here is the screen shot and if I recall correctly even if you type in just a two byte ID it automagically fills in the other two bytes to make it a four byte IDarber333 wrote: ↑Wed May 11, 2022 3:54 pmYes in general you should be fine if Canalyst is blinking its leds. That means it is receiving CAN traffic. Can you post a screenshot of your send/receive screen? I dont have that DCDC with me anymore, but i can still help...
EDIT: that got me thinking... when you post ID into Canalyst you have to make it in 4 byte length for "standard" length (not extended).
So for ID 0x1D4 you would put it like 00 00 01 D4. Try it...
Msg repeats in 100ms intervals. Example from my AC compressor...
AC compressor CAN.jpg
I think I CAN, I think I CAN... Topic is solved
- Gregski
- Posts: 2348
- Joined: Tue Sep 14, 2021 10:28 am
- Location: Sacramento, California
- Has thanked: 360 times
- Been thanked: 481 times
- Contact:
Re: I think I CAN, I think I CAN...
"I don't need to understand how it works, I just need to understand how to make it work!" ~ EV Greg
- EV_Builder
- Posts: 1199
- Joined: Tue Apr 28, 2020 3:50 pm
- Location: The Netherlands
- Has thanked: 16 times
- Been thanked: 34 times
- Contact:
Re: I think I CAN, I think I CAN...
Type in : 1D4 (one byte and one nibble)
Converting an Porsche Panamera
see http://www.wdrautomatisering.nl for bespoke BMS modules.
see http://www.wdrautomatisering.nl for bespoke BMS modules.
-
- Posts: 3295
- Joined: Mon Dec 24, 2018 1:37 pm
- Location: Slovenia
- Has thanked: 81 times
- Been thanked: 242 times
- Contact:
Re: I think I CAN, I think I CAN...
You do not receive anything from DCDC. You did not even open the interface. You need to open the interface in Operation menu.
Can you connect another component to the same bus that you know it transmitts some telegrams? Maybe a charger? It seems like you are having similar issues than me from the start.
Also try changing CANH and CANL cables position. Maybe you mixed them up. There is no danger in doing that. You just dont receive anything from the line.
I am not directly familliar with this interface. It is a different software.
When you connect Canalyst to your bus does its CAN1 port LED lite up and blink continuously? That means communication is being received regardless if you have connection open or not.
Also try opening CAN bridge between both channels at 500kbaud.
I got my software here
https://www.zlg.com/can/down/down/id/42.html
Here is also some English version
https://www.hklrf.com/CANalyst-II-USB-t ... _5071.html
- Gregski
- Posts: 2348
- Joined: Tue Sep 14, 2021 10:28 am
- Location: Sacramento, California
- Has thanked: 360 times
- Been thanked: 481 times
- Contact:
Re: I think I CAN, I think I CAN...
The interface was open as you put it, this was just a screen shot I captured of the traffic I was trying to send to show you the format of the frames
Question: does the converter expect a minimum voltage from the battery pack?
"I don't need to understand how it works, I just need to understand how to make it work!" ~ EV Greg
-
- Posts: 1801
- Joined: Sun Apr 03, 2022 1:57 pm
- Has thanked: 102 times
- Been thanked: 348 times
Re: I think I CAN, I think I CAN...
I'm building up my GVRET based CAN dongle and have so far got to this point:
I've gone this route because I would like two can ports so that I can break into the CAN feed to a particular module (ICE engine ECU) and definitively see what it's receiving and what it's transmitting (one port goes to the bus and one port goes to the module under test). It's a bit more intrusive as you need to cut wires but should make it much easier to see which packets are being sent by the ECU which is what I'm interested in.
I've spent all morning trying to get it to work and discovered one very nice feature but also a few gotchas that are probably worth sharing.
