up to higher power levels--don't try to stomp it with 330 Amp limits on a 300A device. That limit was obviously of no use with your 300A IGBT.

Every piece of wire or buss bar has an associated resistance and inductance and even capacitance depending upon neighboring wires. All those factors become important when dealing with high frequency switching of high power devices. Every build is different and requires tuning and adjustment to work--it's not a one-size-fits-all situation.


Pauls' explanation doesn't make any sense: The purpose of the controller is to regulate and control the voltage and current such that damaging "dI/dt" events do not occur no matter what inductance (motor plus wires) you may have. You should start with a slow ramp-up of current, i.e. setting the "dt" component to be a longer period of time.

Google is your friend for looking up unknown terms or acronyms.

One approach might be to get some help. Once you get it assembled maybe you can find a local EE-type with an oscilloscope to take a look at the waveforms and verify that it's working okay before hitting the highway. You don't want a failure as you pull out making a left turn across heavy traffic, etc...
 

·
Registered
Joined
·
152 Posts
Thanks for your advice Kenny. I am definitely getting help from this Forum! And I really appreciate it. I like your idea of reducing the current limit. Once I get the Serial working I will do that. Finding someone with an oscilliscope in my neighborhood? haha... maybe - edit: well actually I may know someone! Good idea. I would love to have one myself - but all the cheap little ones seem to have low voltage limits.... too low to see possible spikes etc. One day.

The idea with only using 1 device right now instead of 3, is that it is simpler to work the bugs out with less chance of blowing 3 devices. Once you get 1 channel working, then you will know how to set the second and third channels.

If anything, set the current limit to 50 Amps and see if you can get the motor to run, and monitor the voltage, current, temperature readings, current sharing, etc. to make sure that it all looks okay. Then increase the limit to 75A and repeat, etc. work your way up to higher power levels--don't try to stomp it with 330 Amp limits on a 300A device. That limit was obviously of no use with your 300A IGBT.

Every piece of wire or buss bar has an associated resistance and inductance and even capacitance depending upon neighboring wires. All those factors become important when dealing with high frequency switching of high power devices. Every build is different and requires tuning and adjustment to work--it's not a one-size-fits-all situation.

Google is your friend for looking up unknown terms or acronyms.

One approach might be to get some help. Once you get it assembled maybe you can find a local EE-type with an oscilloscope to take a look at the waveforms and verify that it's working okay before hitting the highway. You don't want a failure as you pull out making a left turn across heavy traffic, etc...
 

·
Registered
Joined
·
952 Posts
Arber, I can barely keep up with your language. I'm working with a P&S DC Control Board that was built and populated by Paul. So I don't really know what DIP version is, or what BOM is, or SMD or dsPic. I know these are important terms but I don't really understand them. And I don't have a PICT.

I'd like the boards to work as they are. I assume Paul put the latest software on the board before he sent them to me (I got two at the same time) - about a year ago.

I have not even been able to successfully connect with Serial! And I have tried - really tried.

Paul indicated that he set the maximum current to 330 for each IGBT (if I remember correctly - somewhere back in my email thread). When I explained what happened to Paul, he suggested that the problem was a voltage spike, due to this formula :

V = di/dt * L

He said this:

"Under high current, you have a large di/dt (change in current over time). That can't be helped. The layout caused L to be too big (the stray inductance). L is related to the loop area of the current path. So,you had a large voltage spikes, which caused the insulation to break down inside the igbt, which took some of the driver sections with it on the board."

And when I look back at my layout, I can see what he means... because I had different length cables to different IGBT's both going in and out - my big mistake - and the IGBT that burned out was the one on the shortest path. I think perhaps it was not burned by overcurrent, but by a voltage spike going over 600v (the rating of the IGBT) and finding it's way up into the board taking the Gate driver with it.

ANYWAYS, what I hear you suggest is to try and make it work with 1 IGBT. Connecting all the current sensors wires together??? But if I do this, then I only have 300amps of potential @70degC since that's what I have left - two 300a IGBT's.

Yeah I want to make it work correctly. I understand your idea that it makes sense to have one DC motor driven by 1 big IGBT - but I don't have 1 big IGBT. Also I'm not so sure about hacking Paul's board until I really understand it better. For example, if the three current sensors are put all in parallel all to the same point, would than not confuse the board? Paul indicated that they need to be connected properly or the board see's a hardware fault.

