No readings from shunt on Zombie web interface

[Iceman96]

Hi BrainsTrust

I am after some help with a problem i am having with my IVT-s shunt not displaying values on the Zombie web interface. I only recently received my Zombie, and am trying to do a test run on my full leaf stack with it out of the vehicle. Here is a list of what I have:

Zombieverter received in July this year running firmware version 2.17
IVT-S-1K-U3-I-CAN1-12/24
2014 full Leaf stack. ( confirmed working as I drove the leaf before stripping it)
24 kwh leaf battery pack from the same vehicle.
12 volt 10 amp bench power supply

I received my Zombieverter kit in July this year, and after soldering on the parts supplied with it I was able to power it up and communicate with it via wifi on my ipad. I then configured it to work with the Leaf stack.
Next i connected it up as per the wiki, with the inverter and pdm connected to CAN1 on pins 27 & 28 on the zombie, and the ivt-s shunt on CAN2 on pins 43 & 44. I also tested and confirmed the correct operation of the precharge, Positive, and Negative contactors in the battery pack.
I then supplied 12 volt power to the zombie and the shunt, and when looking st the spot values for voltage on the web interface I was getting random values showing for all 3 voltage readings. I then followed the procedure in the wiki for initialising the shunt with the zombie ( including powering up the shunt 2 to 3 seconds before the zombie), but that made no difference. After looking through the forum pages, i found another suggestion of changing the shunt over to CAN1 selection on the web interface. I did that and tried the initialisation procedure again, and this time it appeared to work. Now when I power it up i am reading a rock steady Zero volts on all 3 voltage readings on the web interface.
Next I applied a 12 volt “ignition” input to pin 15 on the zombie - nothing happened - I checked the spot values and could see the zombie registering the input on pin 15. From my reading of the wiki, I was expecting the relay providing 12 volts to the inverter and pdm to energise, but nothing happened. I then applied a momentary 12 volt input to the “ start” input on pin 52 of the zombie. Immediately the relay energised providing 12 volts to the inverter and pdm, the precharge and Negative contactors in the battery pack closed, the zombie indicated it was in precharge mode, and I saw the battery pack output rise quickly to 375 volts as measured with a Fluke multimeter. Checking the spot values there was still zero volts indicating on all 3 volts displays, even after refreshing the spot values. After approx 5 seconds I heard a contactor drop out in the battery pack, and the zombie went into “precharge fail” mode. Interestingly, when I checked the contactors, it was only the precharge contactor that dropped out - the Negative contactor remained closed. However It did drop out once I removed the power to the Zombie. I repeated this process 3 times, with the same result each time.
I am surmising that the zombie was going into “precharge fail” due to the fact that it was not seeing any voltage reading within the 5 second window after the precharge and Negative contactors closed.
I then decided to try another test where i powered up a 12 volt 4 amp load ( a headlamp bulb) through the IVT-S shunt and monitored the spot values on the web interface. Once again the values for volts and current were steady on zero.
So I still appear to have a problem with my shunt, and am looking for sugestions. Should I try hooking it up to an arduino due with my 4 amp load going through it, and see if I can get some readings off the serial monitor? Or are there any sugestions for something else to try first?

TIA

Andrew

[royhen99]

The CAN1 and CAN2 pinout on the wiki were incorrect up until May last year. Now the pinout and interface selection on the web page match. Check wiring and selected interface match i.e. both CAN2 ( or CAN1 ), and maybe the IVT-S shunt is not initialised so try that again. Also it is not clear if you have connected one of the voltage inputs on the shunt to battery or power supply positive.

On the web interface have you selected ISA under ISA shunt Control - Type ?

[tom91]

Download the parameter .json and attach

Have you used terminated each CAN bus? When they are unpowered each bus should have a resistance of 60 ohm between high and low.

[Iceman96]

The CAN1 and CAN2 pinout on the wiki were incorrect up until May last year. Now the pinout and interface selection on the web page match. Check wiring and selected interface match i.e. both CAN2 ( or CAN1 ), and maybe the IVT-S shunt is not initialised so try that again. Also it is not clear if you have connected one of the voltage inputs on the shunt to battery or power supply positive.

