Audi SQ7 4.0 TDI - electric supercharger.

I

ipepe

New member
I'm very interested in Audi's electric supercharger mounted in Audi SQ7 with 4.0 TDI engine. They are quite cheap for used one and can be bought for around 600-1400 PLN ($150-$350) in Poland or about double that new one. It's produced by Valeo and it runs on 48V system, I think that part number is 057145601F.

Audi promo video:
Engineering explained video about audi's system:
Disassembly video (in polish):

I don't think it's as powerful as Alex's electric compressor but it's quite compact solution with builtin controller and water cooling. If worked out it could probably be used with much success on smaller displacement engines up to 2.0 liters petrol engines.

The only downside at the moment is that I don't know anybody that could figure out how to make it work outside Audi SQ7. Control might be running on encrypted CAN line or even something weirder. I think as best case scenario would be to take out original PCB and rig in some other controller. The problem is that construction is based on reluctance motor for which controllers are not cheap and readily available.

Out of other interesting information is that there was recall for software update because in some cases compressor could pull too much current and burn out: https://www.motor-talk.de/forum/softwareupdate-sq7-t6581274.html

Some (might be true or not) information that could be found on internet:
- spins at up to 70,000 rpm to supply boost so quickly that the lag endemic to turbochargers is largely negated
- a 48-volt system provides the 10 horsepower needed to run the compressor
- The EPC’s response time is less than 250 milliseconds, peak output is 7 kW and its maximum speed is 70,000 rpm. The electric turbocharger’s boost function extends all the way to an engine speed of 1,650 rpm.
- 7kW / 48V ~= 150A

On some sell offers there is information that product was checked before posting. I think there might be some hardware/softwate to test this compressor on bench. If theres something like that, then I'm interesting. Maybe it could be used as "controller of controller" or maybe there is a chance to reverse engineer transmission protocol to control board so that compressor turns on.
 
interesting ... but with compact comes tiny input/output diameters ... this explains the "to an engine speed of 1,650 rpm".

I would say it is easier to build on Alex's prototype, add eMotor cooling and a small (arduino driven?) high power controller which taps into OBDII with software WE write :)
 
That is a REALLY interesting idea. Remember - nothing stops you from running 2 of them. Or 3. Or 4. Or 10...

The price is certainly right.
 
Lol. BTW, ipepe - I can understand a fair amount of that last video. My parents both came from western Ukraine (which at various times in history was Polish) - even though I was born in the states, I learned to speak Ukrainian before I learned to speak English. My grandparents used a lot of Polish words and colloquilisms. Thanks for the links!
 
AlexLTDLX said:
That is a REALLY interesting idea. Remember - nothing stops you from running 2 of them. Or 3. Or 4. Or 10...

The price is certainly right.
Click to expand...
Problem with connecting multiple is problem with how to do it optimally. Do You connect them in parallel or do You connect them in series? Price currently is quite good but I think it will go up once people know how to control them and everyone will want one in their car.

AlexLTDLX said:
Lol. BTW, ipepe - I can understand a fair amount of that last video. My parents both came from western Ukraine (which at various times in history was Polish) - even though I was born in the states, I learned to speak Ukrainian before I learned to speak English. My grandparents used a lot of Polish words and colloquilisms. Thanks for the links!
Click to expand...
I'm glad to hear that You understood it! You speak English so well that I didn't think You understand some Eastern Europe language ;)
 
Last edited:
I think you'd need to connect them in parallel; flow volume is limited by compressor size. Putting them in series is good for raising boost pressure, but it won't give you more flow necessarily. In other words, volume is volume. It doesn't matter how compressed it is at the time. And I tried to get a good explanation of the relationship between two compressors say in a "standard" twin turbo setup (i.e. in series) from my friend Bill Devine (he's an owner at Bullseye Turbo and also a legendary racer - he has a ~3,5000 hp twin turbo mustang he races - that's why he's "Wild Bill" Devine.) Anyway, I never really fully wrapped my head around his explanation, but I think I got the gist of it. If your boost pressure stays the same, in theory with two parallel compressors you should end up with twice as much flow, but in practice it works out to be a bit less. So two 600hp turbos @ 5 psi don't give you1,200 hp at 5 psi; more like 1,000 hp - but I could be wrong; this is another area where we'll get better data once we start seeing more electric turbos in action.
 
Back to topic:
I have read some comments under disassebmly video but I also found one more polish Youtuber that tries to make this supercharger work. First one claims in comment that they managed to get it running but only in on/off mode (for me this is already win because I don't expect it to work other way). He doesn't got into detail how but that's what he claims. Other one claims that they managed to replace control board with their own one but only managed to reach 20'000 rpm.

Also I might know reason why they are so cheap when sold as used ones. Apparently Audi tried to fix some software issue on those and they recalled them for replacement with brand new units? There was also one person that claimed to scrap few of them because he thought that if they are being recalled, they might be worthless.

