Sevcon Gen 4 Errors and Solutions thread

Hello methods !

Sorry for turning to You directly, but I went through the whole internet and called all possible people but You seem to be the only one in this topic who really knows what he's talking about and digging deep enough...

I have a microcar type "Romet 4E" (You can check it in Google Images as it's most probably unknown to You) that uses a Sevcon Gen4 8035 Size4 controller. (setup is mostly the factory default single-traction-motor, but I could figure out some modding as well, although the flashed-in program is unknown. I contacted the manufacturer but they don't want to disclose information)

Original battery setup was 9pcs 8V 135Ah Lead-Acid, but I converted the car to Lithium (24pcs 3.3V 200Ah LiFePO4) for a much better performance, lower weight, and to extend the EV range to more than twice of the original.
So the system voltage went from nominal 72V to 79.2V (which is no problem for the 8035 as the nominal voltage is 80V and the maximal is 116V as far as I know). I also made some internal mechanical modifications but that's irrelevant at this time.

Now the next step to tune the car further is to extend the speed limit and improve the performance of the regenerative braking.

There are two RPM limits programmed into the controller, the one belongs to the "normal" driving mode, the other to the "economy" mode, which allowes You to drive (through the single fixed gear) 60km/h in normal mode and 40km/h in economy mode.
Certainly I never use the economy mode as it can be achieved by handling the accelerator pedal more gently, I use it for regenerative braking as switching the "eco" mode during driving brakes the car back to 40km/h using regenerative.
So my problem is obvious: 60km/h for normal driving is too slow, and braking back to 40km/h only is too less regenerative braking.
I would like to change the two RPM limits so that the normal limit is 80km/h and the economy limit is 10km/h, not because I want to drive 10km/h but this would allow me to make regenerative braking down to 10km/h instead of 40.
Also the 80km/h has some risk because of the fixed gear, it might over-RPM the engine, so I don't want to drive this speed continouosly, but every now and then there might be a need during overtaking to go faster than 60km/h to make the overtaking shorter.

On the default Gen4 setup there is no "normal mode" and "eco mode", so this must have been programmed by the manufacturer, and I was tracing back the wires and it turned out that the "eco-switch" is connected to "Digital Input #5" (pin 20) of the Gen4 connector: it is floating at high impedance (0.4 x Vbatt) when in "normal mode" and pulled down directly to GND in "eco mode". This pin20 must have been mapped to the "economy" variable inside the controller's program.
So far so good.

Now my idea is to use a ClearView display to program the RPM limit(s) if possible. But before I order such an expensive toy I would like to ask someone with experience (this means You) how the ClearView is working and whether it is possible to program the RPM limit(s) of the motor.
I have seen a YouTube Video (https://www.youtube.com/watch?v=2ryWTccXb-I) where some guy was changing the RPM limit of the motor (Menu-->Personalities-->Baseline-->Max Speed FWD), but my controller seems to be programmed to two different limits, and I don't know if it is possible to program more limits with the ClearView? May be this is solved by using the Profiles? (Menu-->Personalities-->Profile1 and Profile2), and the "eco-switch" is just switching between these 2 profiles?
I don't know as the video does not step into the Profiles1 and Profiles2 menu...

The other thing I would like to program with the ClearView is (beside the two RPM limits) is the efficiency of the regenerative braking. At the moment I am not satisfied with the braking power of the regenerative braking.
My suspection is that the "maximum deceleration rate" is limited in the controller, not to overcharge and blow up the original lead-acid batteries, as they can only be charged slowly with a smaller current. But now the car has Lithium that is able to absorb much more power and much more charging current, allowing the car to decelerate much harder, therefore the maximum deceleration rate could be set up for a higher value, so I could brake better with the engine and regenerate more energy instead of using the hydraulic brakes...

So any idea or suggestion or experience is very welcome that helps me further!
Thank You very much! (sorry for writing a complete roman out of it...)
---
Oliver
 
Hello Oliver!
Can You save your dcf from sevcon and send it for me?
I will try to understand your motor parameters.
Also you has abbility to rise up votage to 100 V an example and take more power from your motor!
 