The above shows my test setup with the ports looped back to each other. The problem was that every time I sent a CAN message both ports would co crazy. Initially I thought that this was because the bus was unterminated so added a couple of links to enable the on board terminations. This changed matters but not in a good way. I then tried disconnecting the two ports, at this point only one port went crazy when sending a message. It turns out that:
1.) The above CAN boards do not have a link to enable the termination. What looks like a link is a second set of bus connections - I'd shorted out the bus. There is a termination on the board that is always connected. Going to have to mod the boards to change this which is a pain because they are now soldered down!
2) GVRET has a feature where a couple of processor pins are configured to enable bus forwarding, pin 11 enables bus 0 to 1 and pin 12 enables 1 to 0. This is very nice as originally I was expecting to have to write a script to do the forwarding in SavvyCAN which is a very inefficient way of doing it. The problem is that I didn't know it was there and it is enabled by default (need to you short the pins to ground to disable). When I looped the ports and sent a message it was effectively ping ponging the one message back and forth forever!
3) If you send a message to and unconnected GVRET port it doesn't get an ACK back. The board then retries sending the message forever. Not found the code controlling this yet but guessing it is in the due_can library. I'd like to either disable this or limit the number of retries so need to keep looking.
I'm sure all the above are documented somewhere but have no idea where. If anyone knows please could they let me know as there may be other nice features hidden away in there.
With all these corrected the board now appears to be working a treat. Now need to order a switch to allow the port forwarding to be controlled without opening the box and get it all buttoned up.
I've gone this route because I would like two can ports so that I can break into the CAN feed to a particular module (ICE engine ECU) and definitively see what it's receiving and what it's transmitting (one port goes to the bus and one port goes to the module under test). It's a bit more intrusive as you need to cut wires but should make it much easier to see which packets are being sent by the ECU which is what I'm interested in.
I've spent all morning trying to get it to work and discovered one very nice feature but also a few gotchas that are probably worth sharing.
The above shows my test setup with the ports looped back to each other. The problem was that every time I sent a CAN message both ports would co crazy. Initially I thought that this was because the bus was unterminated so added a couple of links to enable the on board terminations. This changed matters but not in a good way. I then tried disconnecting the two ports, at this point only one port went crazy when sending a message. It turns out that:
1.) The above CAN boards do not have a link to enable the termination. What looks like a link is a second set of bus connections - I'd shorted out the bus. There is a termination on the board that is always connected. Going to have to mod the boards to change this which is a pain because they are now soldered down!
2) GVRET has a feature where a couple of processor pins are configured to enable bus forwarding, pin 11 enables bus 0 to 1 and pin 12 enables 1 to 0. This is very nice as originally I was expecting to have to write a script to do the forwarding in SavvyCAN which is a very inefficient way of doing it. The problem is that I didn't know it was there and it is enabled by default (need to you short the pins to ground to disable). When I looped the ports and sent a message it was effectively ping ponging the one message back and forth forever!
3) If you send a message to and unconnected GVRET port it doesn't get an ACK back. The board then retries sending the message forever. Not found the code controlling this yet but guessing it is in the due_can library. I'd like to either disable this or limit the number of retries so need to keep looking.
I'm sure all the above are documented somewhere but have no idea where. If anyone knows please could they let me know as there may be other nice features hidden away in there.
With all these corrected the board now appears to be working a treat. Now need to order a switch to allow the port forwarding to be controlled without opening the box and get it all buttoned up.
- Gregski
- Posts: 2348
- Joined: Tue Sep 14, 2021 10:28 am
- Location: Sacramento, California
- Has thanked: 360 times
- Been thanked: 481 times
- Contact:
Re: I think I CAN, I think I CAN...
so I wanted to let you know that I also gave the Microchip a second shot (wanted to test and see if it will work with only sending one command) this time with the Chevy Volt first gen DC-DC converter and it still did not work
"I don't need to understand how it works, I just need to understand how to make it work!" ~ EV Greg
-
- Posts: 1801
- Joined: Sun Apr 03, 2022 1:57 pm
- Has thanked: 102 times
- Been thanked: 348 times
Re: I think I CAN, I think I CAN...
