[bernext]

Hi,

I'm Bernardo from Mexico City. I bought an EV kit from ElectricMotorSport, but I messed up with the configuration inside the Sevcon Drive.

Does anyone know how to use the configuration software named DVT from Sevcon ?

My EV is not moving anymore, and I have checked all the cables. I think it's a problem related to the objects or registries inside the drive.

My battery pack shows a voltage of 79V. I'm using a Gen4 Size 6 Sevcon Drive. My motor is PMAC DLC28, ME1302.

Digital inputs from FS1, FWD and REV work fine. Analog input from throttle varies between 0.02V and 11.2V, when actuated. The velocity object reports values when spinning the wheels manually. Temperature reading is ok. The main contactor is closed, when turning the key switch. The EV is in Operational Mode. CAN bus freq is at 1Mhz. I have no clue what did I change by mistake.

Regards.

 

[frodus]

Did you call them to see if they can give you a hand?

What did you change when you were in DVT?

What faults are showing?

 

[bernext]

Hi frodus,

There are no faults shown at all. The EV won't simply move, the wheels are not turning.
Sevcon says I need to contact the dealer. The dealer wants to charge me 2000 USD for a diagnostic service. I would like to figure this out without paying 2000 USD.

I don't remember what I changed in the configuration. That's why this is so misterious.
Today, I've already checked these items:

-Encoder, works perfectly, Velocity object shows speed of wheels when turning them manually, CW and CCW. Enconder is RMB29AC01SS1, works at 5V.
-The thermistor works, motor temperature is ok. Temp sensor is PTC KTY84.
-The motor control gains are: Kp->0.059, and Ki->0.019
-The values from motor plate are 550A max for stator current, Stator Inductance Ls 81.95uH, number of poles 4.
-Number of inputs and outputs are correct. Three digital inputs, FS1, FWD and REV. Analog input , just one, the throttle. The analog input from thermistor, doesnt seem to be able to be mapped in the screen. Maybe is being read by default.
-All the cables are in place, and have passed the continuity test.
-The battery pack is at 79V, and it's fully charged. I have an Elithion BMS, and I've also seen the individual voltages of every single cell, and 23 out of the 24 report 3.32V, and the other one 3.2V. THe SOC is 100%, and so as the SOH. There are also no leaks towards the chassis.

I can't find a jogging function, as in industrial drives. I want to verify if the motor is not fried. I think I'm going to connect two windings into an oscilloscope, and turn the wheels around, and check if I see some voltage.
Or maybe, connect the batteries directly to the two windings, just for a sec, to see if the shaft moves a little.

What do you think ?

 

[frodus]

I do consulting on the side for Sevcon stuff (see my site). Let me know, and we can set up a gotomeeting. I do charge, but it's far from $2k. I have the files for that motor if we need to reload.

I'm surprised Electricmotorsport wouldn't help.

 

[bernext]

Hi frodus,

How much would it cost ? Are we talking about a diagnostic service ? Do we need video ? skype ? I have to set up a lot of things for that to happen, because my EV is downstairs in the parking lot, and I live in a 6 floor building.
Would you accept bitcoin as a payment mechanism ? litecoin ? credit card ?

I've been struggling with this topic for quite a while, and I would just like to make sure that my EV problem has a solution in the horizon, because we could end up hearing ... "man, after spending 8 hours, I gotta tell you..., your PMAC motor has been demagnetized, u need a new one", or what about "you need a new characterization of your motor, the current one is of no use anymore". If any of these two scenarios were the reason why the EV is not moving, I would really just throw the project to the garbage. It's of no use for me, at least, having motors that can be demagnetized or can loose properties or change their intrinsic variables, so much, which would imply to perform future characterizations or would need an adaptive control technique, which the Sevcon drive cannot perform due to fix gains in PI control in the actual DCF file.