Hello sanfox,

I have no any tools for the Gen4 controller, so I surely can not extract the data out of it...
My questions for the Gen4 in my previous post were to be able to decide if I should buy a ClearView Display, to be able to adjust some parameters.
The Gen4 came pre-programmed by the car manufacturer, who do not produce the car any more, and they refuse any help when I tried to get some over the phone...
 
Oliver, I can give you contacts of a guy from Poland, which has DVT dongle and his oun progect with sevcon powered car.
If you intresting that send me PM.
 
Is any one still active on this thread??
I am trying to do a similar project with a Me1114 motor Sevcon Gen 4 size 6 controller
Learning how to this from scratch is not for the faint heartted. Getting no help from Sevcon either.
I try one thing at a time and then re save the dcf file, just to go back if needed.
I have the motor spin for a second and then get an over current error.

Would like to know if the Sevcon Clear View worked out for you project as well. I need to
get a display panel but not sure what one to go with??

Thanks, for any help.. advice..
Mark
 
Hey Mark,
I'm sure there are many things that could cause your error, but I have been dealing with similar symptoms on a Parker GVM motor (3ph PMAC, sin/cos encoder) and gen4 size 6 controller combo that I've 90% narrowed down to EMI to the motor encoder. While any of the experts can jump in to make corrections, I'll let you know my best recommendation based on what I did SO FAR. Note, I haven't completely fixed my problems so I'm still lurking this page once in a while.


1. Have you done the things in this thread? https://endless-sphere.com/forums/viewtopic.php?t=88647
Still plenty you can do without an oscilloscope, but it's better if you do have one.

2. Is it under load, or freewheeling in the air? (applying the brakes counts as load).
If it trips while freewheeling in the air - I'm at a loss, that sounds like some bad parameters and you'll need to get the specs from the manufacturer.

3. If its fine when freewheeling, but trips the moment you're under load then you could be in a similar boat to what I had: EMI messing with the encoder values. How do I know? Because I mapped the encoder min and max values to TPDOs, and watched them on my laptop while spinning up the motor. Everything is happy and unchanged until I apply sufficient brakes and throttle (load) to the point where the motor starts to chatter and is about to trip on said Overcurrent fault: and guess what? My peak and trough encoder values spread wider JUST as that happens, when the load cranks up.

Even with Champion EXRAD braided shielded cables, grounding the shielding I still had this problem - but each time I added additional shielding, the problems slowly dissipated (encoder peak/trough values remain unchanged) until more and more amps were flowing. So if you're using unshielded cable (like welding cable) or if you haven't ground your cable shields, then this is at least something to fix. Parker recommended shielding each phase separately, and then the lot of the 3 together, those shields all going to chassis ground. The encoder should be shielded separately, going to a DIFFERENT ground point, and likewise with your traction pack cables being shielded and ground separately from the encoder. Simply having my encoder shield ground on the same chassis point as my 3x AC cables caused all sorts of hell for me as well. I was surprised to learn that.

4. Next, I found that the braided shields inside my 3 phases and DC cables weren't good enough for the level of noise generated, so I wrapped copper shielding tape around the outside of the cables and connected that to the braids, wrapping that with electrical tape. That helped a ton.

5. Finally, it wasn't completely resolved, so the mechanical engineer in me (as in, NOT EE) was questioning how well copper can absorb DC EMI. So I added steel wool to the shielding. It blew my mind, but this actually made a big difference too, the bike even SOUNDS quieter on the road. Tomorrow I'll wrap steel wool on the DC cables (my positive cable is near the encoder cable) to see if that finishes the job. Now, I only start to have these problems when I'm riding over 80 km/hr.

So, hopefully that helps, and if anyone else is reading this and has better advice for me to fix as well, I'd love to hear it!

