[info] microRusEFI

[JRD McLAREN]

One hint: ..

Many aftermarket ECU use two (or more) voltage regulators.

One is for "main power" ... for cpu, switches, and other..
Second is High precision / High accuracy ... for sensors, TPS, MAP, HALL or other "critical" components.

..

[AndreyB]

Even a regular LD1117 would work very well, and is only $0.20 at qty 100.

LD1117S33TR from mouser would save us a $1 here. What do we think about LD1117S33TR from mouser?

[mck1117]

Many aftermarket ECU use two (or more) voltage regulators.

We already do most of this. The TLE8888 actually contains 5 regulators:

1) 6 volt pre-regulator. Uses an external transistor, Q1. This just supplies the other regulators, getting heat dissipation out of the tle8888.
2) 5v main regulator. This supplies the 3.3v regulator for the MCU, 5v output drivers, etc.
3) 5v tracker supply 1. This tracks the 5v main supply, but is independent and protected.
4) 5v tracker supply 2. Same as #3.
5) 5v standby supply. For persisting memory when the car is off (not used on microRusEfi).

Each of the tracker supplies are routed to the 5v pins in the connector.

We could use a more complex power supply scheme with fully split analog/digital supplies, but based on the testing I've done so far, it just isn't necessary for our use case.

[mck1117]

LD1117S33TR from mouser would save us a $1 here. What do we think about LD1117S33TR from mouser?

The LD1117 would certainly work fine. However, it has a different pinout, so we'd at least need a respin of the board (which will happen anyway).

[JRD McLAREN]

We already do most of this. The TLE8888 actually contains 5 regulators:

Yes, I'm read TLE8888 pdf document ...
but, TLE8888 is not "on the stock" in my country ..
(it is good "all-in-one" chip)

[pk386]

Digging this form factor

[AndreyB]

With one fix and many controversial changes we now have release candidate of version 0.2

See https://github.com/rusefi/hw_microRusEfi/blob/master/microRusEfi_Schematic.pdf
See https://rusefi.com/docs/ibom/microRusEfi_latest.html

[lumpensack2003]

Hi there,

I really like the form factor - fits perfectly the bike crowd.
But in my opinion it would be a shame if there aren't any internal pressure sensors.
Imagine a clean motorcycle install an then, for MAP and AAP, two external pressure sensors with cables / connectors.
I would vote for one, better two, on-board pressure sensors. One with a possibility to connect a pressure line at the case.
Isn't the case already prepared for this?

Regards,
GJ

[kb1gtt]

There is CAN bus. Is there any chance external sensors can be installed? AKA not everyone needs a pressure sensor. MRE is based on minimal parts. CAN bus allows for expansions for those that need more features.

A 10 second duckduckgo search produces this pressure sensor.
https://www.stw-technic.com/products/sensors-and-measurement-technology/pressure-transmitters/pressure-transmitter-mo1-can/

[JRD McLAREN]

If you need any kind of MAP sensor, then try to use "stock" automotive sensors ...

Any type from BOSCH or Delphi / Delco can be used. (from cars)

[Simon@FutureProof]

I don't think there is any room in the case for a MAP sensor, the case has also been designed with the ability to seal it up watertight in mind. A MAP sensor through the case would probably prevent this.

In any case a MAP sensor inside the case means vacuum tubes and that means messy tubes around the bike with more chance of failure and the signal delay that runs of vacuum pipe introduce.

It would be better as JRD suggests to use an OEM MAP sensor direct into the inlet and wire back as would be required for all the other sensors on the bike.

[kb1gtt]

I thought they were looking for MAP and AAP. If you only have one signal, then yeah, just connect that to the input pins of the ECU connector. I understand that mechanical folks some times don't like wires. However there is going to be a bunch of wiring, so it's really just adding one more sensor to that harness that already has a handful of sensors.

What is AAP? We have a list of acronyms here.
https://rusefi.com/wiki/index.php?title=Engine:acronym-phrases

[AndreyB]

Absolute atmo pressure i.e. baro? Guessing from context

[kb1gtt]

Just added AAP to the wiki.

[JRD McLAREN]

.. I saw it somewhere ...

rusEFI can measure atmo pressure ... from MAP sensor ..
.. on stopped engine, or when the ignition is powered on (and engine is stopped) ..

Right ..??

[AndreyB]

We are getting further away from the subject of microRusEfi hardware board

[JRD McLAREN]

OK .. OK ...

so get it back ...

Is CJ125 "onboard" palned ..??
..or not .. ??

(it is one of "standards" for me)

[andreika]

.. on stopped engine, or when the ignition is powered on (and engine is stopped) .. Right ..??

The option is called "Use fixed baro corr from MAP".

[AndreyB]

Only mouser sells TLE8888 in low quantity, digikey does not. This makes mouser the default distributor for MRE.