EDIT: also on second youtubers video they mention that there is 32-bit STM controller on the board with exposed programming pins, but after half a year they replaced control board so I assume they didn't manage to program it or get enough information to create a new software from scratch dedicated for that PCB
 
Frankly, if it's just a brushless motor, I'd try just using an external RC type ESC. If I had just a *bit* more time, I'd order one myself and see if I could get it working. But to be honest, I haven't looked at them in any detail at all. I'm desperately in need of cleaning out my garage and I have to build my new battery packs first.
 
ipepe said:
Back to topic:
I have read some comments under disassebmly video but I also found one more polish Youtuber that tries to make this supercharger work. First one claims in comment that they managed to get it running but only in on/off mode (for me this is already win because I don't expect it to work other way). He doesn't got into detail how but that's what he claims. Other one claims that they managed to replace control board with their own one but only managed to reach 20'000 rpm.

Also I might know reason why they are so cheap when sold as used ones. Apparently Audi tried to fix some software issue on those and they recalled them for replacement with brand new units? There was also one person that claimed to scrap few of them because he thought that if they are being recalled, they might be worthless.

EDIT: also on second youtubers video they mention that there is 32-bit STM controller on the board with exposed programming pins, but after half a year they replaced control board so I assume they didn't manage to program it or get enough information to create a new software from scratch dedicated for that PCB
Click to expand...
So, do we know the turbo is using a CAN, and are they bypassing it?

I mainly deal with pulsed RF, so this is some new stuff for my head to wrap around. I just can't see how a turbo they claim spins up to 70k rpms, but only used to 1650rpms? Meaning how much control are you going to need at under 1650 RPMS. Like you ipepe, I am looking to just go on/off, and control with a Hobb. I went to school for this stuff back in the 80s, so I have a bit of a learning curve.....

When I got my job working the WSR88d, my boss told me the worst thing they did was put a computer on it, I tend to think the same way on some electronics. Yes, I am OLD.
 
SkylineUSA said:
So, do we know the turbo is using a CAN, and are they bypassing it?

I mainly deal with pulsed RF, so this is some new stuff for my head to wrap around. I just can't see how a turbo they claim spins up to 70k rpms, but only used to 1650rpms? Meaning how much control are you going to need at under 1650 RPMS. Like you ipepe, I am looking to just go on/off, and control with a Hobb. I went to school for this stuff back in the 80s, so I have a bit of a learning curve.....

When I got my job working the WSR88d, my boss told me the worst thing they did was put a computer on it, I tend to think the same way on some electronics. Yes, I am OLD.
Click to expand...
Diesel ... and the eCharger is only used (in the original application) from idle up to 1650 to hide the turbo lag ...
 
re CAN ... (the device has a 4 pin connector). It meamwhile is very easily possible to reverse engineer the messages with a small can bus sniffer (arduino ... again). But that would need the full vehicle and some tapping into a working setup!
 
AlexLTDLX said:
Frankly, if it's just a brushless motor, I'd try just using an external RC type ESC. If I had just a *bit* more time, I'd order one myself and see if I could get it working. But to be honest, I haven't looked at them in any detail at all. I'm desperately in need of cleaning out my garage and I have to build my new battery packs first.
Click to expand...
On the disassembly video they say that propeller axis isn't magnetized. That is why I assume that BLDC controller will not work in this case. I think this construction is called reluctance motor and controllers for those are quite expensive and not so readily available.
 
I did notice that "screwdriver" test while I skipped through the video. Didn't even put two and two together - but of course, you're right. Some Teslas use induction motors, btw. I don't even know where you'd start looking for a driver for that motor that could handle 70,000 rpm. I wonder why they went with that motor (they're less efficient) than a pm brushless motor? I'm sure there's a reason - it might be because it's cheaper to manufacture, or maybe they were worried about heat (neodymium magnets lose their magnetism when they get hot), or maybe there's some performance advantage we're not aware of.

Here's an interesting article on the differences and relative efficiencies:
www.machinedesign.com

Motors for efficiency: Permanent-magnet, reluctance, and induction motors compared

• It is possible to find superefficient high-horsepower motors but, at this point, many of these are exotic designs that are practical for limited applications. • Special motor...
www.machinedesign.com www.machinedesign.com
 
When I originally was planning on using this supercharger I was thinking about converting it to BLDC construction. For price of less than $200 You could basically scrap everything except case and compressor wheel and put TP Power motor inside with BLDC controller
 
I finally looked into these a bit more and they're expensive over here - well over a grand used. If you're gonna go through all that, it seems like you'd be money ahead with that Speedmaster unit. I have a feeling in the next year or two there's going to be more of these things (not just Audi) in OEM applications that we can use.
 
Lol - that is badass. I have not seen it already, but I'm smiling as I type this. Awesome! I don't speak German, but I'd love to know how he got the computer to work, and is that the actual battery that Audi uses?
 
Toli0002 said:
Here is an interesting video where they use the Audi SQ7 electric turbo on a smaller engine. :)
(Maybe you have seen it already...)

O yes the language is German:)
Click to expand...
highly customized 2.0 TURBO (!) 4 banger ... where the Audi eblower was ADDED for lower rpm boost ( not running exclusively )!

Did not go through the whole clip ... curious if he has explained the mystical eblower interface and control ... (will check later and report).
 
You must log in or register to reply here.
Top