Toyota Prius Gen 2 Inverter Controller
-
- Posts: 1801
- Joined: Sun Apr 03, 2022 1:57 pm
- Has thanked: 102 times
- Been thanked: 347 times
Re: Toyota Prius Gen 2 Inverter Controller
Can't really help a great deal with this as I don't have any experience with those boards/inverter but I do have a fair bit of experience diagnosing electronic/electrical problems.
From what you describe above I'd be looking for either a signal loading issue or an issue with system ground connections.
On the signal loading it is quite possible for you to get the results you have if either the resistance of the inverter inputs are low, or the controller outputs too high. In this case a wire feeding the inputs (very low resistance) will drive the inverter inputs but the lower drive strength (higher resistance) of the controller may not be able to. You could possibly test this with a multimeter by measuring both the AC and DC voltage of the PWM output/driver lines in isolation and then when connected to the inverter to see if there is a significant change due to loading? A scope would be better here though.
With the grounds if there is not a good connection between the two systems it's possible for the controller output to not meet the high or low level input thresholds of the inverter (for example if for some reason the inverter ground is a couple of volts higher than the controller it will never see a low level input). If you haven't already tested the ground continuity it would be well worth doing.
What you really need is an oscilloscope, can you borrow one from anywhere? A battery powered scope is the ideal as it doesn't introduce an extra mains earth reference into the system. The ability to see what the signals are doing on the real system and to follow the signal through the drive stages and inverter makes everything so much more straightforward. I wouldn't recommend doing it on a live high voltage system but as long as you are powering the inverter from a low voltage test supply it should be fine. A soundcard can do some things but the limited input range, low sampling rate and the AC coupling make diagnosis of a lot of issues much more difficult.
From what you describe above I'd be looking for either a signal loading issue or an issue with system ground connections.
On the signal loading it is quite possible for you to get the results you have if either the resistance of the inverter inputs are low, or the controller outputs too high. In this case a wire feeding the inputs (very low resistance) will drive the inverter inputs but the lower drive strength (higher resistance) of the controller may not be able to. You could possibly test this with a multimeter by measuring both the AC and DC voltage of the PWM output/driver lines in isolation and then when connected to the inverter to see if there is a significant change due to loading? A scope would be better here though.
With the grounds if there is not a good connection between the two systems it's possible for the controller output to not meet the high or low level input thresholds of the inverter (for example if for some reason the inverter ground is a couple of volts higher than the controller it will never see a low level input). If you haven't already tested the ground continuity it would be well worth doing.
What you really need is an oscilloscope, can you borrow one from anywhere? A battery powered scope is the ideal as it doesn't introduce an extra mains earth reference into the system. The ability to see what the signals are doing on the real system and to follow the signal through the drive stages and inverter makes everything so much more straightforward. I wouldn't recommend doing it on a live high voltage system but as long as you are powering the inverter from a low voltage test supply it should be fine. A soundcard can do some things but the limited input range, low sampling rate and the AC coupling make diagnosis of a lot of issues much more difficult.
-
- Posts: 898
- Joined: Fri Apr 26, 2019 5:40 pm
- Has thanked: 291 times
- Been thanked: 177 times
Re: Toyota Prius Gen 2 Inverter Controller
I have an old Tectronics tube scope packed away in a box... somewhere. I might just buy a cheap portable scope rather than climb through everything I packed away "for later".
The last time I used a scope, I was unaware it wasn't isolated, and I wanted to look at the waveform from the PFC circuit in a light dimmer. It didn't end well. I've been too chicken to use it ever since, knowing that I don't have the right context of what is safe or not safe to test with them.
I'll maybe reassemble everything (I now have a proper wiring harness) first, if you think it could be signals dropping below threshold, maybe the duponts are giving inconsistent results. Real wires shoved into terminals could maybe help.
Also, I'm feeding the board 11.3v instead of 12v, because it's dropping through a diode. I guess I could eliminate that as a potential point of confusion.
-
- Posts: 1801
- Joined: Sun Apr 03, 2022 1:57 pm
- Has thanked: 102 times
- Been thanked: 347 times
Re: Toyota Prius Gen 2 Inverter Controller
Quite like the old tube scopes, they're what I leant on. Early digital scopes (and current cheap scopes) have trouble with aliasing and you had to be careful setting them up to avoid misleading results, the tube scopes just show what's there. Mains powered scopes are fine as long as the circuit you are testing is fully floating (i.e. has no earth reference of its own - battery powered or fully isolated power supply). I've just bought a cheapish battery powered scope myself to avoid the isolation issues, it looks promising but not used it in anger yet so the jury is out on how it performs.
TBH you could do with figuring out the right context of safe before working on a HV DC system! I've had a few incidents with mains voltage and ICE ignition systems which have taught me to treat them with respect. The trouble is these are relatively safe compared to HV DC . If you're not confident it's safe don't do it. Sorry don't mean to preach, and you have much more experience in the EV field than me, but working around HV DC scares me.
Just had my first look at the blue pill schematic and there is not much on it to go wrong. Sure you have already seen it but just in case I did come across this https://openinverter.org/forum/viewtopi ... 2767#p2767. Are you sure the transistors are in right, correct part, not blown as that could explain most of what you are seeing?
Can't see the 11.3V being an issue. The leads might not be helping depending on what kind of quality they are.
I'd get it connected up properly on a LV test supply to the inverter and follow the PWM signals though using a scope. It should be pretty clear where it stops working and if it isn't posting scope pictures here would make it a lot easier for others to help diagnose.