That's a shame, it looks like it should be a good option.
Have you tried connecting it up to one of your other CAN dongles to see if that sheds any light on what it's doing (or not doing!)?
Have you tried connecting it up to one of your other CAN dongles to see if that sheds any light on what it's doing (or not doing!)?
- Gregski
- Posts: 2348
- Joined: Tue Sep 14, 2021 10:28 am
- Location: Sacramento, California
- Has thanked: 360 times
- Been thanked: 481 times
- Contact:
Re: I think I CAN, I think I CAN...
the darn thing itself claims to be sending CAN messages, I have a case open with Microchip Tech Support and they have replied and asked me to try it with a different device, so I emailed them back and said now I have tried it with both the Chevy Volt first gen charger and the first gen DC-DC controller, neither seem to work
"I don't need to understand how it works, I just need to understand how to make it work!" ~ EV Greg
-
- Posts: 1801
- Joined: Sun Apr 03, 2022 1:57 pm
- Has thanked: 102 times
- Been thanked: 348 times
Re: I think I CAN, I think I CAN...
I had the opposite problem, SavvyCAN claimed it wasn't sending anything but I could see 1000s of messages on the canbus lines! You need to able to see what's actually happening on the bus to diagnose it, fortunately you happen to have a few other canbus dongles!
I'd connect the Microchip one directly to one of your other dongles (with nothing else on the network). That way you can see exactly what it sends out by monitoring it using the other dongles software and also send messages back in to see if it receives them.
Might be a faulty unit?
Got to say the more I use SavvyCAN the more it impresses me, you can tell the software was written by someone who also used it.
I'd connect the Microchip one directly to one of your other dongles (with nothing else on the network). That way you can see exactly what it sends out by monitoring it using the other dongles software and also send messages back in to see if it receives them.
Might be a faulty unit?
Got to say the more I use SavvyCAN the more it impresses me, you can tell the software was written by someone who also used it.
-
- Posts: 1801
- Joined: Sun Apr 03, 2022 1:57 pm
- Has thanked: 102 times
- Been thanked: 348 times
Re: I think I CAN, I think I CAN...
Just noticed from your pictures that it's reporting an "error passive".. I think this means that it's not getting an ack back from the charger (same problem I had), is it definitely set to the right baud rate and are the can bus wires the right way round?
Edit - passive error can be caused by the Tx seeing bit errors as well as missing ack, appears to mean it can't get a message onto the bus - no real surprise there then!
Edit - passive error can be caused by the Tx seeing bit errors as well as missing ack, appears to mean it can't get a message onto the bus - no real surprise there then!
- Gregski
- Posts: 2348
- Joined: Tue Sep 14, 2021 10:28 am
- Location: Sacramento, California
- Has thanked: 360 times
- Been thanked: 481 times
- Contact:
Re: I think I CAN, I think I CAN...
I believe so because it reads messages just fine, which makes me think the high and low wires are connected properly and the termination is on, this guy has software controlled resistor terminator thingie and I tried it with both on and off, it is on in this case
"I don't need to understand how it works, I just need to understand how to make it work!" ~ EV Greg
-
- Posts: 1801
- Joined: Sun Apr 03, 2022 1:57 pm
- Has thanked: 102 times
- Been thanked: 348 times
Re: I think I CAN, I think I CAN...
If it receives messages then that suggests it is connected right, comes back to not getting an ack back from the charger. This could be the dongle isn't actually putting message on the bus or it is getting it onto the bus but the charger doesn't like something about it. In the later case connecting it to another dongle might give more information,
It's starting to sound like a duff transmit line on the unit (given that the chargers work with other dongles). Don't suppose you have a scope handy?
It's starting to sound like a duff transmit line on the unit (given that the chargers work with other dongles). Don't suppose you have a scope handy?
