openinverter forum

celeron55 wrote: ↑Wed Jun 09, 2021 3:59 am

chrskly wrote: ↑Sun May 23, 2021 7:53 pm I'm seeing an issue where my main contactor is not closing. I have UDCSW set to zero. The precharge contactor closes. No errors in the web interface. It seems like the inverter thinks it's closing the main contactor.

How are you starting the inverter? Are you using the start input or manual mode?

Previously, I was starting the inverter in manual mode. I burned out two precharge resistors this way until I realised that the main contactor wasn't closing.

In this last test, I didn't manually start the inverter, as I presumed that closing the main contactor was part of the normal start up sequence. Do I have to pull the 'mg2 start' input high or hit 'start in manual mode' to get the main contactor to close? I thought the 'start' just started the pwm signals.

The contactor drive does not work in manual mode. In order for the main contactor to close you need :

1)UDC>UDCSW
2)No errors or faults. Check the web interface.
3)Momentary signal on DIN_START.

Check your mode. does it say : RUN?
The precharge contactor should switch off after 5 seconds if RUN mode is not achieved.

Damn, OK, this is fully a misunderstanding on my part. Thanks all for the help. Turns out my suspicion of me being dense was a good instinct

Confirmed that when I give DIN_START (MG2_start) 12v momentarily, I see ~3.5v on pin 2 of the STM32F.

Manual run != run

Don't sweat it. I'm the guy who can't wire up 2 contactors:)

I have been doing some bench testing and mild test driving with the v1d small board and wanted to share the problems (and solutions) I encountered:

First I tested the board with a 2015 Yaris Inverter. It works well together with a Prius 2 transaxle, however, I had to add an additional 5V supply for the Wifi modules. The onboard supply could not keep up the power when adding two ESP8266 modules.

Currently I am working with a 2015 Prius III inverter. Same problem with the 5V supply and this time the -5V supply also stopped working and outputted 0.7V. The same board that worked with the yaris did not work with the Prius III inverter, even though both use the same power board and current sensors (at least according to the part numbers). Adding an additional LM2776 in parallel to the one on the board solved that problem and works well so far, although I do not know if the LM2776 is actually designed for parallel use.

excellent remarks. I had to add an extra 5V power supply for the wifi modules.
but here -5 volts for current sensors is something suspicious.
Thank you so much. we will know in the future.

The Block 4 version of the V1d board addresses these problems and some others. 5v supply is now upgraded and capable of 2 amps. -5v supply has been changed to a MAX735 switch mode device as some -5v charge pump devices have a problem starting up into a load despite their respective datasheets claiming they can do just that.

Also I'd like to advise folks waiting on boards that they will be dispatched in the next few days. It's been a bit of nightmare pulling all the parts together with the current shortages.

Alerted by your posts I started to check my power converters, because I didn't used them at my stage of development yet.
And I must say:

1. XL1509-5.0 is a complete piece of crap. Firstly it supposed to give 5V, but it gave 9V. After adding external resistor to its internal voltage divider it gave 5V, but under little current draw (30mA from 3x STM32 and linear regulator) it dropped under 3V.
There was earlier TPS54331. Why you switched to XL1509?

2. I had two days of nightmare with MC34063A. It simply won't switch in boost mode. It only puts thru voltage. I've changed few ICs, but from this same seller. Build another converter. Nothing! It just won't switch.

I've ordered from Ali some assembled converters to swap my ICs and see if they are faulty. Will see.
https://www.aliexpress.com/item/1005002494251409.html

3. I have no issue with -5V yet, but I didn't used it under load.

The XL1509 works perfectly. Added Block 4 PDFs to the repo :
https://github.com/damienmaguire/Prius- ... master/V1d

My inverter blowup meant I could take pictures of my board mods:

1. Barely visible: 560R resistor to 12V pulling up IGBT. Also connected to precharge FET meaning IGCT is low at startup and thus DC-DC converter off
2. Single wifi module mod
3. CAN transceiver for MG2 CPU. Sits on top of the Atmel one. Not sure if this would work simultaneously with the onboard CAN to MG1. Have deleted that.

I never wired up PB9 to the DC-DC converter for setting voltage via user PWM.

Would these be mods for the next revision? Is this even the right place to discuss?

What have you done to blow this inverter?

zilion wrote: ↑Wed Sep 01, 2021 2:07 pm What have you done to blow this inverter?

viewtopic.php?p=31165#p31165

I added the following text to the wiki:

Be super careful never to program the Atmega while high voltage is applied and caps are not discharged. When cycling through the boot loader, it seems to do something strange that will blow up the otherwise bullet proof buck/boost converter! Also be aware that Arduino also cycles through the boot loader when closing the serial terminal!

I blew up another inverter today

Due to an "upgrade" on Olimex wifi modules I can no longer supply them as open networks. By default all charger kits will have SSID : inverter PASSWORD : inverter123

If someone could please update the wiki with this would be much appreciated.

Jack Bauer wrote: ↑Wed Oct 20, 2021 10:52 amIf someone could please update the wiki with this would be much appreciated

BigPie appears to already have that covered.

Hello All,

I need a little help if possible I've just finished soldering all the parts to my V1D board and I seem to be short C93 this was a kit from the BMW webshop and there is a greater than 0% chance it was included and I've launched it into the atmoshpere without noticing it even existed...

is anyone able to confirm that it is required to be populated and that as per the BOM its supposed to be a;

Kemet 0805 1μF Ceramic Multilayer Capacitor, 16V dc X7R Dielectric ±10% SMD




thank you in advance!

Hello,

apon looking further into this i believe saw that c93 on the schematic needs a 47uf cap which I have placed in the c93 slot but on the voltage check im only getting -1.47v where i should be getting -5v? any help please?

also the v1d board BOM isnt up on the github as far as i can see only v1c?

