[dkubus]

I'm dreaming up (inspired by this forum) a 2,500A Continuos -3,500A Peak capable DC controller based on multiple IGBT modules bridged together with refrigerant chilled water cooling system preferably 400V battery input so that Sag is less issues toward the end of a 65 second steep Hillclimb run. I'd love some input into what might be possible using the "Paul&Sabrina" control board? Cheers guys for any help or advice you might have for this optimistic project.
Cheers Mike.

 

[Hollie Maea]

I think Evwest is still trying to find someone to buy their Shiva. That would probably be the easiest way (though finding $11,000 might not be easy). Thus said, what brushed motor could do 3000A for 65 seconds?

 

[dkubus]

Agreed that's the challenging part however the use of multiple motors opens up loads of possibilities regarding wiring configuration and therefore voltage/amperage handeling abilities. Plus the ability of the controller is there to have "headroom" from over heating and catastrophic failure mode at say 1,500-2,000 Amp "continuos" abusive racing use. The idea isn't to run at 3,000+ Amps for more than a second or two of course and "spread over multiple motors" all equipped with super cooling ability for short bursts. I currently have (in my road car) a Soliton1 and I do really like it! But I'm not keen on spending thousands of dollars on a big bad-ass controller from Evnetics as they are no longer able to support it and assist the development or repair if we ever damage it. I'd rather do the hard work and be able to "back ourself" so that we can add the features we need/want... plus then it's possible we can assist others have amazing power too?...

I think Evwest is still trying to find someone to buy their Shiva. That would probably be the easiest way (though finding $11,000 might not be easy). Thus said, what brushed motor could do 3000A for 65 seconds?

 

[Duncan]

I'm using a P&S IGBT controller - I think it's good for 1400 amps - but I'n limiting it to 1200 amps

This uses three IGBTs - I don't see any reason why you could not use six IGBTs - or bigger ones

I'm a mechanical engineer (so I could be wrong) - but the genius level trick that Paul used was that each IGBT has a meter of cable before the three connect to each other - this means that the usual problem of the three IGBTs have slightly different closed resistance and the current then going through one of them and not the other two is eliminated

The IGBTs bolt to a thick aluminium plate - on mine I drilled it and pass water through it to keep it cool

 

[dkubus]

Hi Duncan, thanks for joining in here. I was hoping you would. Exactly what I was thinking might be the best way forward. The one meter cables are of what size and do they pose any heating problems? Efficiency is of course important too but my main questions regarding the P&S board is what's the control interface options for it like? If I ran 6 igbts then should I run two P&S boards? Or twin up three pairs of igbts? 2,400A would be great!!

Thanks again for joining in. 🙏

I'm using a P&S IGBT controller - I think it's good for 1400 amps - but I'n limiting it to 1200 amps

This uses three IGBTs - I don't see any reason why you could not use six IGBTs - or bigger ones

I'm a mechanical engineer (so I could be wrong) - but the genius level trick that Paul used was that each IGBT has a meter of cable before the three connect to each other - this means that the usual problem of the three IGBTs have slightly different closed resistance and the current then going through one of them and not the other two is eliminated

The IGBTs bolt to a thick aluminium plate - on mine I drilled it and pass water through it to keep it cool

 

[Duncan]

This is Paul's website

https://www.pandspowerelectronics.com/shop

I'm sure that he will be able to help you

 

[dgilbert]

I used the P&S control board with another IGBT driver (2 actually) to drive 6 x400A IGBT. Current limit is set at 1200A. I've been running 170V but will be 220V or so next year.
And built another very similar with 4 IGBTs set for 800A.

 

[dkubus]

Hey Dgilbert, they sound like great setup's you've put together! And they should have loads of safety margin too. I'm also considering attempting to try and stay under 50% of each IGBT's rated abilities (if I can get hold of enough meaty units to use)...?

Would love to see some pics of your setup if your willing to share? How configurable is the control boards firmware? Is there plenty of options for settings of battery/motor voltage & current limits etc?

Cheers for joining the conversation, really gives me hope that my goal might not be as crazy as I first thought. 🤘🤘👍👍

Merry MegaWatt Christmas 🎁😜

I used the P&S control board with another IGBT driver (2 actually) to drive 6 x400A IGBT. Current limit is set at 1200A. I've been running 170V but will be 220V or so next year.
And built another very similar with 4 IGBTs set for 800A.

