Enova(!) Parameters

EVElvis
Posts: 31
Joined: Thu May 07, 2020 9:17 pm
Location: Leicestershire

Re: Enova(!) Parameters

Post by EVElvis »

that's promising news Tom. Im pleased to hear as i hope the 50 pulse encoder ive ordered helps me too.
Have you tried running at 900 rpm driving? Im curious as to torsional resonance you can get in drivetrains and wonder if at 900 rpm you have hit a drivetrain natural frequency - the firing frequency of the phases matches the natural frequency of the drivetrain? I guess unless you have a clutch plate still in use there is little drivetrain damping? I did wonder if my slightly sloppy drivetrain with torsional accelerations and the original 256ppr were not ideal.
tom3141
Posts: 157
Joined: Sat Dec 22, 2018 2:29 pm
Location: North Yorkshire

Re: Enova(!) Parameters

Post by tom3141 »

I’ve not tried driving it yet with the new encoder. It was fine at 900rpm with the old encoder so I don’t think that’s the problem but I’ll definitely give it a try when I get a chance.
tom3141
Posts: 157
Joined: Sat Dec 22, 2018 2:29 pm
Location: North Yorkshire

Re: Enova(!) Parameters

Post by tom3141 »

I’ve done some more testing this morning.

No matter what I do with parameters, whether I have the wheels off the floor or driving on the road, it always starts behaving strangely at approx 900rpm.

I now have my speedo working and the rpm graph and speedo match up. I haven’t seen any spikes on the rpm graph either.

It feels almost like a rev limit as it won’t accelerate past this speed. I can actually hold it at this speed and it runs ok but there are random knocks and bangs. If I try to accelerate more it becomes worse and trips.

I’m not sure if it could be related but I noticed something strange going on with the fmax parameter. I tried setting it really low (5Hz) and found it didn’t seem to make any difference. I could still spin up to 900rpm (approx 30Hz). The other thing that’s strange is that if I try to set it to anything above 21Hz it will let me do it but I get a message saying fmax is now limited to 21Hz. See image below. Any ideas what this is about?
6378BE9A-CD16-43A0-9556-AC1983F49D55.jpeg
The good news is that it’s the first time I’ve tried driving it with the new encoder and it feels really good apart from not being able to go above 20km/h.
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Enova(!) Parameters

Post by arber333 »

tom3141 wrote: Wed May 20, 2020 9:14 pm Arber, what alterations did you make to the main board in order to use the optical sensor?
Sorry i missed that post.

Like Johannes states in his instructions i connected A and B channels with a jumper and removed R1, R2, C1, R4, R6 and C3. Basicaly i connected channel A onto channel B and remove other pullups. With newer sine firmware inverter expects to see same signal on both channels.
See schematic for gen2 board. https://openinverter.org/docs/index.htm ... -2,28.html

However i do not know if Enova board is maybe Gen1 built? In that case you need to folow the very first schematic. In latest firmware inverter autodetects hardware. What does your interface tells you about version?
tom3141
Posts: 157
Joined: Sat Dec 22, 2018 2:29 pm
Location: North Yorkshire

Re: Enova(!) Parameters

Post by tom3141 »

The Enova board was V2 I think. I had some problems when initially testing it and it was recommended to remove a resistor to revert it to v1. The interface now says it is v1.

There is only one encoder channel on the Enova board. I wondered if I’d need to change the value of any resistors or remove any but it seems to be working other than the issue at 900rpm. I’ve got a lot less pulses than the original encoder so the low pass filter shouldn’t be causing a problem.

One thing that did confuse me, was that the pinout I got with the Enova board mentions that the encoder input is for an npn open collector encoder which I don’t think the original Enova encoder was.
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Enova(!) Parameters

Post by arber333 »

tom3141 wrote: Sat May 30, 2020 1:04 pm One thing that did confuse me, was that the pinout I got with the Enova board mentions that the encoder input is for an npn open collector encoder which I don’t think the original Enova encoder was.
Well if you use open collector signal you need pullups on the main board. If not you just need the 10K/1nf filter.
You can test NPN/push pull to connect a LED to encoder output from signal to gnd and rotate motor. If It is push pull LED will blink because output is generating voltage by itself. Maybe encoder is pushing voltage into the pullup and filter cant drain it fast enough?
tom3141
Posts: 157
Joined: Sat Dec 22, 2018 2:29 pm
Location: North Yorkshire

Re: Enova(!) Parameters

Post by tom3141 »

That’s an interesting thought. The encoder I’m using now is definitely push/pull. As it seemed to work I’d assumed that the pull up resistor was not causing a problem but maybe removing that would help.

