Troubleshooting WB Projects' E Turbo

AlexLTDLX

AlexLTDLX

Administrator
Staff member
Hi All,

WB Projects has been talking with me about his project through PMs. Since this is a "let's all learn together" type of forum, I asked him if I could move our conversation to a thread here.

I'll start by going be to a point in our PMs that makes sense.

WB Projects:
Good new for me 🤘 I've made a new set up with the intake and the maf sensor and now it work like it suppose to! The turbo still surge on low rpm.. the tunning will be this week 🤘

AlexLTDLX:
That's great to hear. If the turbo's surging at low rpm, then maybe your engine isn't flowing enough at air that point? If it was me, I'd start the turbo above your torque peak at first (peak torque is peak cylinder pressure, and it can be tricky to tune things there; you have more leeway at high rpm). Once you get high rpm flow/boost worked out, then work your way down, rpm-wise. With all my screwing around with positive displacement superchargers, I've learned that lesson the hard way. Several times in the form of destroyed pistons.

It would be easy to modify the arduino code to ramp in boost with rpm; you could use maybe a tach output, or maybe even tap into a crank (or cam speed sensor) - or, possibly easiest still; use a cherry hall-effect sensor running off your balancer bolt bosses or something. When I first put the Megasquirt in my car, I couldn't get the magnetic sensor in the distributor to work reliably; especially under boost. What I'm using now is a cherry sensor getting triggered by the 4 bolt bosses on the harmonic balancer - that's literally what triggers my EFI setup:

And here's the mount in the car:


WB Projects:
Programming the Arduino is an excellent idea! If I understand, the cherry sensor count the rpm and send a signal. With this signal I can tell the Arduino to be 0-100% depending of the rpm! I will check exactly how work the cherry sensor. Is it a magnet inside ?

Correct me if im wrong. I bought a bigger turbo because for the ones I search for, the efficiency was way to high in RPM for the TP motor I have. So the bigger one have my goal way down in the compressor map. Maybe this is where my surge problem is! Too much flow and not enough compress? Do you think that's possible?

I want to turn the ESC up for more boost but the compressor surge.. so I try to figure how Im I stuck at 3psi before surge

AlexLTDLX:
Yeah, the sensor is a hall effect type and detects when anything metal passes in front of it. And you have the surge concept a bit backwards - what you're describing is choke - the right hand side of the map (too much flow and not enough to compress). Surge happens when there's not enough flow, and the air is forced to go back through the compressor - that's literally what causes the surging - the air pushes back through the compressor (and loads it further) and then when enough air goes back through the compressor to relieve enough pressure from the discharge side, the compressor can then speed up until it happens again. It's a cycle that repeats and we call it "surge."

To get rid of the surge for now, just start your turbo later in the RPM band. Do you have a compressor map for your turbo? We can look at it and see exactly what's going on. Can you tell from the ESC how fast it's spinning?

I'm thinking of maybe moving some of this conversation to a public forum post - I think there's a lot of good info here, and you're the only other person that actually has a working unit. Would you mind if I did that?

WB Projects:
For sure we can discuss public! For now Im at work and the wheel of idea keep spinning in my head.. I will make a first tune, then tune the ESC with the help of MGM then after we will see wath really happen.. but for what I can feel is when im at low rpm and I activate the supercharger, I hear a surge noise and the gauge boost go up and down like crazy. When I try on the bench, put my hand in front and start the turbo, I hear the same noise and I feel no pressure on my hand. I feel some flow of air on the intake and the exhaust of the compressor housing, but no pressure. I can block it super easy from 0 power but when I start the supercharger, then try to put my and in front to block it is way more difficult..

For the compressor map is a chinese GT35 so the compressor map is a supposition from the real deal 😂
 
That actually doesn't sound like surge to me. It sounds more like the ESC is having trouble sync'ing to the motor under any load. When I was having that issue (when my motor cables were long, and battery cables were short, using a Flier ESC), it put a lot of stress on the setscrews holding my hex-drive socket on the motor shaft - so much so that the setcrew points got beaten up and the hex drive socket got loose.