TBH you could do with figuring out the right context of safe before working on a HV DC system! I've had a few incidents with mains voltage and ICE ignition systems which have taught me to treat them with respect. The trouble is these are relatively safe compared to HV DC . If you're not confident it's safe don't do it. Sorry don't mean to preach, and you have much more experience in the EV field than me, but working around HV DC scares me.
Just had my first look at the blue pill schematic and there is not much on it to go wrong. Sure you have already seen it but just in case I did come across this https://openinverter.org/forum/viewtopi ... 2767#p2767. Are you sure the transistors are in right, correct part, not blown as that could explain most of what you are seeing?
Can't see the 11.3V being an issue. The leads might not be helping depending on what kind of quality they are.
I'd get it connected up properly on a LV test supply to the inverter and follow the PWM signals though using a scope. It should be pretty clear where it stops working and if it isn't posting scope pictures here would make it a lot easier for others to help diagnose.
-
- Posts: 1801
- Joined: Sun Apr 03, 2022 1:57 pm
- Has thanked: 102 times
- Been thanked: 347 times
Re: Toyota Prius Gen 2 Inverter Controller
Just come across your later post on this issue so it has been corrected. Still think it's worth checking the transistors out with a multimeter on diode test. You should read 0.6V with red lead on the base and black on either of the other two and open circuit for all other combinations (base is connected to the resistor on the blue pill board, emitter to ground and collector to the output pin).Pete9008 wrote: ↑Fri Jun 03, 2022 12:57 pm Just had my first look at the blue pill schematic and there is not much on it to go wrong. Sure you have already seen it but just in case I did come across this https://openinverter.org/forum/viewtopi ... 2767#p2767. Are you sure the transistors are in right, correct part, not blown as that could explain most of what you are seeing?
-
- Posts: 1801
- Joined: Sun Apr 03, 2022 1:57 pm
- Has thanked: 102 times
- Been thanked: 347 times
Re: Toyota Prius Gen 2 Inverter Controller
Just come across the picture of your board here https://openinverter.org/forum/viewtopi ... 284#p39284
Not quite clear but there is something around the legs of T1, what is it?
Also can't make out whether the base resistors colour codes are brown, black, orange (10k) or brown, black, red (1k). Could you either check the resistance with a meter or post a clearer picture of that area of the board? They should be 1k, if they are 10k it might not drive the inverter inputs hard enough.
Not quite clear but there is something around the legs of T1, what is it?
Also can't make out whether the base resistors colour codes are brown, black, orange (10k) or brown, black, red (1k). Could you either check the resistance with a meter or post a clearer picture of that area of the board? They should be 1k, if they are 10k it might not drive the inverter inputs hard enough.
-
- Posts: 898
- Joined: Fri Apr 26, 2019 5:40 pm
- Has thanked: 291 times
- Been thanked: 177 times
Re: Toyota Prius Gen 2 Inverter Controller
Bumper repair epoxy.
So, there's an error on the silkscreen. There was no note with the board when I bought it, so I soldered it up wrong. Then, trying to flip the transistors, I broke one leg off at the case. So I soldered it back on, but I was worried about its frailty, so I gobbed the only epoxy I had on hand onto it to encapsulate it. The other transistors looks weird because, rather than risk lifting a pad, I cut the legs mid way before I flipped the transistors and just soldered them.
If the most questionable transistor didn't work, I would still be getting *something* on the other two motor phases, but I'm getting nothing.
They're 1k. Double-checked. So that's good. Also, IIRC I measured them a few weeks ago and they were 960-980Ω.Also can't make out whether the base resistors colour codes are brown, black, orange (10k) or brown, black, red (1k). Could you either check the resistance with a meter or post a clearer picture of that area of the board? They should be 1k, if they are 10k it might not drive the inverter inputs hard enough.
Looks good to me. One DMM measured ~.809-.830v. Usually pretty close to 0.820v. Did all 6 combinations, no other directions had a drop. Probably just out of calibration. Tried it on another DMM and got ~0.690v drop instead.Still think it's worth checking the transistors out with a multimeter on diode test. You should read 0.6V with red lead on the base and black on either of the other two and open circuit for all other combinations
Transistor 1:
B-E: .820
B-C: .820
E-C: 0
E-B: 0
C-B: 0
C-E: 0
Transistor 2:
B-E: .820
B-C: .820
E-C: 0
E-B: 0
C-B: 0
C-E: 0
Transistor 3:
B-E: .820
B-C: .820
E-C: 0
E-B: 0
C-B: 0
C-E: 0
Pretty sure I'm getting output from the Blue Pill board. I even tested the back side of the inverter harness, to make sure there's the right voltage on the PWM lines. Voltage alone hides the waveform, but, I would be getting *something*, some kind of power draw on the motor if it was getting through. It might not spin properly, but, I just getting a no power, floating voltage on the inverter output lines.I'd get it connected up properly on a LV test supply to the inverter and follow the PWM signals though using a scope.
After the harness the signal disappears into the inverter board itself and, I've done no tests there.
But, aside from obviously blown components (like the power transistors, which appear fine), there's a lot more going on there than there is on the blue pill. I dunno that it's even reasonable to troubleshoot Toyota stuff beyond "Is my inverter bad?" yes/no type of tests. If the inverter's bad, then it's a matter of going and getting my money back or a replacement.
Thanks for the help so far, I'm going to try reassembling the inverter and re-wiring up the blue pill a bit more directly, cut down on the questionable jumpers next.
-
- Posts: 1801
- Joined: Sun Apr 03, 2022 1:57 pm
- Has thanked: 102 times
- Been thanked: 347 times
Re: Toyota Prius Gen 2 Inverter Controller
That all sounds good.