On the web interface have you selected ISA under ISA shunt Control - Type ?

Hi Royhen99,

Thanks for taking the time to try and help me. I connected eveything up exactly as per the zombieverter wiki for the Leaf stack connected to a zombieverter. On there it shows the inverter and the pdm on the “ ev system can” ( pins 27 & 28 on zombie), and it has the shunt on “CAN 2 “ - ( pins 43 & 44 on zonbie). Then in the web interface parameters I selected Can1 for the inverter and the pdm, and Can2 for the shunt. ( There is only the option to select Can1 or Can2 - there in no Can0 to select ). As per my earlier description of my problem, I found that I could not initialise the shunt when I had it on Can2, but after changing the selection to Can1, it then appeared to initialise ok. So I left it on Can1. I have since tried various combinations of Can selections, and the only thing I know for sure is that I have to select Can1 for the shunt, otherwise the 3 voltage readings in the spot values show random sporadic values. But when it is on Can1 selection i am getting steady state values of zero for the voltage and current.

I have tried re-initialising the shunt today, but am still reading Zero volts and amps.

And yes, the voltage input to the shunt was connected to the positive output from the battery pack, and I was reading approx. 375 v on there with my fluke multimeter until the precharge contactor drops out. I refreshed the spot values page while the meter was reading 375 v, but the web interface still displayed 0 v. I have my voltage input to the shunt connected to the top right input on the shunt as per the ones I have seen on youtube, but I have also tried it connected to the other voltage inputs, but still no volts reading on the web interface.

And Yes, I have selected ISA as the shunt type, and I have turned the “ shunt initialise” off as well after I performed the initialising of the shunt.

Any other ideas? I hope I dont have a faulty shunt?

Thanks - Andrew

[Iceman96]

Download the parameter .json and attach

Have you used terminated each CAN bus? When they are unpowered each bus should have a resistance of 60 ohm between high and low.

Hi Tom91

Thanks for taking the time to try and help me. I have downloaded and attached the json file showing my parameters. And that was a good call to check the CAN termination resistors. I checked both CAN buses, and the shunt one was good, showing 60 ohms. But the EV system can that the pdm and inverter were on was showing 120 ohms, so no termination resistor on one end. I split the Nissan factory splice on the ev system can bus wiring and can confirm that the zombie has a termination resistor, but neither the pdm or inverter have one. I had assumed that 1 of them would have had a termination resistor in them, but obviously when they are installed in a Leaf, neither of them is at the end of the line. I will get a 120 ohm resistor and connect it across one of their can inputs and see how that goes.

Also, do you know if the voltage and current values displayed in the web interface spot values come from the shunt ( as I assumed), or are they provided by the pdm or the inverter?

params 4.json

(1.95 KiB) Downloaded 94 times

Thanks - Andrew

[royhen99]

From your description the shunt should be on CAN2 ( both connection and web page selection ) and is currently not initialised hence the erratic readings.

[Iceman96]

From your description the shunt should be on CAN2 ( both connection and web page selection ) and is currently not initialised hence the erratic readings.

Thanks for that. I thought it should be on Can2 as well. And that is how I originaly connected it up. But when I tried to initialise it on Can2 I got no result, and I kept getting nonsense values. Yet as soon as i changed it to Can1 on the web interface ( with no changes to the physical wiring), it appeared to initialise ok, as the voltage readings were then stable on zero volts. But then when it has a voltage applied it remains at zero volts.
The other thing I am confused about, is I also followed Tom91’s sugestion, and when I looked at the json file of my parameters, it shows the devices i put on Can1 as having a value of 0, and the devices on Can2 as having a value of 1 ?? I would really like to talk with someone that has a zombie connected up to a Gen2 leafstack, so I could confirm exactly how they connected everything. But following Tom91’s other suggestion, i found the the can bus I have my pdm and inverter connected to wasn’t terminated correctly at one end, so I am going to fix that today and see if that makes a difference

[Iceman96]

Hi Brainstrust,

Success at last!! I not only got my shunt to supply values, but I even got the motor to spin in both directions! So I am a very happy chappy, and I’d like to say many thanks to Royhen99 and Tom91 for your help. This is how I solved it.