FWIW, I personally wouldn't be putting my hand in anywhere near that thing when it's running. If that thing comes apart, your hand will likely be ripped apart. I've seen turbo failures first-hand, and in some cases, shrapnel made it through the sides of the volute. Certainly a lot of shrapnel goes out the discharge. Here's a illustrative pic I found online (I have no pics of my own, for once):
i-kcfzb3P-L.jpg

This might be a question best asked to Tomas Palan at MGM. Also, if I'm looking at the right compressor map, 50,000 rpm will limit you to about 4 psi at peak hp and about 230 hp. You might eventually need a faster motor. I could also be looking at the wrong compressor map.

And congratulations on having the second working DIY e turbo (that I know of) in the world!
 
When I try on the bench, put my hand in front and start the turbo, I hear the same noise and I feel no pressure on my hand. I feel some flow of air on the intake and the exhaust of the compressor housing, but no pressure. I can block it super easy from 0 power but when I start the supercharger, then try to put my and in front to block it is way more difficult..
Click to expand...

I agree with Alex

1)
It sounds like it's loosing sync to me. Blheli32 has many settings to deal with this. I don't know what MGM has.. but at the very least you should drastically increase motor timing that helps to reduce scync loss.

2)
I also agree that you probably shouldn't be boosting at low RPM which already implies that your "Lugging" your engine (screwing it).
I believe that basing max rpm/ Boost should be relative to engine RPM. Im intending on pulling the signal form the cam sensor.

3)
We go for over sized turbo's because they can achieve "more" at a lower impeller speeds. Achieving high RPM is difficult with a high powered motor because as you increase motor size power you reduce max RPM.
Surge maybe a real problem though that im worried about myself.

Also a "bigger" turbo is less restrictive when it's off.

Really using a BIG turbo on a small engine because you can spin it slower is not the right design for that. you'll notice that "Super changers" that are designed for low RPMS still have rather small inlets but larger outer diameters which is more ideal for the job.



Turbo
1635397165132.png

super charger
1635397221431.png
 
I totaly agree with loosing sync! I was telling me that yesterday. maybe is the sync and it look like it! I boost more in 4th then in 2th gear. For sure I will see with MGM in the datalog to reprogramme the ESC! You are correct alex about 230hp, that was my first goal! I will probably put a BOV because it surge after letting the throttle. I don't put "brake" in the turbo programming the ESC and I don't know if I should. Other then that, the surge at low boost is that big of a deal?
Thanks Alex 😁 you and mkgnstuff should have your name on the e-turbo for the help you gave me! 🤘
For the number of the datalog can't remember the exact number of the load, I can briefly read the graph and I can only say that my MAF sensor signal jump up ou my load too! But 1.15 of load is that make sens? 0.98 load N/A and something like 1.15 load with the turbo if I remember. Can we translate load into HP?
 
For the surge problème, I need to note that I have 3" from the throttle to the turbo! So the air stuck after the throttle close is not like when you have 5feet of tubbing, including the intercooler! So the air stuck is not huge
 
****The datalog with the Turbo is off****
The tunner take the wrong last tune so the fuelling is a lot off.. I feel something wrong between the first tune and the second.. I'll be back with the real datalog when this is going to be fixe.. so don't waste your time to read the number because is not correct
 
9.8 to 115 load makes sense since this is normally relative to EV and you always have a vacuum with NA so your always under 100%. Now you are boosting beyond atmospheric pressure (100%) so that is reflected as 115% . This means at that point your 7 points ahead of the engine with out the e-turbo .. So it works!

Do not allow motor braking that will turn the vast majority of the kinetic energy into heat (some is turned back into power and put back in the battery) this can lead to extreme motor heating. Also if you slow the motor you'll want to spin it back up again. You're far better off with a blow off valve than trying to slow the motor, which will still result in some surge.

Surging at low RPM as said before should be dealt with by not flooring it at low rpm Or electronically reducing the motors rpm below a certain rpm.

Do you have an Anti surge housing ?
 
Last edited:
Yes I have an anti surge housing. It help for sure but I just ordered a cheap BOV

I will keep the brake to "off" because all you said make sens and that's what I thought
 
You can actually translate load into horsepower. Your engine is rated at 181 horsepower. That's theoretically equivalent to a load reading of 100. Basically, you're adding 15% more horsepower. Assuming your timing and air fuel are close to where they should be, you're at 208 horsepower.

I had a good conversation with member AES on here yesterday. Turns out he's an electrical engineer. According to him, having long motor cables makes the system susceptible to emi; particularly in a car environment where there's a lot of ignition RF spraying everywhere. If you were to shield your motor cables and ground them to your battery pack for the e-turbo, that should help.