-
- Posts: 3295
- Joined: Mon Dec 24, 2018 1:37 pm
- Location: Slovenia
- Has thanked: 81 times
- Been thanked: 242 times
- Contact:
Re: I think I CAN, I think I CAN...
Hm... now that you mention. I did in that order:
1. I simply connected complete HV battery from Chevy Volt i had in my workshop. Do be carefull and make sure + and - lines are correct with a polarity test! viewtopic.php?f=4&t=632&p=8198&hilit=polarity#p8215
2. I connected AUX battery to output.
3. I commanded my Canalyst with ID 00 00 01 D4 msg A0 B2 at 100ms
4. I connected enable to 12V battery and i saw the light! Actually 12V light bulb got really bright .
- Gregski
- Posts: 2348
- Joined: Tue Sep 14, 2021 10:28 am
- Location: Sacramento, California
- Has thanked: 360 times
- Been thanked: 481 times
- Contact:
Re: I think I CAN, I think I CAN...
Full disclosure I am only using six 60 volt BMW 530e cell modules at this juncture [don't ask what happened to the other two] so a total of 240 volts going to the DC-DC Converter, could this be the issue?
"I don't need to understand how it works, I just need to understand how to make it work!" ~ EV Greg
-
- Posts: 3295
- Joined: Mon Dec 24, 2018 1:37 pm
- Location: Slovenia
- Has thanked: 81 times
- Been thanked: 242 times
- Contact:
Re: I think I CAN, I think I CAN...
True! I belive Volts charger and DCDC start to work from 280Vdc onwards. However i know for sure Volt inverter innards can work with 60Vdc just fine.
- Gregski
- Posts: 2348
- Joined: Tue Sep 14, 2021 10:28 am
- Location: Sacramento, California
- Has thanked: 360 times
- Been thanked: 481 times
- Contact:
Re: I think I CAN, I think I CAN...
Alright I just tried SaavyCAN with five 60 volt cell modules and a measured 289 volts and no dice, still does not work. Is there a fuse inside the DC-DC Converter? If so maybe that is blown?
"I don't need to understand how it works, I just need to understand how to make it work!" ~ EV Greg
-
- Posts: 3295
- Joined: Mon Dec 24, 2018 1:37 pm
- Location: Slovenia
- Has thanked: 81 times
- Been thanked: 242 times
- Contact:
Re: I think I CAN, I think I CAN...
There is no fuse that i know of inside. It however has PTC surge suppressors instead of precharge. But they do not protect from reverse connection!
First you determine if reverse diode inside DCDC is working. Perform polarity procedure i mentioned before:
viewtopic.php?p=8215#p8215
It works for chargers as well as for all HV device connections that have reverse diodes in their boost stage transistors.
I also found photos of its insides
https://www.gm-volt.com/threads/apm-pictures.144298/
If you open it you can inspect input stage for any damage...
- Gregski
- Posts: 2348
- Joined: Tue Sep 14, 2021 10:28 am
- Location: Sacramento, California
- Has thanked: 360 times
- Been thanked: 481 times
- Contact:
Re: I think I CAN, I think I CAN...
wow what an awesome find, thank you so mucharber333 wrote: ↑Mon May 16, 2022 7:06 amI also found photos of its insides
https://www.gm-volt.com/threads/apm-pictures.144298/
If you open it you can inspect input stage for any damage...
so here are the guts of mine, and although I don't know what I am looking for at least nothing seems smelly or black burned
"I don't need to understand how it works, I just need to understand how to make it work!" ~ EV Greg
- celeron55
- Posts: 780
- Joined: Thu Jul 04, 2019 3:04 pm
- Location: Finland
- Has thanked: 29 times
- Been thanked: 111 times
- Contact:
Re: I think I CAN, I think I CAN...
Looks pretty much the same as any automotive 300VDC->12VDC converter.
