EcoModder.com

EcoModder.com (https://ecomodder.com/forum/)
-   Open ReVolt: open source DC motor controller (https://ecomodder.com/forum/open-revolt-open-source-dc-motor-controller.html)
-   -   Vehicle Ground Problem (https://ecomodder.com/forum/showthread.php/vehicle-ground-problem-24252.html)

J-Ro 12-08-2012 06:54 PM

Vehicle Ground Problem
 
Hi Guys,

I've been doing my first few test power-ups and finally moving my EV under its own power.

I noticed my Iota AC-DC (Now DC-DC Converter) wasn't charging my 12 volt system battery. In diagnosing this I discovered (not sure if this is the iota issue or not) that the vehicle chassis ground is at a high voltage in reference to the high voltage ground (i have a 144v pack). Now I'm less concerned about the Iota working and the fact that my grounds are somehow intermixed.

I disconnected everything, the DC-DC converter, Precharge circuit, Contactor circuit, low voltage wires going to the Open Revolt Controller down to the point that only all the heavy cabling is connected. Still I have this very large potential on the vehicle chassis.

Right now I have 134.2v sitting in my controller capacitors because the Iota had been hooked up for a period of time and drew them down to that level from the ~155ish v full battery charge and the chassis ground is at 107.7v. I haven't been able to develop a relationship between the CAP voltage and vehicle ground but I do know that when there's less volts in the caps there is less vehicle ground potential as well.

Any ideas?

JoeG 12-08-2012 08:51 PM

Hey J-Ro,
It sounds like somewhere on your batteries ground cabling you have a short/contact to the chassis . I would start by removing the negative cable on your pack and see what a volt meter reads bettween the chassis and the most positive terminal on your pack. It should be zero. If it isn't zero then somewhere in your pack you have a ground issue. If you have a lead acid pack, it could be an acid and dirt film causing a surface short to the frame. You can find the cell that is shorting to the frame by testing from the battery terminals to chassis ground and when the voltage reading is zero you found the area of the short. If the voltmeter reads zero when you disconnect the negitive battery cable, then you just have work down the line disconnecting things until you find thing that is grounding out.
Hope this helps,
Joe

J-Ro 12-09-2012 12:32 AM

I am using AGM batteries. They are pretty new and in pretty good shape. My pack cabling is mounted on stand offs and in convoluted tubing and I even have gone under and re-verified that there is an airgap between both pack wires and the vehicle chassis the entire run forward.

Does anyone know if the Open Revolt controller needs to have it's Aluminum heat sink isolated from chassis ground? Besides the motor itself (a brand new FB1-4001), this is the next closest place i can think of that the pack gets to touching ground.

J-Ro 12-09-2012 01:30 AM

I unbolted the controller from the chassis ground its hink sink is in contact with and tried to jam a peice of plywood between the controller heat sink plate and the chassis. Becasue of the bad angle and the time of day I'm not completely sure i got the contact 100% eliminated but it did reduce the voltage between pack neg and chassis neg down to about 3 volts with about 140 volts in the caps....does this help anyone out???

MPaulHolmes 12-09-2012 11:11 AM

I haven't heard about anyone isolating the controller base from the car frame. The controller base shouldn't be connected to anything electrical on the controller. I would check to see that resistance from B- to aluminum base, B+ to base, and M- to base is infinity.

J-Ro 12-09-2012 05:48 PM

i Drained the caps and measured the resistances:

B+ to Heat sink is 6ohms and seemed to be slowly falling the longer I held the multimeter there but may have been more of a hover.

M- to heat sink is 20 ohms steady

B- to heat sink started at 41ohms and quickly rose up to 427 ohms and stayed there. After going B- to heat sink I now had 0.296v stored in the caps. I assume the voltage and the rising resistance had to do with the multimeter pumping power in to measure the resistance?

So what does all of this mean? I feel like I knocked the assembly of the controller out of the park so I'm not instantly ready to say the issue is from shotty workmanship. Do these readings point to a specific area where the problem may be?