I'm staring down the barrel of a similar problem so I hope you get that sorted...
 
I thought of shielded wire! I think I will try with this since Im planning to get bigger cable anyway. Do you still want to have longer battery cable? Or you considering changing?
 
BTW, there's a member on here named AES. I spoke with him on the phone yesterday. He's an electrical engineer. He's probably a better person to go to to try to figure this out than I am. I'd probably figure it out eventually, but it would be a more expensive process that would probably involve more fire and plasma arcs.

I'll point him to this thread.
 
HHAHA I just start talking Marek from MGM about it and we will see where i'm going with this! Winter comming faster and faster 😰 so maybe it going to be only next summer that I will do some other test. I want to make it work nicely this year and improve next year!
My plan is to get maybe a bigger battery (depending if I'm able to stop the big voltage drop), maybe get better cable if needed (probably shielded and bigger awg), crank the ESC up a little bit and maaaayyyybeeee a new motor.
 
Hello, I discussed a possibility of losing the synchronization with our developers and I think that the motor cables would have to be really long for synchro loss to occur. We have tested it on much longer cables with no loss. :) I don't want to say that losing synchro is not possible though, but there would be more factors influencing it at the moment.

As I spoke with WB, it would be good if we have data from the run, we will analyze it and I am sure that we will find exact settings for you guys to make these setups work flawlessly.
 
Thanks for taking the time to join Marek. One of our members on here, AES, is an electrical engineer. He gave me some insights into this issue. He said to think of the motor cables as extensions of the windings. In a car, you have a good amount of EMI happening thanks to the car's ignition system. He advised shielding and grounding the motor cables if they're long. But the problem I've experienced is lack of sync without the car's engine running. I simply couldn't get the motor to sync at all, and William can't get his to sync with a relatively small load on it. There's an automotive professor from Weber University on youtube that takes apart EVs and hybrids: https://www.youtube.com/channel/UCtr07mdKhsUwVJjL8Kw_q5A

Virtually every one he's taken apart has the inverter by the motors with a large capacitor bank and the power cable from the batteries are much longer. Even Torqamps have short motor cables and long power cables. I can't really give you a technical reason (short of guessing) why that is. But it is consistent with our findings on here. Once I moved the ESC by the motor, things starting working for me. I don't have an issue building a housing for the ESC keeping cooling and water ingress in mind. I'm not sure how much room William has to do that.

FWIW, we may be a small (for now) group of people doings this, but we are mighty. William is a CNC machinist, AES is an electrical engineer, cmoalem is a software engineer, MkgStffAwesome is great with arduino coding, etc. I'm leaving a lot of people out, but you get the idea. I'm sure we'll get this sorted.
 
Ok guy I'm surprise with the test! This is the data of some pull at more or less battery capacity! I'm surprise because this ESC is a BEAAASSTT so check this out

As you can see I top at 450 kph hahaha im jocking look the other pictures
received_217169587060122.jpeg

this is the battery internal voltage. I mentioned some volatge drop but this is where it get interestingreceived_457233712491357.jpeg

the input voltage drop a lot. But even with thisreceived_625761588432659.jpeg

the input and peak current is verry good. The peak is a little hight for my taste but again look the other graphreceived_304426617931393.jpeg

THIS is why the ESC is a beast. I had a goal of 50k RPM and even at 55% power I am chockedreceived_266370812084851.jpeg

and this is the incomin power (W)received_1949141301933911.jpeg

So, what do you guys think? Alex, this is where I'm getting confuse. Go back and see the green graph (motor rpm) remember when I tell you in 2nd gear I had less boost compared to 4th gear? I really thought that higher gear means more load on the turbo but it look like it's not the case! The last 3 spike on the graph is 2-3-4 little pull. With the tunning of the car, the turbo surge a lot less then before at a wide range of RPM. BUT the only time that I hear fast pulse of the ESC (loocking like it getting out of sink) is when Im in second in hight RPM. When I start it low like 3k RPM it goes really well. But when Im at 4k rpm+ this is where the ESC macking fast pulse sound. *Only at 2nd gear*
So maybe, this is not what we think? For some reason the 2nd gear look harder to push air. So maybe to much current in 2nd gear?
 

Attachments

  • received_457233712491357.jpeg
    received_457233712491357.jpeg
    148.1 KB · Views: 5
You must log in or register to reply here.
Top