[mudder]

Here's the complete pinout. That'll be useful as you probe.

Unfortunately we've looked long and hard for information on the microcontroller, but have never found anything.

I'm not certain how much information from the OEM MCM is necessary for an LiMCM adaptation... in fact, back in 2016 I made a drop-in replacement PCB that fit in the OEM MCM enclosure. That product was never released.

One use for an MCM schematic is for continued hacking without replacing that PCB.

 

[mudder]

I posted the HVDC portion of the MCM schematic somewhere in this thread:

G1 Honda Insight Lithium BMS Discussion

@retepsnikrep and I are considering designing a simple battery management system that supports Honda's 3rd generation Insight lithium battery packs. We've discovered that these packs mechanically fit in the G1's OEM IMA battery bay, which makes them an IDEAL candidate for simple lithium...

www.insightcentral.net

 

[mudder]

@mudder I recall this thread and your reversing of the circuit to determine how it was doing voltage measurement. Did you note how the car was detecting whether the HV pack remained isolated from the LV system?

Yes, it switches an engine-ground-referenced current source alternatively into HVDC+ & HVDC-, then measures the resulting voltage. It then calculates the insulation resistance between the HVDC bus and engine ground.

 

[mudder]

Great, you found it! It's possible the resistor values changed between versions (I inspected a 1K000-PHM-070 unit). Another possibility is that I made mistakes; I yield to your exhaustive (and destructive) MCM analysis.

I do have more notes in a notebook, but not in any presentable format. It'll probably be easier for you to just continue reverse engineering, and then compare your results to mine (at the link you found).

 

[mudder]

@eprom:
Here are all my notes on every signal in the IMA bay. That's an OpenOffice spreadsheet. Click the 'MCM' tab and that's every wire going to the MCM.

 

[mudder]

I've referenced this schematic several times this week. Thanks for taking the time to reverse engineer the I/O.

FYI:
C25 is 220 nF
C56 is 2.2 uF

 

[mudder]

'IL OZ' could be 2SA1463IL
'OK OZ' could be 2SC3736OK

 

[mudder]

From the draft MCM schematics attached to this post, it looks like the MCM can transmit on the METSCI BUS but cannot receive.
And it seems that on BATSCI, the MCM can receive, but not transmit.

Does that make sense to someone familiar with communication on these channels?
@mudder @retepsnikrep ?

Yes, that is correct. Note that some data the MCM transmits on METSCI actually ends up being retransmitted back to the MCM on BATTSCI.

 

[mudder]

Thanks for continuing to chip away at the MCM I/O!

 

[mudder]

Yes, the MCM'E' input also monitors the high pack isolation.

 

[mudder]

Thanks for looking into whether the MCM has an ambient temp sensor.
Die temp wouldn't be useful for measuring ambient temp.
NF8 is probably a pi filter, with a series inductor and a parallel capacitor (to ground) on both sides.

At this point I suspect it's the ECM that's commanding regen when it's cold outside. This is easy to test if you're experiencing the 'background regen when cold' issue and have a manual IMA controller.

 

[mudder]

I don't quite understand the purpose of transistor Q16 which is bypassed by D31.

Q16 is an LDO regulator. When IC22 drives pin 10, the voltage drop across Q16||D31 will be much less than it would be (~0.7V) when Q16 is off.

One point that I noticed the capacitor C32 is electrolytic, it is always powered, so to date on the insight 2000, this capacitor has been powered continuously for 23 years, it is possible that it is starting to age.
I found mine a little rounded but I couldn't unsolder it to measure it because there is a large thermal mass underneath.

You'll need a board preheater and/or a heat gun to remove it. Note that capacitor aging is a function of the actual applied voltage, versus the maximum specified value. For electrolytics, typically you'd want Vrms to be less than 65% of nominal rating.

 

[mudder]

Maybe you can glean information from these pages:

TA8903SN IC17 Toyota Replacment/Emulator Honda OBD1 Apollo Tuning | eBay

Find many great new & used options and get the best deals for TA8903SN IC17 Toyota Replacment/Emulator Honda OBD1 Apollo Tuning at the best online prices at eBay! Free shipping for many products!

www.ebay.com

IC17 Replacement

TA8903SN emulator. This fixes the problem with many dead obd1 honda ecu's that will not activate the main relay/fuel pump or its delayed. Available in bulk just select quantity from the drop down menu. Reference to install direction in second picture.

jasonwaterstuning.com

TA8903SN TOSHIBA | TA8903SN Electronic Chips Distributor | TA8903SN MG100J6ES52 TC551001CFTI-70L 2SC4685 TC4069UBFN stock available from veswin.com

TA8903SN, MG100J6ES52 TC551001CFTI-70L 2SC4685 TC4069UBFN from TOSHIBA Electronic Chips at Veswin Component Distributor, TA8903SN large in-stock quantities. PayPal accepted, inquire TA8903SN Specifications online at veswin.com or RFQ Email [email protected]

www.veswin.com

Looks like it might be a quad transistor IC.