ONE possibility I can see is to put all three current sensors on one single output from one IGBT - lining them all up in a row so they all sense the output of a single IGBT, but they are feeding back to the three sensor inputs on the board. Leaving the other IGBT's disconnected (either removed completely, or still plugged into the board but with no big cables connected). I sill wonder though if the Gates are sensing anything through their connection to the IGBT and if two were missing, would the Gate's feed back a HW fault to the brain of the board?

Thanks for your thoughts Arber - and also for the feedback on that Ebay listing.

-Ocean
Hah, so glossary it is then

I say DIP when i mean through hole version of board
SMD is surface mount version what i think you have got now.
BOM is bill of materials
PICKit2 or 3 is Microchip interface that you need to upload new software to the brain of the board.

What i mean is to use one IGBT to run your motor and also use one current sensor to provide feedback. Rest of them would be disconnected or at least not used.
You would have to ask Paul abaout paralelling sensors, but i think that idea would cause current reduction because of paralel feedback. I think you would get like 1/3 of the current.

If you use only one IGBT you can fool desat detection on other (unconnected) drivers by connecting C1 + C2+ C3 in paralel and E1+E2+E3 also and then connect one single wire to DClink +HV and one wire to -HV. I guess ill IGBTs have the same potential when connected to DClink.

I have 3phase version of Pauls controller so at least two current sensors have to be connected for the logic to work. I dont know how it is for DC version.
Last time i tried to run both Leaf motor and my ACIM motor was in 2017. I got good control up to 100Vdc on both. Upwards from that i got terrible EMI that prevented Leaf motor from running with ABI encoder. I think i used too small capacitors on top of everything. I since have DClink taken apart and put that board aside to play with Volt/Ampera inverter. I have direct access to gates from there so i can just connect control board and drive it.

Did software change in any way from 2017?

tnx

A
 

·
Registered
Joined
·
1,478 Posts
The PWM (pulse width modulation) gate driver signals are low voltage and could be monitored by almost any oscilloscope, as are the current feedback signals from the current sensor(s). Those are the signals of most interest--one is the command and the other is the response.

If you wanted to look at the high voltage for spikes for some reason, then a voltage divider circuit could be used to step the HV down to a level that the scope could handle. But there will be lots of noise and spikes in the HV due to the high frequency switching of the PWM. The current is more important to measure than the voltage--the controller doesn't care about voltage and will do whatever it takes to the voltage to make the current respond according to the command.
 

·
Registered
Joined
·
152 Posts
Thanks for glossary lesson Arber ;~)

Yes I have the surface mount version of the board. Actually it would be really great to have the Bill of Materials for that. I would like to pursue fixing my broken board eventually. I don't have a Pickit, but I don't think I really want to go there. Paul's board comes with the program, so I think just getting to Serial communications to work will be enough for me.

So I finally ordered THREE Semikrons for $40 each which seems like a great deal. They are 400 amp @ 25degC (max rating) so actually they are basically the same as the Littlefuse's that I have.

So I think I will stick with using all three IGBT's as the board is meant to. BUT I will be much more careful with my implementation.

Duncan suggested using long (1 meter) output cables to the motor (he attributed this idea to Paul as a brilliant way of helping to keep the current balanced between the three IGBT's) - it makes sense intuitively, because even if I make every effort (which I will) to keep the current path equal through the IGBT's, having long output cables spreads any resistance-difference (between the IGBT's) into a smaller percentage over the length of the cable. There may be even more to than that but it's pretty nebulous. But I like it.

Also, I will do as (I think Kenny suggested) to lower the Current Limits to like 50 amps to start out. I don't have a scope yet (thanks Kenny for the suggestion that I could use a cheap one to monitor the current sensors and PWM). BUT I do have a nice DC Clamp Meter. I can use this to monitor (and "hold") the highest current reading.

So my plan is to use the Clamp Meter on each output cable successively... with a 50 amp limit, I reach my top speed. and see what was the max current on each IGBT output. If they are different from each other, I need to make adjustments. If they are the same, I can be more aggressive on the pedal, and see again if there are differences. If no difference is detected, I can raise the current limit and do the test again.

But all this will happen after I rebuild the controller with much better / more balanced connections.

Paul is helping too - he has been really kind and patient with me. I really don't know if the software has changed since 2017.

Here's one question I've been wondering about. So on the IGBT's, Post 1 is the motor output. Post 2 has the Battery & Cap NEG. Post 3 (nearest the Board) has the Battery POS & Cap POS. So, if I want to put a "snubber cap" to help with voltage spike protection - would it go on Post 1&2 (parallel with main caps) or between Post 2&3 - or between Post 1&3?

