rusefi.com

Time to introduce the project.

It's a PNP Ecu made for all ME 7.1 and ME 7.5 cars.

It has:

2x TLE7209 H-bridge drivers for E-Throttle
2x CJ125 LSU Drivers
2x L9651 Injector Drivers, 8 x 2.2 A outputs
10x Digital IO
16x Analog Inputs
22x Low Side outputs in addition to the injector drivers
TLE8888-1QK For Power supply and rest of the outputs
STM32F767 MCU

Looks like this project has a lot of HW engineers and leaks of programmers.

The STM32F767 is a new chip for rusEFI. I understand it's currently blinking a LED, which is a huge first step. Porting the the existing rusEFI software is underway. As well this chip offers several diagnostics features via the SPI, so lots of new software will be beneficial for this PCB.

Dron_Gus wrote: ↑

Mon Feb 25, 2019 8:31 am

Looks like this project has a lot of HW engineers and leaks of programmers.

Well, that's how it's usually done.

Hardware needs to be in place before programming.

Developing and debugging such complex software and enhancing its functionality requires a lot of effort and leaves much more room for mistakes and errors, which might be sort of discouraging (when diving in some projects people usually have certain expectations about the expected outcomes, and if they are not sure about the future results, they usually tend not to step up to such endeavours.

On the contrary, since it's not a GHz-domain project, with no RF technologies involved, building HW for it is probably not that difficult, once you are aware of and comply with the well-defined good pcb design practices.

All this, as well as the the obvious need for highly customizable aftermarket PnP ECUs lead to the current proliferation of various PCB designs (there seem to be six or seven independent designs based on rusefi firmware that have been launched over the past couple of years).

Back to the topic. What are the model years that used those ME ecus? Seems those newer engines have much higher requirements in terms of the quality of control and are prone to any slightest mistakes in tuning, compared to older engines. I could be wrong though.

960 wrote: ↑

Mon Feb 25, 2019 7:43 am

STM32F767 MCU

Did you get it to blink status LED with STM32F767 or any other chip?

russian wrote: ↑

Mon Feb 25, 2019 4:58 pm

960 wrote: ↑

Mon Feb 25, 2019 7:43 am

STM32F767 MCU

Did you get it to blink status LED with STM32F767 or any other chip?

I have one with a F4 that most is working.

But I had to order new pcb's cause of the lack of mask between pads.

It's impossible to get right as it is.

I think a F7 will work when that is resolved.

russian wrote: ↑

Mon Feb 25, 2019 4:58 pm

960 wrote: ↑

Mon Feb 25, 2019 7:43 am

STM32F767 MCU

Did you get it to blink status LED with STM32F767 or any other chip?

I have a F767 soldered properly, but still the same as last time.

No led blinking, just a weak constant light.

Both SWD and DFU working.

Do you have any idea?

What is the exact part number of stm32f4 which has worked and what is the exact part number of stm32f7 which does not work?
How sure are you about pin compatibility between these two chips?

russian wrote: ↑

Fri Mar 01, 2019 6:29 pm

What is the exact part number of stm32f4 which has worked and what is the exact part number of stm32f7 which does not work?
How sure are you about pin compatibility between these two chips?

STM32F407ZGT7

STM32F767ZIT6

I recall comparing the datasheets, but maybe it needs another look.

960 wrote: ↑

Fri Mar 01, 2019 6:33 pm

STM32F407ZGT7

STM32F767ZIT6

Pins 95 and 121 show minor differences, no idea if this is important, You probably want to look into pins 95 and 121 and confirm that these differences are important or not?

They are connected to VDD.

I see they are both connected to VDD in the Nucleo schematics as well.

https://www.st.com/content/ccc/resource/technical/document/user_manual/group0/26/49/90/2e/33/0d/4a/da/DM00244518/files/DM00244518.pdf/jcr:content/translations/en.DM00244518.pdf

960 wrote: ↑

Fri Mar 01, 2019 6:33 pm

STM32F767ZIT6

Since you have SWD working you need to start debugging the firmware - you shall run openocd and gdb and start stepping over the code and see how far it goes within the firmware. Let me know if you want me to help you set openocd via teamviewer - please install teamviewer and we would need some channel for audio. google hangouts or skype or whatsapp or viber.

Since 960 is not posting

A little update:

The F767 are working fine, and a lot have been sorted out.