Hello,

I have the same problem with my V1d B4 board. Check if the inductor L7 is on the board, on my board this inductor is missing, but it is absolutely necessary for the -5V converter to work. In case it is missing on Your board, You have to solder it manually.

@Damien: on my board the capacitor C93 is also missing. Could You please share with us an information which capacitor should it exactly be? (capacitance, type, voltage and foorprint) I would really appriciate the help.

PS: I have noticed, that this capacitor is 47uF on the schematic, but in the datasheet of the MAX735CSA+ there should be 1uF ceramic capacitor parallel with 47uF electrolytic capacitor. Could You please confirm which one should I solder here?

Great thanks in advance for Your help!

C93 should be a 47uF 0805 6.3v ceramic cap such as :
https://www.lcsc.com/product-detail/Mul ... 35553.html

Hi,

I have finally became my inductor and capacitor soldered, but the DC/DC converter for -5V (IC17) was still not working properly (1.5V at output instead of the -5V).

But I have eventually found out what was wrong:
THE DIODE D9 WAS SOLDERED IN THE WRONG DIRECTION!!! I have resoldered it and now it works perfectly fine!

The bar on the diode indicating the cathode should point right in the direction of the inductor, but it was soldered differently.

So guys, be carefull with Your boards, as this error may be potentially repeated in other boards as well.

Hello all. New to the forum as a member but been lurking for years.
Have few questions about the gen 3 dual motor board before i comit to buying one.

1.) Can mg1 inverter be used to drive the dumb non can bus prius ac compressor? Can it run a motor sensorless. And what inputs are available to run mg1 inverter. I watched in the video that mg1 gets its info can from mg2 stm32 chip. But can i run the mg1 seperately and not tied at all to mg2.

2.) Is the coolant pump controlled by the dual motor board?

3.) Dc/dc converter controlled by the board? Can an output generate pwm or whatever the dc/dc converter needs to turn on.

4.) Can the atmega on board be used for anything other than just the buck boost converter? Are there other pins acessable?

5.) Does the dual motor board have ability to communicate either can or analog with a bms?

6.) Will gen 3 dual motor board work with 2012+ prius gen 3 inverters or only the 2010-2012

7.) Is there inputs for motor temp sensors? Motor i have doesnt have them but id like to add them.

8.) Is there any support setting wise for honda ima motor?

Im sure ill have other questions. Im still reading on it.
Now the project..
Mazda rx8 with 3.2l v6 honda vtec engine. Using the honda accord ima motor on back of the v6 going to a bmw 7 speed dual clutch transmission in conventional rwd layout. I am going for a performance hybrid. Using the ima motor for assist and regen to charge batteries. Plans to run prius gen 3 inverter and use mg2 power stage for the ima motor and hoping to use the mg1 power stage for the dumb non inverter prius ac compressor. If this wont work does anyone know the can data for the prius compressor with can bus inverter on it.
For batteries planning on running a prius prime pack. Which brings up another question can i just hook up the prius prime pack to the inverter? Or is 350vdc too much for the gen 3? And require changes to dc dc converter or require the buck converter to lower voltage for the dc dc converter?

The honda ima system is pain to control and its components would possibly hold up but the open inverter is a better system.
In stock form the honda runs a 144vdc pack and up to about 250vdc has been put through an insight ima motor. No one has pushed further as oem dc dc converter shuts off and they are using stock inverter.

I plan to run either the full prime pack or 4 of the 5 modules.
Can the boost converter be used to up pack voltage to be ysed by mg2 inverter? Say you had a normal driving mode and another mode wjere you wanted more power. Could the boost converter be enabled to provide more voltage from the pack.
Id like to do multi position potentiometers on both assist and regen so on the fly i can with click of a knob change the amount of assist and regen.

As for throttle and regen the controller will have to comunicate eith the engine ecu. I have analog, digital and can bus available.

And lastly ill need to create an engine start profile, parameter or sequence as its replacing the starter. Is there an extra input for this and ability to put it in code that when input pin goes high to rotate engine 1000rpm for given amount of time to start the has engine.

Thats all for now. Any help would be greatly appreciated. I have 2 gen 3 prius inverters on their way. And looking for the info listed above before i buy a board so i buy the correct one that will fit my needs.

slociviccoupe wrote: ↑Fri Mar 25, 2022 11:07 pm Have few questions about the gen 3 dual motor board before i comit to buying one.

1.) I think you could run an AC motor sensorless using the sine firmware, but I am not sure about this. You can control MG1 separately by using CAN. The CAN bus on the board is only between MG1 and MG2, so you have to break it up and add two transceivers to control it from the outside.
2.) No
3.) Yes, it is controlled via an enable signal (check out the schematic on github https://github.com/damienmaguire/Prius-Gen3-Inverter)
4.) Check the schematic linked above
5.) Yes, via CAN (see 1.) )
6.) It has been successfully tested with a 2010 inverter (https://openinverter.org/wiki/Toyota_Pr ... el_Numbers)
7.) Yes, there are temp inputs but they need a little rewiring (see wiki)
8.) I don’t think so

@leonB so for wanting independent control over mg1 and mg2 the dual motor board isn't it. Seems like other work would have to be done to the dual motor board such as cutting traces between the 2 microcontrollers and adding external circuits. That wanting independent control over mg1 and mg2 to run 2 seperate motors not linked together that (2) main boards or 2 mini main boards would accomplish that. One board for each power stage.
Can one of the mini main boards control the mg1 igbt and the buck boost if at later date decide not to use mg1 for driving a motor.
I really want to run a prius gen 2 compressor from mg1 and honda ima from mg2.