I'm still very suspicious about those transistors. A wrong part, or one the wrong way round, might act like a diode rather than a transistor. A low hfe (low current gain) part might not pull fully to ground. Both of these could give a signal that doesn't pull low enough to wake the inverter. Are they definitely the right part, they haven't been swapped for an alternative due to availability issues or anything like that?
It doesn't sound like a fault in the inverter as you have seen it working but I suppose it is possible.
The other test to try is the base to collector transistor test with the multimeter but measured from the base resistor all the way through to the inverter board. That way it is also testing the wiring between the two boards.
I'm still very suspicious about those transistors. A wrong part, or one the wrong way round, might act like a diode rather than a transistor. A low hfe (low current gain) part might not pull fully to ground. Both of these could give a signal that doesn't pull low enough to wake the inverter. Are they definitely the right part, they haven't been swapped for an alternative due to availability issues or anything like that?
It doesn't sound like a fault in the inverter as you have seen it working but I suppose it is possible.
The other test to try is the base to collector transistor test with the multimeter but measured from the base resistor all the way through to the inverter board. That way it is also testing the wiring between the two boards.
-
- Posts: 898
- Joined: Fri Apr 26, 2019 5:40 pm
- Has thanked: 291 times
- Been thanked: 177 times
Re: Toyota Prius Gen 2 Inverter Controller
So am I and they're on my list to replace "next order". But I haven't ordered electronics parts in 2+ years and shipping is killer. On my next order I want to have the stuff that makes the current sensors work, which, I think I can copy from Johannes' more recent design. I have bins of scrap parts, but, none I've identified as opamps, or, that I'd trust to work in a vehicle after harvesting them. Since I don't really know what to order yet for the current sensors, I've been putting off replacing the transistors.
Nope, 2.5 years ago there weren't part shortages :p. They're the exact part ordered off of Digikey. Someone here even helped me pick the correct ones.Are they definitely the right part, they haven't been swapped for an alternative due to availability issues or anything like that?
Ehn, I've seen the power transistors default connect to HVDC+, and then toggle to HVDC- when a PWM line is pulled high, after a wakeup. I dunno that I qualify that as "seen it working", but, it seems like if I fed it PWM it should.It doesn't sound like a fault in the inverter as you have seen it working but I suppose it is possible.
I'm just about dead certain it's not the wiring either.The other test to try is the base to collector transistor test with the multimeter but measured from the base resistor all the way through to the inverter board. That way it is also testing the wiring between the two boards.
What it sounds like is a busted MSDN connection, but, I've quintuple-checked that every possible way there is to check it or wire it differently.
-
- Posts: 1801
- Joined: Sun Apr 03, 2022 1:57 pm
- Has thanked: 102 times
- Been thanked: 347 times
Re: Toyota Prius Gen 2 Inverter Controller
What's a MSDN connection? All I can think of is Microsoft Developers Network and I'm sure it's not that!MattsAwesomeStuff wrote: ↑Sat Jun 04, 2022 9:19 am What it sounds like is a busted MSDN connection, but, I've quintuple-checked that every possible way there is to check it or wire it differently.
-
- Posts: 898
- Joined: Fri Apr 26, 2019 5:40 pm
- Has thanked: 291 times
- Been thanked: 177 times
Re: Toyota Prius Gen 2 Inverter Controller
Heh.
"Motor ShutDowN".
If it's not connected to 12+, your MG2 inverter doesn't work. (Anything with the prefix "G" [i.e. GSDN] tends to be MG1, the prefix "M" tends to be MG2, they're both motors but one is sometimes called "generator" and the other "motor" in Toyota literature).
There's also a MFIV, a fault signal which you connect to pin 27. Not sure what it does or how it works or what it expects.
It's on the wiki, which, I have spent the last couple days gutting and adding new content to.
https://openinverter.org/wiki/Toyota_Pr ... 2_Inverter
Speaking of which, new post...
-
- Posts: 898
- Joined: Fri Apr 26, 2019 5:40 pm
- Has thanked: 291 times
- Been thanked: 177 times
Re: Toyota Prius Gen 2 Inverter Controller
I'm probably the last person in the world to ever read or use this info, but, I got a "new" Gen2 inverter from the junkyard, and wiring harness, etc, to finally help determine whether it's the inverter or whether it's my controller board. And meanwhile, because I can't make progress on my own, I just documented the hell out of everything.
... I spent a couple days giving the Prius Gen 2 wiki a complete overhaul:
- What used to be the main page with a smattering of general description and specific projects and development notes all mashed together, is now (I've decided) deprecated in favor of 3 new pages:
https://openinverter.org/wiki/Toyota_Prius_Gen2_Board -- Old and busted, "board" doesn't mean anything.
https://openinverter.org/wiki/Toyota_Pr ... 2_Inverter -- New "main" page, with all the inverer-specific knowledge that then only links the 2 following project pages
https://openinverter.org/wiki/Toyota_Pr ... hole_Board -- Damien's old board (the one I'm building in this thread), and all its notes, and,
https://openinverter.org/wiki/Toyota_Pr ... Controller -- Johannes' new all-in-one kit.
- Since I hate when years later I stumble across an open source project and have no idea how active or useful it is or what state of functionality it was abandoned in (I still have to explain to people that the Open ReVolt project hasn't been a thing for 10 years, nor does its successor exist, and, how did they even come across this, and this isn't the cheap solution anymore...), I wrote up a history section for this through-hole board. Just so people understand the context of how it came about, and why it's no longer used. I think it's correct, but it's editorialized so maybe I got some stuff wrong.
- I explained Johannes' project on its own page, and linked the two product pages where to buy it.
- The main Gen 2 Inverter page now describes and links to each of the projects and tells people straight up what to use and not use. No more broken record of me telling people they're mixing up projects.