Thanks for the help!

kennybobby 12-09-2012 07:53 PM

You have a short circuit
 
Quote:

Originally Posted by J-Ro (Post 344545)
i Drained the caps and measured the resistances:

B+ to Heat sink is 6ohms and seemed to be slowly falling the longer I held the multimeter there but may have been more of a hover.

The battery + is shorted to the heat sink, maybe a buss bar on the capacitor bank, etc. Your vehicle is floating at B+ due to some short circuit to vehicle chassis on the high side. Post a schematic of your wiring and the controller, otherwise there is not much else anyone can do to help--you're going to have to ring it out.

In the meantime i went to the wiki page to look at schematics for the open revolt. The only set i could download was for the 500A rev 2B, all the rest were locked up in some sort of online file saver (useless POS). Anyway the Mosfet section looks totally flaky--i've never seen a totem pole or H-bridge motor driver circuit more off base and fubar. What's the point of the dc link capacitors when you're not switching the buss but the low side? What's the point of the 10 diodes when they are not protecting the FETs? Plus putting a reverse diode on the the battery ground terminal of the dc/dc converter? Where are you guys getting this crap?

MPaulHolmes 12-09-2012 10:08 PM

I'd first disconnect M+, B-, and B+ bus bars from the battery pack and the motor. Also, just for the heck of it, disconnect the 5 low voltage wires that go to the control board too. Then measure resistance from each bus bar to base plate. If you want to be really thorough, maybe do the measurement with the whole controller out of the car, just sitting on a piece of wood or something. In that context, the resistances should be infinity. The isolation sticky pad stuff can insulate from 7000 volts.

J-Ro 12-10-2012 09:56 AM

Hey Paul,

I took the controller out of the vehicle and re-measured everything. I still see some continuity but they are in the several hundred Kohm range.

Whats my next step? Dis-assemble the controller and try to better isolate the mosfets and diodes from the heat sink?

I visually inspected the mosfets and diodes and can clearly see the isolation material covering 100% of the area behind all of the mosfets and diodes so im not really sure what I would get starting over on that.

MPaulHolmes 12-10-2012 11:25 AM

No, don't take it apart. If it's several hundred kOhm, I'm sure that's fine. That can just be related to the volt meter sometimes. My fluke can show several hundred kOhm, or infinity, measuring the same 2 points at different times. The leakage from that would be very small. I think your short to the chassis is somewhere else... It could be the motor causing it too. Sometimes carbon dust can make a connection where it shouldn't. Try the motor posts to the car frame also, and then each battery post to the frame of the car. It's got to be somewhere... I love mysteries like this.

kennybobby 12-10-2012 08:50 PM

If you wired up your controller using the 500Amp cougar mosfet schematic rev 2D that is found on the revomit wiki page--it has a serious labelling error on connector J5 from the controller board to the power board. J1 and J12 are incorrectly indicating their polarity and connection to the battery pack B+ and B-.

J-Ro 12-10-2012 09:38 PM

If I Isolate the Controller from the chassis the problem goes away.

Seems like there should be a better solution than this but I'm not oppsed to this being the solution.... :confused:

J-Ro 12-21-2012 11:59 PM

Hey paul,

Is it at all possible that a failed mosfet or diode could be causing this?

Is it possible forr either of these failures to occur without knowing it?

Ive pulled in the pretty high 400 amps from my battery pack (according to my own external data logging equipment) on some of my recent hard accelerations so Idunno if that would be doable with a failed mosfet.

Thanks!

MPaulHolmes 12-26-2012 01:53 PM

A failed mosfet would show itself by being a dead short, which would then blow the source leg clean off. If isolating the controller from the chassis makes the problem go away, there's no way it's just a failed mosfet or diode. So I think you are good there.


All times are GMT -4. The time now is 03:59 PM.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.
Content Relevant URLs by vBSEO 3.5.2
All content copyright EcoModder.com