 

[brian_]

I'm a mechanical engineer (so I could be wrong) - but the genius level trick that Paul used was that each IGBT has a meter of cable before the three connect to each other - this means that the usual problem of the three IGBTs have slightly different closed resistance and the current then going through one of them and not the other two is eliminated

While this wouldn't eliminate the potential imbalance situation, it would reduce the severity (perhaps to the point of eliminating the problem) by making the IGBT closed resistance a less significant part of the resistance of each output current path - slick idea.

The one meter cables are of what size and do they pose any heating problems?

I would hope that this configuration would not increase the overall output wiring resistance, but that's only an educated guess. Ideally (for balance) the individual (one per IGBT) output cables would be matched in length, and all run all the way to the motor terminal or a point very close to it, with a total cross-sectional area comparable to what would be used as a single conductor.

 

[brian_]

Agreed that's the challenging part however the use of multiple motors opens up loads of possibilities regarding wiring configuration and therefore voltage/amperage handeling abilities.

The obvious possibility is to use one controller per motor. That allows you to scale up the power by increasing the number of controllers (from one to two), instead of by increasing the capacity of a single controller to a level which is difficult to build and to find. It does have the inherent cost disadvantage of duplicating the logic portion. It also eliminates the possibility of switching the motors from in series with each other at low speed to in parallel with each other at high speed... but is that a practical option anyway?

 

[dkubus]

Yes it would be good to put a single controller per motor however the application I have in mind means the advantage of twin motors with various contactor configurations will be of more benefit for the initial launch torque, and minimizing the voltage stress on the brushes too.

Mike.

[/QUOTE]
The obvious possibility is to use one controller per motor. That allows you to scale up the power by increasing the number of controllers (from one to two), instead of by increasing the capacity of a single controller to a level which is difficult to build and to find. It does have the inherent cost disadvantage of duplicating the logic portion. It also eliminates the possibility of switching the motors from in series with each other at low speed to in parallel with each other at high speed... but is that a practical option anyway?[/QUOTE]

 

[Duncan]

Hi

I am very unconvinced about the advantages of being able to change parallel/series - with a modern high voltage controller I think the advantage evaporates

Two (or more) controllers strikes me as a very sensible idea

The Logic board is cheap - the expensive parts are the IGBTs and the big capacitor

So you would need the expensive bits anyway!!

I'm thinking about two of the P&S controller I'm using - one for each motor

You can see the "brain" card there at the front - there is also a 24v supply card that mounts above that
Then it's just the three IGBTs and the huge round capacitor - and the current sensors

You can see the long output leads I mentioned

https://ecomodder.com/forum/showthr...cheap-diy-144v-motor-controller-6404-689.html

This link takes you to the conversation I had with Paul when he sent me the controller - it arrived a bit bashed up


1400 amps is probably all you want to feed each motor anyway!

 

[dkubus]

Any ideas if these or similar might be the way to go ??

https://rover.ebay.com/rover/0/0/0?mpre=https://www.ebay.com.au/ulk/itm/192701402023

3600A 😜 1200 volt max!

I used the P&S control board with another IGBT driver (2 actually) to drive 6 x400A IGBT. Current limit is set at 1200A. I've been running 170V but will be 220V or so next year.
And built another very similar with 4 IGBTs set for 800A.

 

[dkubus]

That's a sweet setup, and I'm thinking you may well be correct regarding the value of contactor switching etc.. unfortunately my second WapP9 is stuck in transit (merry Christmas).. so I can't test the "torque per Amp" experiments I have in mind prior to deciding on a final configuration.

I was recently given a box of 20 x 1000V DC link caps that I was thinking should all team up in parallel real well into one mother of a DC link Cap. (20x 100uF)...

Very experimental of course but this is a race car project after all so experimenting is the fun bit. 🤘

Hi

I am very unconvinced about the advantages of being able to change parallel/series - with a modern high voltage controller I think the advantage evaporates

Two (or more) controllers strikes me as a very sensible idea

The Logic board is cheap - the expensive parts are the IGBTs and the big capacitor

So you would need the expensive bits anyway!!

I'm thinking about two of the P&S controller I'm using - one for each motor

You can see the "brain" card there at the front - there is also a 24v supply card that mounts above that
Then it's just the three IGBTs and the huge round capacitor - and the current sensors

You can see the long output leads I mentioned

https://ecomodder.com/forum/showthr...cheap-diy-144v-motor-controller-6404-689.html

This link takes you to the conversation I had with Paul when he sent me the controller - it arrived a bit bashed up


1400 amps is probably all you want to feed each motor anyway!