- The Gen 2 wiring harness (the car side that goes into the inverter), soon turns into 5 cables and some loose wires, and wire colors are re-used often. So I traced out all the wires in the harness and labelled which cable/bundle they belong to. I imagine many people have done this before, and no one ever wrote it down so we all repeat the same work. No more of that.
- I added 7 new pictures to the wiki, showing closeup labelled photos of every wire color and pinout, including the i9. I know most people from here out will probably just buy Johannes' kit and toss the wiring harness, but for anyone who would ever put a little time into reverse engineering or contributing to his development efforts, or for anyone opting to buy the board-only kit from Johannes and not the whole enclosure and connectors, hopefully this saves them some time and reduces chances of error. People can see what is being referred to rather than read text about it and risk getting it wrong.
![Image](https://openinverter.org/wiki/images/f/fb/Prius_Inverter_-_Pin_Numbering_2.jpg)
![Image](https://openinverter.org/wiki/images/e/e9/Prius_Inverter_Wire_Colors_5.jpg)
![Image](https://openinverter.org/wiki/images/e/e5/Prius_Inverter_Wire_Colors_4.jpg)
![Image](https://openinverter.org/wiki/images/6/61/Prius_Inverter_Wire_Colors_6.jpg)
![Image](https://openinverter.org/wiki/images/5/54/Prius_Inverter_Wire_Colors_7.jpg)
![Image](https://openinverter.org/wiki/images/5/52/Prius_Inverter_Wire_Colors_8.jpg)
![Image](https://openinverter.org/wiki/images/a/a8/Prius_Inverter_Wire_Colors_9.jpg)
... oh and I tried doing a quick test-wiring of the new inverter, but, didn't get it to work, though, I only put in a couple minutes and hardly even checked settings.
... I spent a couple days giving the Prius Gen 2 wiki a complete overhaul:
- What used to be the main page with a smattering of general description and specific projects and development notes all mashed together, is now (I've decided) deprecated in favor of 3 new pages:
https://openinverter.org/wiki/Toyota_Prius_Gen2_Board -- Old and busted, "board" doesn't mean anything.
https://openinverter.org/wiki/Toyota_Pr ... 2_Inverter -- New "main" page, with all the inverer-specific knowledge that then only links the 2 following project pages
https://openinverter.org/wiki/Toyota_Pr ... hole_Board -- Damien's old board (the one I'm building in this thread), and all its notes, and,
https://openinverter.org/wiki/Toyota_Pr ... Controller -- Johannes' new all-in-one kit.
- Since I hate when years later I stumble across an open source project and have no idea how active or useful it is or what state of functionality it was abandoned in (I still have to explain to people that the Open ReVolt project hasn't been a thing for 10 years, nor does its successor exist, and, how did they even come across this, and this isn't the cheap solution anymore...), I wrote up a history section for this through-hole board. Just so people understand the context of how it came about, and why it's no longer used. I think it's correct, but it's editorialized so maybe I got some stuff wrong.
- I explained Johannes' project on its own page, and linked the two product pages where to buy it.
- The main Gen 2 Inverter page now describes and links to each of the projects and tells people straight up what to use and not use. No more broken record of me telling people they're mixing up projects.
- The Gen 2 wiring harness (the car side that goes into the inverter), soon turns into 5 cables and some loose wires, and wire colors are re-used often. So I traced out all the wires in the harness and labelled which cable/bundle they belong to. I imagine many people have done this before, and no one ever wrote it down so we all repeat the same work. No more of that.
- I added 7 new pictures to the wiki, showing closeup labelled photos of every wire color and pinout, including the i9. I know most people from here out will probably just buy Johannes' kit and toss the wiring harness, but for anyone who would ever put a little time into reverse engineering or contributing to his development efforts, or for anyone opting to buy the board-only kit from Johannes and not the whole enclosure and connectors, hopefully this saves them some time and reduces chances of error. People can see what is being referred to rather than read text about it and risk getting it wrong.
![Image](https://openinverter.org/wiki/images/f/fb/Prius_Inverter_-_Pin_Numbering_2.jpg)
![Image](https://openinverter.org/wiki/images/e/e9/Prius_Inverter_Wire_Colors_5.jpg)
![Image](https://openinverter.org/wiki/images/e/e5/Prius_Inverter_Wire_Colors_4.jpg)
![Image](https://openinverter.org/wiki/images/6/61/Prius_Inverter_Wire_Colors_6.jpg)
![Image](https://openinverter.org/wiki/images/5/54/Prius_Inverter_Wire_Colors_7.jpg)
![Image](https://openinverter.org/wiki/images/5/52/Prius_Inverter_Wire_Colors_8.jpg)
![Image](https://openinverter.org/wiki/images/a/a8/Prius_Inverter_Wire_Colors_9.jpg)
... oh and I tried doing a quick test-wiring of the new inverter, but, didn't get it to work, though, I only put in a couple minutes and hardly even checked settings.
-
- Posts: 898
- Joined: Fri Apr 26, 2019 5:40 pm
- Has thanked: 291 times
- Been thanked: 177 times
Re: Toyota Prius Gen 2 Inverter Controller
Well, I gave it a month's rest and to my surprise the problem didn't just fix itself. So I tried tackling it again.
Lemme go through my symptoms/actions again, just to be sure I'm not getting what I'm supposed to be getting.
- Got a new inverter. Got lots of wiring harness. Plugged the wiring harness directly into the terminals on the board, no more Dupont connectors.
- Following Johannes' procedure from this post: https://openinverter.org/forum/viewtopi ... 248#p39248 . Boost 0. Fweak 100. Fslpspnt 1. Ampnom 1 (or 100, doesn't matter).
- I'm supposed to get ~1.65vdc PWM output from the blue pill's phase lines. I do.