At the moment mouser does not have X1 crystal NX3225GD-8MHZ-STD-CRA-3 in stock. Obviously digikey is an option but it's not great that we use a unique part with strange availability. Shall we migrate X1 to some more popular package, is there a reason why we use the unusual 3.2 mmx2.5 mm? Would it be better to migrate to cheaper and way more available larger "49smd 8mhz"?

https://github.com/rusefi/hw_microRusEfi/issues/83

[Abricos]

Round Chinese connector terminals not interchangeable with terminals from nissan connector - round terminals would not click into Nissan female plastic connector and vice versa.

Nissan male PCB connector plugs into assembled Chinese female connector with round terminals just fine.

Larger pin:

Screenshot_20190712-193925.jpg

Screenshot_20190712-193914.jpg

Screenshot_20190712-193950.jpg

Smaller pin:

Screenshot_20190712-194055.jpg

Screenshot_20190712-194016.jpg

Screenshot_20190712-194033.jpg

[AndreyB]

Good news: MRE 0.1 assembled by mck1117 has arrived. Thank you Matt!

Good news: USB connectivity works using GND/D+/D- wires from the connector.

Bad news: ECU reboots on start of cranking.

Bad news: I might have damaged my unit

My small Miata battery is charged and sits on 10A automatic charger.

The moment I start cranking ECU status LED stops blinking, after few seconds of
cranking 5v power LED also starts to blink.


Here is battery voltage drop during cranking


Between shotting the videos I was moving DMM from +5v on MRE to +12 power source on the connector.
DMM was on in +20 voltage measurment mode, MRE was on, power was on - this was not smart I guess

The moment I've touched hot +12v wire with my DMM lead MRE has turned off.

I've moved MRE to the bench - MCU is fine, it runs if powered by 5v.

But TLE8888 is not producing 5v any more I get 12v on Q1 and C2 but not 5v output, not sure how to troubleshoot.

[kb1gtt]

Bench top 12V setup? Also what is your setup? Do you have injectors / igniters installed, or are you just trying to decode the crank? I also wonder if you have dodgy wires. Perhaps a poor connection is causing a voltage drop. You know the drill, pictures to help remove my blind fold.

[AndreyB]

Car setup: OEM ECU harness with female connectors plugged into male 64 pin connector on breakout board. Breakout board soldered to 48 pin molex 50cm pigtail. 48 pin pigtail plugged to MRE, connector soldered to red PCB. Everything soldered. The whole car is very OEM, very few modifications.

Bench setup: $35 5A 30V power supply.

Only 9 wires soldered to connector so far: two GND wires , 12V, 5v output, CAM/CRANK inputs, coil#1 and coil#2 outputs, one injector output wired.

[mck1117]

Let's start with the first rule of electronics diagnosis: Thou Shalt Check Voltages

Apply 12v to the 12v pin, then check the voltage at:

1. Anode of diode D1 (end towards connector): expected ~12v
2. Cathode of diode D1 (end towards Q1/away from connector): expected ~11.45v
3. Pin 2 of Q1 (large tab, toward edge of board): expected same as 2)
4. Pin 1 of Q1 (pin closest to big capacitor): expected ~8.1v
5. Pin 3 of Q1 (pin most distant from big capacitor): expected 6.0v
6. 5v test pad: expected 5.0v

Expected voltages are measured on my working MRE board, connected to a 12v supply. Voltages #1, 2, and 3 may vary slightly based on the supply voltage.

That should get us a start for what's cooked.

[960]

As above, measuring those things are first step.
However, the 8888 and the power supply components are REALLY hard to kill.
I dont know how this is soldered and checked? Did you have it under a good microscope?

Also, measure both sides of resistors at Key pin

And, does the prereg or 8888 produce any heat?

[AndreyB]

1. Anode of diode D1 (end towards connector): reads 12.0v
2. Cathode of diode D1 (end towards Q1/away from connector): reads ~11.8v
3. Pin 2 of Q1 (large tab, toward edge of board): reads ~11.8v
4. Pin 1 of Q1 (pin closest to big capacitor): reads 0.0v
5. Pin 3 of Q1 (pin most distant from big capacitor): reads 0.0v

[960]

Can you find resistor for pin Key and measure? I am in car without diagram. But this means gate voltage for prereg missing?

Also measure the other resistors at 8888 that shows 12v in diagram

[AndreyB]

Can you find resistor for pin Key and measure? I am in car without diagram. But this means gate voltage for prereg missing?

Also measure the other resistors at 8888 that shows 12v in diagram

I am not comfortable with this style of instructions, please provide exact instructions if you have time to provide exact instructions. I can measure voltage at exact components if you give me exact part numbers like Rxx or Cxx or Qxx but I am not comfortable deciphering "resistor for pin Key" or "prereg ", these terms are confusing to me and the whole set of instructions is only confusing me and creates tension.

[960]

Ok, sorry. Been driving all day and just got home.

R20 and R5, both ends of them

If you have voltage at R5, and all soldering are ok, you probably have the first 8888 that power supply fails.

I have beaten them so bad in every way, and never damaged any. So I really doubt it.

[AndreyB]

R20 and R5, both ends of them

Both sides of R5 read 6.0v, same 6.0v on bottom side of R20.
Top side of R20 reads 12.0v

@mck1117 has soldered this MRE 0.1 number #2.