 

[kennybobby]

Any ideas if these or similar might be the way to go ??

https://rover.ebay.com/rover/0/0/0?mpre=https://www.ebay.com.au/ulk/itm/192701402023

3600A 😜 1200 volt max!

That is just a diode module, no gate or transistor.

Mouser has a 1200V, 800A single IGBT module for ~$150, FZ800R12KE3.

 

[dkubus]

Hey cheers Kennybobby, I'll check that out. I got a little excited when I saw it listed as an IGBT 🤷‍♂️🤦‍♂️

Any ideas if these or similar might be the way to go ??

https://rover.ebay.com/rover/0/0/0?mpre=https://www.ebay.com.au/ulk/itm/192701402023

3600A 😜 1200 volt max!

That is just a diode module, no gate or transistor.

Mouser has a 1200V, 800A single IGBT module for ~$150, FZ800R12KE3.

 

[dkubus]

Hi Kennybobby, how would these go?

https://au.mouser.com/ProductDetail...=sGAEpiMZZMshyDBzk1/Wi3HsWEt5/iIsxEP0C45rDQ8=

https://au.mouser.com/datasheet/2/196/Infineon-FZ1600R17KE3-DS-v02_00-en_cn-464842.pdf

Any ideas if these or similar might be the way to go ??

https://rover.ebay.com/rover/0/0/0?mpre=https://www.ebay.com.au/ulk/itm/192701402023

3600A 😜 1200 volt max!

That is just a diode module, no gate or transistor.

Mouser has a 1200V, 800A single IGBT module for ~$150, FZ800R12KE3.

 

[kennybobby]

Those are massive indeed, maybe 3 or 4 in parallel could handle the current.

It is not a trivial task to take a design and scale it up to the level you want. And this is just the controller. What motor can handle that power? And what batteries can supply that energy?

It sounds like you need a Shiva controller, designed and built by experts in motor drives.

https://www.evwest.com/catalog/product_info.php?cPath=1&products_id=148

 

[dkubus]

I originally didn't go into details regarding the motors or battery as this was a "controller topic" I'm pretty new here so please excuse my ignorance. I'm learning how to post etc (I still don't get how my posts always look different with regard to "replying or quotes"... here goes... bellow ⬇

Those are massive indeed, maybe 3 or 4 in parallel could handle the current.

I'll keep researching ideal components then but I know that these are available here in Australia so these might be a good choice.

It is not a trivial task to take a design and scale it up to the level you want.

100% agree with you on that one! This is why I started this conversation, there's much more experience and knowledge here than I currently posses on the topic and as a DIY guy I'm keen to learn (probably the hard way) but learn all the same.

And this is just the controller. What motor can handle that power?

There will be a minimum of two motors (I have room for more in this vehicle design) if required.

And what batteries can supply that energy?

Battery will be 100Ah (5Ah x 20P Lipo pouch cells) power density cells not energy dencity chemistry. More than likely 88S and more than capable of 30C continuous (till exhausted) with minimal sag. (Hillclimb racing is like drag racing but with corners and normally one mile in distance up a 10-20% grade of mountain road).

It sounds like you need a Shiva controller, designed and built by experts in motor drives.

Whilst I love the idea of the Shiva and I truly enjoy the Soliton1 I have in my daily driver unfortunately Evnetics is no more and I'm not keen on spending money on a device that will have no backup support. On the other hand if I spend double, triple or more getting a controller designed and built that works well then by the time I have been able to achieve that goal, I'll definitely be equipped to support it and keep it running and perform any future modification I ever dream up....

That's why I'm here in this diy forum. Cheers for all of the help and guidance so far. I really appreciate it.

Cheers Mike.

https://www.evwest.com/catalog/product_info.php?cPath=1&products_id=148

 

[Duncan]

Hi
The high current end is very important - Paul (of P&S) does know what he is doing
He learnt a lot from his earlier 500 amp controller

The ring capacitor on my controller has it's connections to the IGBTs in the form of copper sheet - and this is something to do with the inductance!

Which is why if I was going to do as you are I would be using his design - and his control board in the form of two controllers

The SHIVA looks awesome and is probably a level above the P&S in sophistication

But it is five times the cost and no longer supported

Hill climbs here (South Island) are gravel with big trees and big drop offs - I don't have the balls to use my car at full power in those conditions

Back in the UK "Sprints" were great fun - I used to have a twin cam mini

Here we do something similar - But I don't have the maximum speed for the fast ones - I had my mini up to about 130 mph
The fastest I have taken my device is 93 mph at the 1/8th mile drags
100 mph would probably be OK - beyond that will overspeed my 11 inch Hitachi too much