- I'm supposed to get ~7.5vdc output on the board's output phase lines (that go to pins 9, 10, 11, the PWM inputs to the inverter). I do.
- I've got 12v system power, 12v "HV" bus power, the 3 phase lines, MSDN to 12+, MFIV to the board's "FLT", and VH to UDC. Board is set to "forward" and I have a throttle plugged in that we're presumably ignoring. That's all.
- I think I'm supposed to get some voltage across any of the MG2 motor output phase lines from this too? I don't. Zero volts.
Does this (lack of voltage to a motor) already indicate failure or was procedure only to verify the control board PWM output and it's not expected to make it to the motor output?
If it does indicate failure... I suppose I could go to the procedure testing the output transistors on the inverter for failure. I can't imagine they'd be busted though.
We have recently learned about this "wakeup" pulse each phase needs before it starts conducting. Since the PWM is giving the inverter voltage, does that take care of that wakeup? If not, can I manually force a wakeup signal on each phase? If so, can I do this while the control board is operating and feeding PWM?
Not sure what the next step would be, or what the discovery of the "wakeup" might imply for testing or failure modes.
Pete said above that he suspected if grounds weren't common it could have this behavior. I don't think the HVDC ground is supposed to be common with the 12v system ground (in i9). But there is also the case ground to inverter. I briefly tried getting the DCDC inverter working, which only needs 12v to Pin1, and it's supposed to turn HVDC into 13.8v across the case ground and a big 12V wire, but I got nothing when I tried that (cranked my HVDC to 125v, close to the DCDC input's minimum).
Could my problem be that the inverter case doesn't share a ground with System 12v's ground? (I tested between 12+ and the metal on the case... got no voltage. Tested continuity between system ground and case ground, found none). Or are those kept separate on purpose?
Just spitballing here.
Lemme go through my symptoms/actions again, just to be sure I'm not getting what I'm supposed to be getting.
- Got a new inverter. Got lots of wiring harness. Plugged the wiring harness directly into the terminals on the board, no more Dupont connectors.
- Following Johannes' procedure from this post: https://openinverter.org/forum/viewtopi ... 248#p39248 . Boost 0. Fweak 100. Fslpspnt 1. Ampnom 1 (or 100, doesn't matter).
- I'm supposed to get ~1.65vdc PWM output from the blue pill's phase lines. I do.
- I'm supposed to get ~7.5vdc output on the board's output phase lines (that go to pins 9, 10, 11, the PWM inputs to the inverter). I do.
- I've got 12v system power, 12v "HV" bus power, the 3 phase lines, MSDN to 12+, MFIV to the board's "FLT", and VH to UDC. Board is set to "forward" and I have a throttle plugged in that we're presumably ignoring. That's all.
- I think I'm supposed to get some voltage across any of the MG2 motor output phase lines from this too? I don't. Zero volts.
Does this (lack of voltage to a motor) already indicate failure or was procedure only to verify the control board PWM output and it's not expected to make it to the motor output?
If it does indicate failure... I suppose I could go to the procedure testing the output transistors on the inverter for failure. I can't imagine they'd be busted though.
We have recently learned about this "wakeup" pulse each phase needs before it starts conducting. Since the PWM is giving the inverter voltage, does that take care of that wakeup? If not, can I manually force a wakeup signal on each phase? If so, can I do this while the control board is operating and feeding PWM?
Not sure what the next step would be, or what the discovery of the "wakeup" might imply for testing or failure modes.
Pete said above that he suspected if grounds weren't common it could have this behavior. I don't think the HVDC ground is supposed to be common with the 12v system ground (in i9). But there is also the case ground to inverter. I briefly tried getting the DCDC inverter working, which only needs 12v to Pin1, and it's supposed to turn HVDC into 13.8v across the case ground and a big 12V wire, but I got nothing when I tried that (cranked my HVDC to 125v, close to the DCDC input's minimum).
Could my problem be that the inverter case doesn't share a ground with System 12v's ground? (I tested between 12+ and the metal on the case... got no voltage. Tested continuity between system ground and case ground, found none). Or are those kept separate on purpose?
Just spitballing here.
-
- Posts: 1801
- Joined: Sun Apr 03, 2022 1:57 pm
- Has thanked: 102 times
- Been thanked: 347 times
Re: Toyota Prius Gen 2 Inverter Controller
First, apologies if I pitch the level of this wrong. Based on your earlier posts I'm assuming that you have a good understanding of electrical systems but electronics is still pretty new to you. If you need more. or less. detail please say.
First I think you really need to get a scope on this. The voltages you have measured are great and would normally give good confidence that everything is OK but it obviously isn't! A scope would immediately tell you whether the signal going into the inverter is good allowing you to rule that side of things out. TBH I can't imagine diagnosing any electronics without a scope, I might try occasionally if I'm too lazy to go and get one but it would be with the expectation of having to in the end. There's a bit of a learning curve but once you get the hang of using one you will wonder how you ever managed without.
For me a good analogy is trying to diagnose a charging problem on a car with nothing but a test light, you might get lucky but you will largely be guessing. The test light will give you a very rough indication of what is going on but will be missing a lot of detail. A multimeter on the other hand will immediately tell you whether you have alternator output, what the regulator is doing, how the battery is responding, etc. The extra information makes the diagnosis so much easier.
The same is true with a multimeter, as soon as you introduce electronics into the system a it becomes a fairly limited piece of test equipment and a scope provides a much better view of what is actually going on.
Next I think it's worth spending a little time on grounds. Grounds are a pretty simple concept but like most simple things can have a lot of hidden complexity. For a start there are too many terms used interchangeably - ground, earth, 0V, etc which can all be the same or different!. For the following I'm going to start using the term 0V instead as this is probably clearer in regard to signal levels on control lines. If two circuits have different 0V references it should be pretty clear that they can't share signals.