After following Tom91’s advice and checking my can buses, I found that Can1 did not have a 120 ohm termination resistor at the inverter end of the bus. So I found a resistor and added it in and then checked and confirmed that with the power off I was now reading approx 60 ohms between CanH and CanL on that bus. I powered everything up and ran my tests again, but unfortunately still no change. No voltage or current readings.
So I then went back and re-read Royhen99’s suggestion that I check my wiring, and then make sure it is wired up as per the wiki, with the shunt back on Can2. I would like to clarify here that I am a retired Electrical technician, and am very experienced with PLC’s and control systems, so am quite used to following wiring schematics, in fact that is what I have always worked off. And that is exactly what I did when I connected up my Leaf stack. I followed the wiring schematic on the wiki page which clearly shows that the shunt is by itself on Can Ext 2 and that the PDM and the Inverter are together on Can Ext . But then I decided to read the directions on the wiki that explained where things were connected. And this is where I found a discrepancy. In the directions for how to connect everything it says that the Inverter is connected to Can Ext ( pins 27 & 28 on zombie), and that the PDM is connected to Can Ext2. There is no mention at all in the directions about how to connect the shunt to the zombie.
So with nothing to lose I decided to try a mixture of the directions and the wiring schematic. I changed my physical wiring so that now the Inverter was on Can Ext ( zombie pins 27 & 28) by itself, and I wired the pdm onto Can Ext 2 with the shunt. I then went into the web interface and selected CAN1 for the inverter, OBD2Can, and the CanMapCan. And I selected CAN2 for the shunt and everything else. I then followed the procedure to initialise the shunt again and Eureka! My shunt was now showing real values.
So emboldened by my success, I then turned the inverter on, and gave it a start pulse. The contactors did their thing and the zombie went into Run mode. I then selected Forward direction and got feedback that I was in “Drive” mode. A gentle application to the throttle pedal ( I am using the one I took out of the Leaf), and my motor smoothly spun up. I then tried it in Reverse and that worked too. So I decided to quit while I was ahead, and I shut everything down.
Have not tried charging yet, so not sure if that is working yet, but that can be a job for tomorrow.

[mattes]

My ISA shunt init went without a glitch and now it reports the correct voltages. Have not tested the amps yet.

It should work regardless if you use CAN2 (pin43/44) or CAN1 (pin27/28).

I was tracing the CAN message on both can busses and I can confirm both selection send out the needed initialization for the ISA shunt
Should look something like this:

Code: Select all

$ candump can0 |grep 411
  can0  411   [8]  34 00 01 00 00 00 00 00
  can0  411   [8]  20 42 00 64 00 00 00 00
  can0  411   [8]  32 00 00 00 00 00 00 00
  can0  411   [8]  21 42 00 64 00 00 00 00
  can0  411   [8]  32 00 00 00 00 00 00 00
  can0  411   [8]  22 42 00 64 00 00 00 00
  can0  411   [8]  32 00 00 00 00 00 00 00
  can0  411   [8]  23 42 00 64 00 00 00 00
  can0  411   [8]  32 00 00 00 00 00 00 00
  can0  411   [8]  24 42 00 64 00 00 00 00
  can0  411   [8]  32 00 00 00 00 00 00 00
  can0  411   [8]  25 42 00 64 00 00 00 00
  can0  411   [8]  32 00 00 00 00 00 00 00
  can0  411   [8]  26 42 00 64 00 00 00 00
  can0  411   [8]  32 00 00 00 00 00 00 00
  can0  411   [8]  27 42 00 64 00 00 00 00
  can0  411   [8]  32 00 00 00 00 00 00 00
  can0  411   [8]  28 42 00 64 00 00 00 00
  can0  411   [8]  32 00 00 00 00 00 00 00
  can0  411   [8]  34 01 01 00 00 00 00 00