I’m also going to try updating the firmware as I noticed some of the fmax things I mentioned above have been fixed in later releases. Tried to update tonight but something went wrong and seem to have corrupted the firmware. 🙁
tom3141
Posts: 157
Joined: Sat Dec 22, 2018 2:29 pm
Location: North Yorkshire

Re: Enova(!) Parameters

Post by tom3141 »

I’ve updated the firmware this afternoon (ended up having to swap back to an old Olimex as the one I was using which I tried to update on Saturday seems to be bricked).

The firmware update made no difference. I was able to limit fmax though and gradually increase it. The problem occurs at 35Hz.

I then removed the pull-up resistor, R11, on the schematic below. This also didn’t have any effect. Are there any other resistors in this encoder circuit that it would be worth removing/bypassing/altering?
1D506DCE-CB3C-4B1D-890E-CA2CE7040E1E.jpeg
If I’m gentle with the throttle it just feels a bit ‘lumpy’ at 35Hz and will sit at that speed as if it’s a frequency limit. It won’t accelerate past this speed whatever I do. If I increase throttle input the noises get louder and there will be an over current trip.

Any suggestions are welcome!
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Enova(!) Parameters

Post by arber333 »

What is your push current rating? Maybe 1K R1 is too much and transistor is saturated?
I think i now use 1nf capacitor with 10K resistor inline connected directly to Olimex PD2 pin. True I use this with open collector only.

You can just eliminate R11 and replace R1 with 10K. Then remove T1 and R3 and connect R1 directly to C1, but replace C1 with 1nf cap.
What is the voltage output of your encoder? Is it 5Vdc? I think Olimex pins can survive 5V if protected by 10K.
tom3141
Posts: 157
Joined: Sat Dec 22, 2018 2:29 pm
Location: North Yorkshire

Re: Enova(!) Parameters

Post by tom3141 »

Thanks Arber.

I’m not sure what the push current rating is, the encoder was just a cheap eBay one so I have very few specs.

If I do what you suggest, do I leave R4 unchanged?

If this doesn’t fix it then it seems like the only other option is that there’s some kind of frequency limit on the output of the encoder but it seems pretty low to me. The encoder signal will be about 1.3kHz at the point I’m having the problem.
tom3141
Posts: 157
Joined: Sat Dec 22, 2018 2:29 pm
Location: North Yorkshire

Re: Enova(!) Parameters

Post by tom3141 »

And yes the encoder output is 5V
arber333
Posts: 3241
Joined: Mon Dec 24, 2018 1:37 pm
Location: Slovenia
Has thanked: 74 times
Been thanked: 223 times
Contact:

Re: Enova(!) Parameters

Post by arber333 »

No just connect from R1 to C1 directly without pull up or transistor. We'll see what happens.

Anyway to test inverter integrity I suggest you find another ACIM motor and try to run it with encoder.
tom3141
Posts: 157
Joined: Sat Dec 22, 2018 2:29 pm
Location: North Yorkshire

Re: Enova(!) Parameters

Post by tom3141 »

I did the changes this morning, it now has a 10k resistor and a 1nf capacitor. T1, R11, R1, R3, and R4 have been removed.

It’s still doing exactly the same thing. 🙁

I’m going to try to borrow an oscilloscope this afternoon to have a look at the encoder signal.
Ian
Posts: 3
Joined: Fri Jun 26, 2020 12:49 pm
Location: Gloucester, UK

Re: Enova(!) Parameters

Post by Ian »

I bought some cheap reflective opto sensors from ebay. The response time of the device itself was terrible, I'm pretty sure it was a fake.
User avatar
Peter
Posts: 310
Joined: Fri Dec 14, 2018 9:07 pm
Location: North West Lancs, UK
Been thanked: 8 times

Re: Enova(!) Parameters

Post by Peter »

Hi tom3141. I am running an Enova with CombiV8 board in my vehicle which runs fine. What are your parameters ? Maybe I can help with that part of the set up. I am using a HEDS 50ppr encoder module with purpose made steel disc too. Peter
tom3141
Posts: 157
Joined: Sat Dec 22, 2018 2:29 pm
Location: North Yorkshire

Re: Enova(!) Parameters

Post by tom3141 »

Thanks for the responses. Glad to hear yours is working Peter.