Every electronic device needs a 0V reference, without it any measurement of voltage is meaningless (it's like saying the top of a building is at 100m without referencing where you are measuring from (sea level, ground level, foundation level etc)). You can also have multiple 0V references in a system. For example on a EV you have a HV-0V and a 12V-0V (plus a load of others hidden away in the electronics!). These should be isolated from each other. Isolation means that if you put a multimeter measuring voltage between the two 0Vs you could measure pretty much anything - it depends on leakage currents, what systems are active, etc - A voltage measurement between isolated systems is usually meaningless (if you need to measure isolation then measure resistance not voltage).
If two electronic circuits are connected they must share a reference (this doesn't have to be 0V but normally is). Without this a 1V output by one circuit may not be seen as a 1V input by the other, the signals only make sense if they share a common 0V.
In you case HV 0V should be isolated.
You need to make sure that there is a good connection between 0V on the blue pill to 0V on the motherboard and 0V on the motherboard to 0V on the inverter. Regarding the inverter the documentation for Damien's board on the Wiki uses GND for what I'm calling 0V. There is the comment that all GND's are common and should be connected to the black wire on the 2-pin connector next to the 32-pin connector. I'm guessing that this is the connection to the inverter power supplies and all electronics tied to 12V-0V
Also on the inverter connector there appear to be two connections GINV and GCNV. The inverter will contain isolation circuitry to keep the 12V-0V and HV-0V separate, I'm guessing that these are the 0V references for those sections in the inverter and converter respectively. I believe that these will also need connecting to the motherboard/bluepill 0V - can anyone using the Gen2 confirm this?
One other way to confirm which 0V is relevant is to use a multimeter on resistance between a signal line and what you suspect is the relevant 0V - if it's the wrong 0V you will see and open circuit, if it's the right one you should measure something (make sure you are not touching the probes while doing this as there is enough resistance in the human body to throw these measurement off).
Once all 0Vs are all connected together (and continuity checked with a multimeter) you can start looking at signal levels.
When using a scope you need to be conscious that it also has a 0V reference too. On a scope this is the shell of the BNC connectors on the front of the scope and this needs connecting to the 0V reference of the circuit under test (99% of the time just using the clips on the scope probes is enough to do this). The scope 0V also tends to be connected to mains earth on most scopes/test equipment. This shouldn't be an issue unless you have another earth reference connected to your system - if you do then connecting the scope will short things out (this is probably why you had a negative experience with a scope and a light dimmer circuit in the past!).
The point of all this is that the scope 0V (or multimeter) also needs connecting to the correct 0V reference for the measurement you are making. If you connect it to the bluepill 0V, and see the signal, that just means that the bluepill output is OK. If you leave the 0V connection on the bluepill 0V and move the scope probe to the inverter input it means that the bluepill output, and the wiring, is OK. To check whether the signal at the inverter input is OK you need to move the measurement 0V connection to the inverter input 0V reference and move the probe to the inverter input. If you still see the signal then the inverter input is OK too. If your 0V connections are OK these measurements should all give the same answer but if they aren't you could see differences and those differences can be the difference between working and not working. This may be where the confusion is coming in, the inverter appears to have multiple 0V references which may or may not be connected.
I suggest that you check all the 0V connections. If that doesn't yield anything dig out the scope and have a look at the signals on the bluepill output (referenced to the bluepill 0V), the motherboard output (referenced to the motherboard 0V) and the inverter/converter input (referenced to GINV/GCNV).
Again, apologies if I've pitched this at completely the wrong level!
First I think you really need to get a scope on this. The voltages you have measured are great and would normally give good confidence that everything is OK but it obviously isn't! A scope would immediately tell you whether the signal going into the inverter is good allowing you to rule that side of things out. TBH I can't imagine diagnosing any electronics without a scope, I might try occasionally if I'm too lazy to go and get one but it would be with the expectation of having to in the end. There's a bit of a learning curve but once you get the hang of using one you will wonder how you ever managed without.
For me a good analogy is trying to diagnose a charging problem on a car with nothing but a test light, you might get lucky but you will largely be guessing. The test light will give you a very rough indication of what is going on but will be missing a lot of detail. A multimeter on the other hand will immediately tell you whether you have alternator output, what the regulator is doing, how the battery is responding, etc. The extra information makes the diagnosis so much easier.
The same is true with a multimeter, as soon as you introduce electronics into the system a it becomes a fairly limited piece of test equipment and a scope provides a much better view of what is actually going on.
Next I think it's worth spending a little time on grounds. Grounds are a pretty simple concept but like most simple things can have a lot of hidden complexity. For a start there are too many terms used interchangeably - ground, earth, 0V, etc which can all be the same or different!. For the following I'm going to start using the term 0V instead as this is probably clearer in regard to signal levels on control lines. If two circuits have different 0V references it should be pretty clear that they can't share signals.
Every electronic device needs a 0V reference, without it any measurement of voltage is meaningless (it's like saying the top of a building is at 100m without referencing where you are measuring from (sea level, ground level, foundation level etc)). You can also have multiple 0V references in a system. For example on a EV you have a HV-0V and a 12V-0V (plus a load of others hidden away in the electronics!). These should be isolated from each other. Isolation means that if you put a multimeter measuring voltage between the two 0Vs you could measure pretty much anything - it depends on leakage currents, what systems are active, etc - A voltage measurement between isolated systems is usually meaningless (if you need to measure isolation then measure resistance not voltage).