For the TLE8888, the outputs are still not working as they need SPI.

Boards are sent off to russian and arrives today, so I guess it's up and running soon

russian will probably post some updates and pictures here as well.

I have the boards and @ has just moved the software forward under https://github.com/rusefi/rusefi/pull/733

Dear Santa, please give me a stereo microscope and 10 extra hours of life every week.

On the picture we see @'s sniffing setup

https://www.ebay.com/itm/AmScope-10X-Widefield-Sharp-Stereo-Microscope-w-Boom-Stand-Gooseneck-LED-Light/382585798395?epid=28024650522&hash=item5913e3fefb:g:p~YAAOSwyPFcFKEx

This is the one you want!

Let's start to donate for russian to get one!?

I just donated $50.

If you all donate a little, russian soon have his microscope

960 wrote: ↑

Thu Apr 04, 2019 9:48 pm

If you all donate a little, russian soon have his microscope

Thank you!

If you all donate a lot, @ simply retires

russian wrote: ↑

Thu Apr 04, 2019 10:43 pm

If you all donate a lot, @ simply retires

Dangle the carrot just far enough...

russian wrote: ↑

Thu Apr 04, 2019 10:43 pm

960 wrote: ↑

Thu Apr 04, 2019 9:48 pm

If you all donate a little, russian soon have his microscope

Thank you!

If you all donate a lot, @ simply retires

That would be the absolute best, but open source usually dont make people rich

Come on folks, we now just need 135$ to reach the goal:)

Post donations here!

#microscopeforrussian

960 wrote: ↑

Mon Feb 25, 2019 7:43 am

2x TLE7209 H-bridge drivers for E-Throttle
16x Analog Inputs
TLE8888-1QK For Power supply and rest of the outputs

Do you use any op-amps on analog inputs - which part number do you use?
Do you have half-bridges wired? Do you expect TLE8888 itself to be able to drive one or two E-Throttles?

russian wrote: ↑

Sun Apr 07, 2019 2:47 pm

960 wrote: ↑

Mon Feb 25, 2019 7:43 am

2x TLE7209 H-bridge drivers for E-Throttle
16x Analog Inputs
TLE8888-1QK For Power supply and rest of the outputs

Do you use any op-amps on analog inputs - which part number do you use?
Do you have half-bridges wired? Do you expect TLE8888 itself to be able to drive one or two E-Throttles?

The analog inputs only have voltage dividers.
I did not have the right ones then, but the part numbers are YC324-JK-0710KL and 744C083153JP.
4 of each.

The Arrays for digital inputs are missing as well, they are 744C083222JP and
744C083333JP

The TLE7209 H-Bridges are connected directly to the MCU.
Wont it work?

960 wrote: ↑

Sun Apr 07, 2019 2:54 pm

The TLE7209 H-Bridges are connected directly to the MCU.
Wont it work?

Do not know. My question was if we expect TLE8888 to work as well or not.

russian wrote: ↑

Sun Apr 07, 2019 4:20 pm

960 wrote: ↑

Sun Apr 07, 2019 2:54 pm

The TLE7209 H-Bridges are connected directly to the MCU.
Wont it work?

Do not know. My question was if we expect TLE8888 to work as well or not.

It shoud be able to, but I dont know how easy it is to implement.

I have not connected them to anything.

I spoke to a guy at Infineon a time ago, and he claimed it was used for GDI engines as well
I recall he mentioned something about IGBT's.

Tested in car today, but got no trigger signal.

I am not sure about the trigger setup.

How do you usually set up crank sensor?

I used 4 5K Resistors in series and one 5K in paralell.

No pull-up or pull down, but I have footprints for it.

This is the original setup:

This is the internal circuit in TLE8888:

did you assemble your frankenso max circuit?

with VR on frankenso one side is usually pulled down to GND

russian wrote: ↑

Wed Apr 10, 2019 8:47 pm

did you assemble your frankenso max circuit?

with VR on frankenso one side is usually pulled down to GND

I have not found any info on it yet.

I have this circuit for the trigger.

I see the frankeso have both resistors on positive side. Do you use pull-down at positive?

It does not count if i try triggering with 1.5V, not 5V, but if I touch wire to 12V it counts.


EDIT: Just removed the Cap across, and now it counts with a 1.5v battery