I haven’t updated my thread but have been discussing here - viewtopic.php?f=2&t=749&start=30

I think I’ve now fixed the problem with an Omron encoder. The eBay encoder definitely seems to have been junk. Not sure why I couldn’t get it working with the original HEDS encoder. Had it driving up to 30mph yesterday and seemed really good. I want to do a little bit more tweaking with the parameters and then I’ve just got some tidying to do before getting it on the road.
tom3141
Posts: 157
Joined: Sat Dec 22, 2018 2:29 pm
Location: North Yorkshire

Re: Enova(!) Parameters

Post by tom3141 »

Would it be possible to send across your parameters Peter? I did a bit more of a test today and although it is driveable I feel the parameters need some further tweaking. Encoder problem seems to be completely fixed though which is great.
User avatar
Peter
Posts: 310
Joined: Fri Dec 14, 2018 9:07 pm
Location: North West Lancs, UK
Been thanked: 8 times

Re: Enova(!) Parameters

Post by Peter »

Hi tom3141. I will copy them later and post here. I have made some interesting discoveries with adjusting parameters away from calculated values too.
User avatar
Peter
Posts: 310
Joined: Fri Dec 14, 2018 9:07 pm
Location: North West Lancs, UK
Been thanked: 8 times

Re: Enova(!) Parameters

Post by Peter »

Hi tom3141. Attached latest parameter file. Please note these are relevant to my setup so treat with caution. Any blown, fried, boiled or char grilled igbt's, etc I can not be held responsible for, ok. :)
Attachments
CombiV8 Settings_01_07_20.txt
(1.86 KiB) Downloaded 91 times
tom3141
Posts: 157
Joined: Sat Dec 22, 2018 2:29 pm
Location: North Yorkshire

Re: Enova(!) Parameters

Post by tom3141 »

Thanks Peter. Of course, I will take full responsibility for any blown IGBTs :)

There’s some quite significant differences from my parameters which is interesting.

Did you find it was best not to scale the parameters with voltage then (setting udcnom to 0)?

What is your nominal pack voltage?
tom3141
Posts: 157
Joined: Sat Dec 22, 2018 2:29 pm
Location: North Yorkshire

Re: Enova(!) Parameters

Post by tom3141 »

Also, am I right in thinking you have it in a Smart car? What is your transmission setup?
User avatar
Peter
Posts: 310
Joined: Fri Dec 14, 2018 9:07 pm
Location: North West Lancs, UK
Been thanked: 8 times

Re: Enova(!) Parameters

Post by Peter »

Hi tom3141. I did not use the udcnom function as such as I wanted to see what happens with known values so kept at 0.
Pack is 220v, 65Ahr cells are from a Fluence, same as Leaf I think.
Enova motor with adaptor plate to Mitsubishi F5M41 gearbox.
Using 2nd gear giving 72mph at 9000rpm
1st is too low... 38mph at 9000rpm
Yes its a Smart :-)
tom3141
Posts: 157
Joined: Sat Dec 22, 2018 2:29 pm
Location: North Yorkshire

Re: Enova(!) Parameters

Post by tom3141 »

Great, thanks.

What’s your performance like?

Mine is direct drive to the diff so 6000rpm will be about 75mph.
User avatar
Peter
Posts: 310
Joined: Fri Dec 14, 2018 9:07 pm
Location: North West Lancs, UK
Been thanked: 8 times

Re: Enova(!) Parameters

Post by Peter »

Hi tom3141. Performance is good, have you ever driven a Bugatti Veyron ? Well... its nothing like that :lol:
I suspect 0-60 is around 10-12 seconds maybe ? I will do some testing this weekend if the roads are dry.
To be honest my original idea was to be able to pop a small wheelie !!Thats no bull droppings either.
Could possibly happen if I select 1st and go on a diet :-)
Peter
Post Reply