The Snubber Cap that I have is designed to fit between two adjacent posts... but I'm not sure how it should be used in this application. Intuition tells me to put it between Post 1&2 (Motor Output & Cap NEG)

Thanks everyone!
-O


Hah, so glossary it is then

I say DIP when i mean through hole version of board
SMD is surface mount version what i think you have got now.
BOM is bill of materials
PICKit2 or 3 is Microchip interface that you need to upload new software to the brain of the board.

What i mean is to use one IGBT to run your motor and also use one current sensor to provide feedback. Rest of them would be disconnected or at least not used.
You would have to ask Paul abaout paralelling sensors, but i think that idea would cause current reduction because of paralel feedback. I think you would get like 1/3 of the current.

If you use only one IGBT you can fool desat detection on other (unconnected) drivers by connecting C1 + C2+ C3 in paralel and E1+E2+E3 also and then connect one single wire to DClink +HV and one wire to -HV. I guess ill IGBTs have the same potential when connected to DClink.

I have 3phase version of Pauls controller so at least two current sensors have to be connected for the logic to work. I dont know how it is for DC version.
Last time i tried to run both Leaf motor and my ACIM motor was in 2017. I got good control up to 100Vdc on both. Upwards from that i got terrible EMI that prevented Leaf motor from running with ABI encoder. I think i used too small capacitors on top of everything. I since have DClink taken apart and put that board aside to play with Volt/Ampera inverter. I have direct access to gates from there so i can just connect control board and drive it.

Did software change in any way from 2017?

tnx

A
 

·
Registered
Joined
·
152 Posts
Intuition tells me to put it between Post 1&2 (Motor Output & Cap NEG)
-O
OK not just intuition. I think of the motor on the Output Terminal Post #1 as feeding back voltage spikes to the IGBT's through this path. So if one leg of the Snubber Cap is connected here it would absorb that vibration. The other leg has to be the opposite right? So since the Motor Output is POS to the motor, the other foot of the Snubber Cap would go on the Bat NEG which is also the Cap NEG and also Post #2. Meanwhile, the Main Cap is between Post #3 (Bat POS input) and Post#2 / Main Cap NEG / Bat NEG.

But it's just theory and I'd like to hear other people here's theory on where snubber caps should be placed in the system. Should they parallel the Main Caps? Or between #3 POS Input and #1 Motor POS Output?

Just wondering...
-O
 

·
Registered
Joined
·
1,478 Posts
What is the capacitance value and voltage and RMS current rating of the snubber caps that you have?

In a dual transistor IGBT the snubber cap is normally connected in parallel to the big electrolytic caps across the POS and NEG posts. (posts #2 and 3)

If the cap were placed as you intuit between the Output to the Motor and NEG, then the transistor would be switching into a capacitance load, which initially appears as a short circuit to NEG and would cause the IGBT to pop and blow.
 

·
Registered
Joined
·
952 Posts
What is the capacitance value and voltage and RMS current rating of the snubber caps that you have?

In a dual transistor IGBT the snubber cap is normally connected in parallel to the big electrolytic caps across the POS and NEG posts.

If the cap were placed as you intuit between the Output to the Motor and NEG, then the transistor would be switching into a capacitance load, which appears as a short circuit and would cause the IGBT to pop and blow.

Since there are 3x IGBTs connected in paralel, i dont see a reason why not connect snubber cap over E2 and C2E1 terminals. If you have the space that is...

Normally ring cap has enough reaction to serve also as a snubber, so DC link should be covered. On the phase side you may have large spikes due to different phase lines.



Here i found some reference to the paralel IGBT design
https://library.e.abb.com/public/64...leling of IGBT modules_5SYA 2098_25082013.pdf
 

·
Registered
Joined
·
152 Posts
JBMan yes I've been studying the 200kw instructable. It's pretty amazing although I don't have the big ring Cap. But I see how the sheet copper plays an important role in spreading the current load evenly over the IGBT's - so I will be using sheet copper in the re-build.

The Snubber Caps I have (4 of them) are ASC x329s 8MFD 600vDC (datasheet PDF attached)

Kennybobby - so I hear you say if the Snubber is put between Post #1 (Motor Output) and Post #2 (Neg) it would be experienced momentarily as a short to the Batter Neg? But since I'm running a DC motor, the Motor Output is a short to the Battery Neg through the motor anyways... so ? Do you mean just from charging up the Snubber Caps when the energy is let through the IGBT's? This event would cause the IGBT to blow? I find it hard to believe that the snubber caps could absorb enough amps to blow a 400 amp IGBT even if I stomped on the pedal - BUT I may be totally wrong about this given the fact I already blew one!