- Gregski
- Posts: 2348
- Joined: Tue Sep 14, 2021 10:28 am
- Location: Sacramento, California
- Has thanked: 360 times
- Been thanked: 481 times
- Contact:
Re: I think I CAN, I think I CAN...
I'm sorry I totally missed this the first time, so here we goarber333 wrote: ↑Sun May 15, 2022 7:26 pm1. I simply connected complete HV battery from Chevy Volt i had in my workshop. Do be carefull and make sure + and - lines are correct with a polarity test! viewtopic.php?f=4&t=632&p=8198&hilit=polarity#p8215
but, But, BUT !!! hold on a second...
"I don't need to understand how it works, I just need to understand how to make it work!" ~ EV Greg
- Gregski
- Posts: 2348
- Joined: Tue Sep 14, 2021 10:28 am
- Location: Sacramento, California
- Has thanked: 360 times
- Been thanked: 481 times
- Contact:
Re: I think I CAN, I think I CAN...
the General makes us use an Intermediate Cable to run from the Converter to the Battery Pack
"I don't need to understand how it works, I just need to understand how to make it work!" ~ EV Greg
- Gregski
- Posts: 2348
- Joined: Tue Sep 14, 2021 10:28 am
- Location: Sacramento, California
- Has thanked: 360 times
- Been thanked: 481 times
- Contact:
Re: I think I CAN, I think I CAN...
so let's unwrap this serpent
Can you see it???
Can you see it???
"I don't need to understand how it works, I just need to understand how to make it work!" ~ EV Greg
-
- Posts: 3295
- Joined: Mon Dec 24, 2018 1:37 pm
- Location: Slovenia
- Has thanked: 81 times
- Been thanked: 242 times
- Contact:
Re: I think I CAN, I think I CAN...
OH, i am sorry, when i got mine it was the unit with the cable attached. It was clearly marked with color stripes, so there was no need to inspect for polarity then.
I guess lesson No. 1 would be to assert HV polarity with all units that you are going to use. If i would be smarter then i would save myself some burnt hands as well as busted charger . I take that lesson to my DUE board too as i use a large diode to protect against polarity reversals.
- Gregski
- Posts: 2348
- Joined: Tue Sep 14, 2021 10:28 am
- Location: Sacramento, California
- Has thanked: 360 times
- Been thanked: 481 times
- Contact:
Re: I think I CAN, I think I CAN...
well bad news, even after straightening out the polarity (I take full blame for that) I had it right at the orange outlet plug but had no idea the wires crossed over later on in the OEM wiring harness, anywho that blew my inline fuse so I replaced it, and with a multimeter I confirmed I am not getting 289 volts inside the converter yet still no dice
"I don't need to understand how it works, I just need to understand how to make it work!" ~ EV Greg
- EV_Builder
- Posts: 1199
- Joined: Tue Apr 28, 2020 3:50 pm
- Location: The Netherlands
- Has thanked: 16 times
- Been thanked: 34 times
- Contact:
Re: I think I CAN, I think I CAN...
You might consider using a cheapo ali express adjustable power supply with current and voltage regulation. Now i know they often max out at 60Volts and 5Amps but 60Volts is anough to check if your polarity is correct or wrong without damaging to much or playing with uncontrolled power.
What i do is set the amps to minimum like 10ma and volts open to 60.
Now if i open up the amps and see that the amps keep riseing and i'm limited on amps then i asume i'm generating heat instead of charging a capacitor etc. I did this with the A/C unit and it even started to work (canbus) and i told me indeed i applied only 60VDC .
What i do is set the amps to minimum like 10ma and volts open to 60.
Now if i open up the amps and see that the amps keep riseing and i'm limited on amps then i asume i'm generating heat instead of charging a capacitor etc. I did this with the A/C unit and it even started to work (canbus) and i told me indeed i applied only 60VDC .
Converting an Porsche Panamera
see http://www.wdrautomatisering.nl for bespoke BMS modules.
see http://www.wdrautomatisering.nl for bespoke BMS modules.