If two electronic circuits are connected they must share a reference (this doesn't have to be 0V but normally is). Without this a 1V output by one circuit may not be seen as a 1V input by the other, the signals only make sense if they share a common 0V.
In you case HV 0V should be isolated.
You need to make sure that there is a good connection between 0V on the blue pill to 0V on the motherboard and 0V on the motherboard to 0V on the inverter. Regarding the inverter the documentation for Damien's board on the Wiki uses GND for what I'm calling 0V. There is the comment that all GND's are common and should be connected to the black wire on the 2-pin connector next to the 32-pin connector. I'm guessing that this is the connection to the inverter power supplies and all electronics tied to 12V-0V
Also on the inverter connector there appear to be two connections GINV and GCNV. The inverter will contain isolation circuitry to keep the 12V-0V and HV-0V separate, I'm guessing that these are the 0V references for those sections in the inverter and converter respectively. I believe that these will also need connecting to the motherboard/bluepill 0V - can anyone using the Gen2 confirm this?
One other way to confirm which 0V is relevant is to use a multimeter on resistance between a signal line and what you suspect is the relevant 0V - if it's the wrong 0V you will see and open circuit, if it's the right one you should measure something (make sure you are not touching the probes while doing this as there is enough resistance in the human body to throw these measurement off).
Once all 0Vs are all connected together (and continuity checked with a multimeter) you can start looking at signal levels.
When using a scope you need to be conscious that it also has a 0V reference too. On a scope this is the shell of the BNC connectors on the front of the scope and this needs connecting to the 0V reference of the circuit under test (99% of the time just using the clips on the scope probes is enough to do this). The scope 0V also tends to be connected to mains earth on most scopes/test equipment. This shouldn't be an issue unless you have another earth reference connected to your system - if you do then connecting the scope will short things out (this is probably why you had a negative experience with a scope and a light dimmer circuit in the past!).
The point of all this is that the scope 0V (or multimeter) also needs connecting to the correct 0V reference for the measurement you are making. If you connect it to the bluepill 0V, and see the signal, that just means that the bluepill output is OK. If you leave the 0V connection on the bluepill 0V and move the scope probe to the inverter input it means that the bluepill output, and the wiring, is OK. To check whether the signal at the inverter input is OK you need to move the measurement 0V connection to the inverter input 0V reference and move the probe to the inverter input. If you still see the signal then the inverter input is OK too. If your 0V connections are OK these measurements should all give the same answer but if they aren't you could see differences and those differences can be the difference between working and not working. This may be where the confusion is coming in, the inverter appears to have multiple 0V references which may or may not be connected.
I suggest that you check all the 0V connections. If that doesn't yield anything dig out the scope and have a look at the signals on the bluepill output (referenced to the bluepill 0V), the motherboard output (referenced to the motherboard 0V) and the inverter/converter input (referenced to GINV/GCNV).
Again, apologies if I've pitched this at completely the wrong level!
Re: Toyota Prius Gen 2 Inverter Controller
Ok as for the inverter case doesn’t share system ground… it does it is 12v ground for the dc dc convertor, it’s not hv grd all high voltages are fully isolated.
As for the inverter it really is super simple, as long as the enable lines are tied to +12v and the inverter ground lines are attached to the same gnd as your control board then the igbts are trigged by pulling the phase inputs that output (15v from memory) down to system ground, this is done on the control board by a fet or transitor, but can be easily tested without a control board by touching phase input to inverter gnd (not the case the gnd line at the inverter multiplug)
I wouldn’t expect to see any voltage coming out of the control board on the phase pwms is they are basically a switched gnd
As for the inverter it really is super simple, as long as the enable lines are tied to +12v and the inverter ground lines are attached to the same gnd as your control board then the igbts are trigged by pulling the phase inputs that output (15v from memory) down to system ground, this is done on the control board by a fet or transitor, but can be easily tested without a control board by touching phase input to inverter gnd (not the case the gnd line at the inverter multiplug)
I wouldn’t expect to see any voltage coming out of the control board on the phase pwms is they are basically a switched gnd
Re: Toyota Prius Gen 2 Inverter Controller
I genuinely can’t find any of the pages for either of the gen2 controller boards in the wiki today, only the general inverter info page, it was always a bit tricky to find but today nothing, I know you split the pages up Matt perhaps the inverter page could have links to the other pages at the top?
Re: Toyota Prius Gen 2 Inverter Controller
Just had a quick skim through the start of this thread, early on Damien mentioned the silk screen for the 3 phase transistors is backwards on the board, if these aren’t correct this will cause you problems. Looks like they need to be backwards to the the silkscreen so flat side of transistors should face out to the edge of the board not inwards
-
- Posts: 898
- Joined: Fri Apr 26, 2019 5:40 pm
- Has thanked: 291 times
- Been thanked: 177 times
Re: Toyota Prius Gen 2 Inverter Controller
Hmm, grammar typo in there I can't parse.
We have the System ground (i9).
We have the HV ground.
We have the inverter case ground.
We have the DC/DC ground.
HV being isolated makes sense.
DC/DC says on the Wiki that it uses the inverter case as the ground (has only a big 12v+ terminal).
So... are System and Case grounds also the same?
Or, is there any reason I would need the DC/DC converter output/case connected to a car battery's + and -?
... they are.I genuinely can’t find any of the pages for either of the gen2 controller boards in the wiki today, only the general inverter info page, it was always a bit tricky to find but today nothing, I know you split the pages up Matt perhaps the inverter page could have links to the other pages at the top?
https://openinverter.org/wiki/Toyota_Pr ... ontrollers
Right there, first heading, "replacement controllers".