Darsey
 
Excuse me.I am a beginner for sevcon gen4 and now i have met some problems about the Gen4.The first one is the node.The default value of the Node is 11 and 33.But i dont know how,now it just has node 33 and i klick the TPDO/RPDO bottom.It shows nothing but Sync Period without TPDO1,2,3,4,5 and RPDO1,2,3,4,5.I cant change the value of PDOs.Could you answer my question or give me some advise how to learn this software. Thanks a lot.
 
Darsey,

Thanks for the information. I have all the shielding grounded but to the same spot on the chassis.
I will repossession it to another ground spot. Makes complete sense that they should be separate and the wires
apart from each other as well. I have a scope but haven't had it out in several years, I guess now is a good time.

I have resorted to paying (OUCH) a tech to help sort the issues. Almost three hours so far. Although the motor
runs well right now it still has a little tweaking to go. Now it is the throttle with the issue, all the RPM's in the first quarter
of travel. I was advised two just use the wiper connection (pin#22) and ground (B-). But it did not work. The Sevcon as
Pin # 34 as power to the pot and it would not work without it. Most likely a software setting issue.

I Should know quite a bit more than when I started, but these controllers have hundreds of different adjustments.
Ming I cant answer your question right now but the tech basically was just playing with different settings till he got it
closer... I cant say it is perfect. I am also a little frustrated he didn't know more than it appeared. I will see about getting
some answers for you. I am paying him by the hour right now so I need to get my issues resolved first.

Mark
 
Hey Mark! Well hopefully you're making progress at least. I'm sure it'll come out alright in the end of the tech is at least familiar with the sevcon Gen 4.

In the last couple days, I found that my D-axis proportional gain was way too low. As soon as I brought that up, I was able to extract a ton more power, though that power diminishes very quickly. By 90km/hr I'm still at negligible torque before it trips on over-current.

Saying that, I'm just pointing out that I was wrong that it was all EMI. Even now, I still watch the encoder values start to spread when i crank on the torque, so it's not impossible... but with such a dramatic change at lower speeds from just a little Kp adjustment I think it's more to do with tuning again. I managed to use a couple of screenshots from other people's settings that were posted on endless-sphere here to start with, and found a messed up old Brammo dcf file floating around on the internet to try and reverse engineer their motor settings to get an idea how other people tune the D-axis here. If that helps you I can share what I've got but I don't think it's going to do much when you're running a different motor. Good luck!

Darsey
 
Darsey,

Yes, I am using the ME1114 motor platform and the Sevcon Gen 4 size 6 to get what I need done.
At this point I could give others with the same motor/controller number and setting to get them close.
But as I get more into this, ever setup is slightly different. So to be a perfectly tuned motor it needs to be
done on each individual setup.

Mark
 
Done this already, only to extranl site with expired password. Tried to register here for myself..
 
Hi,
Could someone help me fix this error with my Gen4 (0x5184, Alg Out On with No FS)
 
Good luck with the Sevcon guys. . .

I wrote a lot, in a lot of threads, and it is mostly garbage. . . but in there you will find some themes. It may be worth your time to sift for nuggets. See where I went wrong over and over.

Be very disciplined about how you mix the files, how you back up changes, etc.
There are a thousand ways that the Sevcon will make you sorry.

I lost a lot of life fooling with the Sevcon. . . . my final answer is that it is engineered to be supported by full time experts. You would have a hard time making something more complicated.

I.E. It is Engineered to not be DIY, so if you are DIY'ing, dont expect a smooth ride. Find an expert. Pay them.

I am not an expert.

-methods
 
hi all,

someone know how to change the spinning directon of a gen4 size6 wit an me1302 ?

i put switch in revers: motor turns fwd, revers switch is activated in software, rev is shown at clearviewdisplay
i put switch in fwd: motor turns rev, fwd switch is activated in software, fwd is shown at clearviewdisplay

swap sin with cos > does not work
swap M1 with M3 > does not work

any idears??
 
Hello electro doctors,

After an update of other ECUs, a Mia electrique C 2012 no longer wants to drive.

I have filtered the CAN bus on a few Sevcon motor controller (VAR) IDs.

Conclusion: the VAR is active on CAN bus.