I once got in contact with John Fiorenza from MotEnergy, and he sent me DCF files for my motor ME1302, "allegedly". Then I realized two things:

1. The version of the files were 28, and my drive had 04, and there's no way I can upload them without incompatibility warnings showing up. After this, I realized there has to be an harmony between the DCF, DLD and ESD versions of files.
2. Those files were for sizes 2 and 4, and my drive was size 6. After that, he pointed me to the dealer again.
3. It seems that configuring an industrial drive from Yaskawa, AB, ABB, Siemens, even Parker, is easier than a Sevcon drive. I understand there are a lot of safety issues involved (because of lawsuits potential), but it seems that Sevcon is adamant to share technical notes. I have a full access license in the DVT software, and I've read the "Gen4_Product_Manual_V3.0" and the "DVT Customer Manual", checking a lot of stuff with no success.

Moreover, given your expertise, would you please tell me:

Is a jogging function nonexistent inside the sevcon drive ? I want to make a little turning test. Is this called "inching" in sevcon jargon?

Are there any technical notes, that are exclusive to dealers and Sevcon people regarding the configuration of their drives?

I would like to know, if I happened to read them, would I be able to figure out what's going on? I do have a Mechatronics Engineering degree, and I've always thought that by reading, I could solve anything regarding machines or systems.
The whole point of this adventure, is to learn how to troubleshoot electric drives in EVs, and I thought that by buying one of these EV-kits, I would really gain this knowledge, but it seems there's a barrier from Sevcon. I once got an email, which basically said, they only do training where there are business opportunities. This made me wonder, ... what about if there were an initiative to create an open source electric drive ?
Would it not be interesting ?

 

[bernext]

Hi again Frodus,

After the oscilloscope testing, I think I'm gonna try playing with the PI gains. At least, if the integral gain goes to 0, I think I can play with the Proportional gain to see if there's a change.

If the intrinsic variables of the motor changed, and I don't have the SCWiz, it seems logic to me, to try to play with the proportional gain, to see if the wheel move a bit.

Also, I'm considering rising the creep voltage.

Regards.

 

[bernext]

Hi Frodus,

I'm sorry to bother you again.
I'm following the instructions from the “Vehicle Setup and Configuration” application note from Sevcon.
I'm trying to use the DVT to control the torque demand being fed to the motor control directly, overriding demands fed in from the Traction Application.
I've reached the part of writing Torque demands to 0x6071, but I cannot get the motor to turn.

Because of this, I've checked the following:

-The controller is in the operational state, and powered up with the line contactor closed and there are no faults at all.
-The statusword of the motor control read from 0x6041 DOES NOT have the following bits set: 0x0027, IT HAS the value of 0x0421 instead.
Do you know the reason for this ?
-The actual torque output read from object 0x6077 is 0x0000, and is different from the value previously set to 0x6071,( i.e., sdo_wnx 1 0x6071 0 15 2 ).
According to the application note, If these two are different between each other, then it is being limited by either the power limit map, low battery or over temperature.

Battery is at 100%, 79V-80V.
Temperature is at 26 °C.
The power limit map, shows a maximum torque of 100NM at 0 rpm, for a motor which is rated to have 133NM tops.

What do you think it's going on ?

The voltage read between B- and any motor phase is 11.36VDC. It never reaches approximately half battery voltage between B- and each terminal, after enabling the bridge in the controller and activating the
motor control.

Regards.

PS.

Drive data:
Type: Gen4 72/80V 550A
Mod# PMAC-DLC Traction, Rotation CW

Motor data:
Motenergy ME1302 Water-Cooled Brushless
PART #DLC28

 

[frodus]

I've emailed you. This level of CAN integration is beyond what I normally do. If you need help, I won't do it without a support/consulting agreement in place, if I even want to take it on. It was not clear from your earlier message you were trying to do this directly, rather than standard throttle control.

Get it working in normal operation first. Then work towards CANbus control.

 

[LethalSkidmarks]

hey Bernext,
I know this is an old thread but I am hoping I could get some help.

I am having a very similar problem to you. I am trying to use the DVT software and follow Sevcons application note in order to make my electric motors spin. Following all their instructions, I get to the portion where I input my torque values. The motors will buzz but do not spin.
-The controller is in Operational mode
-the status word has is reading 0x0437 (i do not know if this is an issue. the bits 0x0027 are set. but there are extras)
-object 0x6077 does not read the same torque value i wrote to 0x6071


I was hoping you were able to figure out the issue with your, and someone would be able to help me. Thank you!