Yep, caught that a couple years ago. They're flipped. And I've tested the legs on each. And I've tested the outputs.Looks like they need to be backwards to the the silkscreen so flat side of transistors should face out to the edge of the board not inwards
Re: Toyota Prius Gen 2 Inverter Controller
Ok that’s me being blind as a bat and missing what’s right in front of me there, my wife would call it looking like a man, so yes you have got links in there! Sorry
And yes dc-dc output ground, case ground, 12v ground for your controller board inverter i9 power ground, inverter ground GINV on the i10 plug all need to be connected to the same ground, all eventually linked be it via wires or chassis back to you 12v system battery, personally I have all my important grounds to the inverter case and a big chunky lead from the inverter case to the battery
And yes dc-dc output ground, case ground, 12v ground for your controller board inverter i9 power ground, inverter ground GINV on the i10 plug all need to be connected to the same ground, all eventually linked be it via wires or chassis back to you 12v system battery, personally I have all my important grounds to the inverter case and a big chunky lead from the inverter case to the battery
Re: Toyota Prius Gen 2 Inverter Controller
Best not to think of hv negative as a ground at all really as hv is completely isolated,
-
- Posts: 898
- Joined: Fri Apr 26, 2019 5:40 pm
- Has thanked: 291 times
- Been thanked: 177 times
Re: Toyota Prius Gen 2 Inverter Controller
My family says "... that's because you boy-looked. Look again it's right there." :p
They definitely aren't right now. Could that be the cause of my problems?And yes dc-dc output ground, case ground, 12v ground for your controller board inverter i9 power ground, inverter ground GINV on the i10 plug all need to be connected to the same ground, all eventually linked be it via wires or chassis back to you 12v system battery
I'll add it to my list of things to try next.
Gotcha, yeah, keeping it separate.Best not to think of hv negative as a ground at all really as hv is completely isolated,
Also of note, the big orange HVDC connectors... only the center pin connects to HV+ or HV-. The whole ring of conductors all around it are all case grounded (makes sense, it's probably tied into the wire mesh cable shield). It is annoyingly close though, if you're shoving gator clips in there. Which is why Is clipped to the bus bars.
Re: Toyota Prius Gen 2 Inverter Controller
Yes if you’ve not got all those on a common ground things won’t work as you need to be pulling the phase input pins down to the same ground as ginv using the transistors on your controller board.
Also yes the outside of those hv terminals is a shielding connection so be careful there!
Also yes the outside of those hv terminals is a shielding connection so be careful there!
Re: Toyota Prius Gen 2 Inverter Controller
Just to be clear when I said dc dc ground in the list of common grounds im talking about the case not hv negative going into the dc dc converter, I know people struggle with me not explaining clearly,
-
- Posts: 898
- Joined: Fri Apr 26, 2019 5:40 pm
- Has thanked: 291 times
- Been thanked: 177 times
Re: Toyota Prius Gen 2 Inverter Controller
Well, tonight I went looking for my bending brake so I could fold a rocker panel. Got distracted. And on literally the last 2 boxes in the entire house for me to check, on the very top, very back, I discovered some old friends:
![Image](https://i.imgur.com/e1Z3DQN.png)
![Image](https://i.imgur.com/mFJe7PZ.png)
![Image](https://i.imgur.com/bwRk66Y.png)
I got them both for free, 10 and 15 years ago. I've used scopes before, but it's been even longer than that.
And no, while the the electrical tape, gator clips, and banana plugs are par for my quality, I didn't do that, they came like that.
Presuming the tubes still warm up, I guess I'm ready to attack some circuits.
Haven't so much as even tried tying the low voltage grounds in common as per last suggestion, so, I guess I'll try that first.
![Image](https://i.imgur.com/e1Z3DQN.png)
![Image](https://i.imgur.com/mFJe7PZ.png)
![Image](https://i.imgur.com/bwRk66Y.png)
I got them both for free, 10 and 15 years ago. I've used scopes before, but it's been even longer than that.
And no, while the the electrical tape, gator clips, and banana plugs are par for my quality, I didn't do that, they came like that.
Presuming the tubes still warm up, I guess I'm ready to attack some circuits.
Haven't so much as even tried tying the low voltage grounds in common as per last suggestion, so, I guess I'll try that first.
Re: Toyota Prius Gen 2 Inverter Controller
Hi All,
So I’ve been struggling to get my car project on the road and there has been a few think restricting this, space and Cooling being the main ones. As I’m attempting to use a Prius gen 2 inverter and gen 3 transaxle in my car along with a full stack battery from a Hyundai Ioniq. It just won’t fit and I’m trying to minimise cutting bits of the BMW 3 in order to get going. So I’ve decided to bite the bullet in install a Leaf stack in it.
QUESTION….
Can I use I’m Prius Gen 2 VCU in the Nissan Leaf stack? (Yes or no will do form anyone who has already tried) then I can research.
Will I need to Hack to Charger (this might be a question for another thread)
Much appreciated.
So I’ve been struggling to get my car project on the road and there has been a few think restricting this, space and Cooling being the main ones. As I’m attempting to use a Prius gen 2 inverter and gen 3 transaxle in my car along with a full stack battery from a Hyundai Ioniq. It just won’t fit and I’m trying to minimise cutting bits of the BMW 3 in order to get going. So I’ve decided to bite the bullet in install a Leaf stack in it.
QUESTION….
Can I use I’m Prius Gen 2 VCU in the Nissan Leaf stack? (Yes or no will do form anyone who has already tried) then I can research.
Will I need to Hack to Charger (this might be a question for another thread)
Much appreciated.
Re: Toyota Prius Gen 2 Inverter Controller
No, Prius gen2 controller, actually controls the inverter logic and intelligent gate drives, the leaf inverter can be controlled with just simple can commands