ID 0x181 VAR_Stat1_EGV: Status OxXX50 => not ready
ID 0x701 VAR_hbeat_EGV (NMT error control): 05 => operational
ID 0x81 VAR_EMCY_EGV:
http://jan.peman.com/hacks/EmergencyErrorCodes.txt
10 82 10 A 53 0 0 0=>
Monitoring; communication; B4&B3: 0x530A PDO Short

0 10 11 82 46 0 0 0=>
Generic; communication; B4&B3: 0x4682 ????

Can someone explain these errors and call a solution?

Sincerely,
Miactief
 
Just thought I'd step in again since I last commented about the noise issues I was having and how it was finally diagnosed.

1. Contactor coil noise at 24v, running a 100v pack. Just like so many comments earlier, the PWM'ing is sending a lot of noise through the system. For me, it looked like it was all over the place. To troubleshoot, I removed the contactor output from the controller and used 3 series 9v batteries to drive the contactor (just to bench test). As soon as I did that, the ground bounce noise cleaned up so much I saw a perfect square wave with overshoot using a scope. That overshoot is still far too much noise that it throws off the sin/cos encoder readings. But it was a "perfect" 8kHz square wave from ground to absolutely anything else on the bike.

2. Since I kept B- isolated from frame ground, I used bypass caps on both my encoder +/- and pack B+ to B-. This smoothed out that ground bounce so much I was able to open up the throttle to 500A for the first time since building this bike. I didn't have resistors to run parallel and properly filter the signal, but that's next. Plus I'll have to clean up the wiring a bit before going riding.

Finally, once that filter is properly set up I'll see if the contactor coils still cause significant noise, and if a similar filter will work on that coil if needed, I'll do the same there too.

The "best" part of it all, after 3 years building a motorcycle during nights and kids' naptimes, I finally opened up the throttle for the first time, came out of the garage, and found Calgary's first major snowfall of the year had just started. The ground's already white and we're getting 2 ft by Monday! Dammit!
 
Can someone explain values versus volts or values versus other scaled decimal entries?
For example when setting up throttle.
Throttle start I am instructed to set to value=0 and throttle end value=1.
Volts start and volts end are straight forward.

I would like to set up user defined throttle.
Here my default settings, volts=value.

Sevcon thr values.gifSevcon thr curve.gif
 
Voltage is the throttle assembly's output voltage that the controller is reading, and the value is the corresponding throttle value as a percentage of full throttle (applicable torque/speed setpoint depending on your application) from 0 to 1.
Also worth applying a little safety factor on your low end, so the throttle doesn't activate accidentally. Set a dead band, say 0.5v applied before anything actually happens.

Darsey
 
I get it. Makes perfect sense that you can pick points on x and y axis.
So a value of .5 is 1/2 throttle. Pt.1
And setting voltage to 1/2 of total will simulate linear response 0 to 1/2 throttle.
Thank you so much.
I'm getting down to the fine tuning of drivability.
 
Another question if I might.
I'm out of 80v size 6 controllers that were used to run 8kw induction motors at 80v.
Will I be able to get by with gen4 model that is only rated for 300a peak, if I settle for lower performance? (Versus 550a peak)
Tomorrow I can turn down my s6 for a performance test. Is that valid?
 
You mean that you have a Gen4 size 4 available instead of a size 6? Yea, you'll have less torque available because the controller will have a lower peak amperage available, holding all other things constant. Do you have the ability to increase your pack voltage? The gen4 size 4 can come in a higher voltage rating, meaning higher top speed (provided the motor is made for it). Then it simply comes down to changing your gear ratio if you have room for the sprockets...
 
Exactly, yes I can use higher voltage. I'm testing 8kw motor at 66v with a gen4-6.
Will turning peak amps down to s4 levels be a good simulation of using s4?
Can run at 96v nominal.
I'm getting a price of some surplus gen4-s4 80v from Spain. Need to know if they will work first.
I could use about 20 controllers if someone finds a source.
Thank you again!
 
Back
Top