I attached two PDF's - one is a brief about snubber's in general - which I try to understand but is a bit nebulous to me, and the other is the datasheet for my particular ASC snubbers.

Maybe they won't be necessary, but if they would help I want to use them which is why I'm asking - how to use them properly.

Is it correct to say that the "Ripple" current / voltage is coming back from the motor through Post #1 Motor Output? If this is so, would it not make sense to put one leg of the snubber on Post #1? Maybe the other leg should be on Post #3 the POS Input? Hmmm.... this would put one snubber leg on BAT POS (after contactor closes) and the other on Motor Output which is actually a short through the motor to Bat Neg.... and This is actually what I initially had in mind - to put the snubbers between Post#1 and #3. But I'm not sure if it's the right thing...

The IGBT_Snubbers PDF has a diagram showing this as one of the possible installations - although their IGBT diagram is not quite the same as what's printed on my IGBT's.

Arber - that PDF of paralleling IGBT's is intense. Over my head but I do begin to get an understanding of the subtle variations that have major impacts - like even heat dissipation of the IGBT's, and the variation of impedence between the paths of current through each IGBT. I think this is why I blew mine in the first place, and also why Duncan suggested long output cables.

So Arber, I here you are saying you think putting a snubber on each of the IGBT output terminals, Post #1 (C2E1 Output) and Post #2 (E2 Bat Neg) would help absorb ripple coming back from the motor through the different long cables...? Again I'm running a DC motor so the phases are the same, but the small variation in impedance of the current path's will exist from the three IGBT outputs to the motor - although hopefully minimized by the long cables.

In this configuration it seems to me like both legs of the snubber start with no potential difference - since at Post #2 there is a direct connection to Bat NEG and at Post #1 is a short to Bat NEG through the motor. So the Snubber would not see any voltage difference until the IGBT let's through current - and even then the voltage difference would have to be what's left between the POST #1 and the impedence of the motor - as if the motor was a resistance - which it is. So I would expect to see some voltage there but not the whole pack voltage ... until ripple current/ voltage came back from the motor... I'm fairly new at this but my brain is trying to work here...

So Kennybobby do you still think there's a possibility of blowing the IGBT with a snubber cap connected to Post #1 (Motor Output) and Post #2 (Bat Neg)? or has the situation changed from how you were thinking about it?

Really just trying to get clear here. But the conversation is really exercising my brain so that's good!
 

Attachments

·
Registered
Joined
·
952 Posts
It's pretty amazing although I don't have the big ring Cap.

The Snubber Caps I have (4 of them) are ASC x329s 8MFD 600vDC (datasheet PDF attached)
Argh! It just dawned on me You are using DClink elcaps without snubbers! Thought you have powering film cap. Well of course you broke IGBT. It was not amps that killed it. It was sudden dV/dt. Since there wasnt anything absorbing transients from sharp switching.

Just put 3x 1.5uF or 2uF snubber caps each on its IGBT. Contacts to use are C1 - E2. If you have spare snubbers and space in controller you can put snubbers to E1/C2 and E2 which can help smooth output on that phase. It is not neccessary though. First snubber is essential though!

A
 

·
Administrator
Joined
·
6,157 Posts
Hi Ocean
I suspect that you are about the same level of electronic knowledge as I am

AND I would not dare to deviate from Paul's drawings and BOM!!

I recommend getting all the same parts as Paul used including that big round capacitor and building it exactly as he shows

I certainly do not have the knowledge and understanding to vary anything - for me it is "Monkey See Monkey Do"

That big round capacitor looks awesome!
 

·
Registered
Joined
·
117 Posts
Hi Ocean
I suspect that you are about the same level of electronic knowledge as I am

AND I would not dare to deviate from Paul's drawings and BOM!!

I recommend getting all the same parts as Paul used including that big round capacitor and building it exactly as he shows

I certainly do not have the knowledge and understanding to vary anything - for me it is "Monkey See Monkey Do"

That big round capacitor looks awesome!
I am currently in the same boat, with the level knowledge side and am trying to build the power stage of Pauls prebuilt AC controller board,

It seems that this Round Cap is very limited availability and only available in the US ,which is strange as the manufacturer is here in the UK, I ask Paul if he could get one and send it to me, he sent me pictures of a alternative readily Capacitor, setup (I can up load Pics if anyone want to see them)

like you say I darn't deviate from the picture

I am still confused as when snubber caps are used, i dont see any on Pauls AC board
 
101 - 120 of 195 Posts
Top