Dmm

Hi. Need some advice, please.
I have caprice '91 with sbc 350 & 700r4 & 6965. For rise up driveability i replaced it with a '89 305 camaro's 8746. Running great as i want, but tcc won't lockup.
I suspect a fault in the tcc inner ground circuit at pin a7. Because grounding pin F in obd connector force tcc lockup. With 6965 tcc lockup working properly.
How to check the transistor switch or else what controls the ground? I want try to repair it.

What i checked: circuit from battery + to pin a7, brake switch, tcc solenoid, 4th gear swith, ecm ground&ecm case ground&ecm sensor pins ground. Hi gear switch polarity in 8746 prom doesn't affect.

Some winaldl logs:

6965 working correct, 100km/h = 1850rpm (correct for 3,42 rear gear+255/70/15 tire), tps fits to load limit table, tcc 1 = locked. Everything is fine, for now...

Replaced 8746, rpm floating = tcc unlock, BUUUUT tcc indicate lockup!

8746, tcc disengage in prom, forced pin F grounding for 5 seconds reduced rpm to correct spd/rpm value 100@1850
Probably don't need to say that after returning 6965 tcc lockup works again...

Maybe someone already had the expirience of recovering a ecm tcc circuit?

Dmm

From a camaro, of course, only ecm, prom is ANLU with changed BPW vs. IAT parameter and EGR is off.

Dmm

I found no difference between a 4l60 camaro&caprice, except caprice's temperature switch in tcc circuit.

RBob

iTrader: (1)

It just may be that the quad driver in the ECM is bad. Pop the cover off and look at the two chips near the ECM harness connectors.

They are a 16-pin chip (2 of), and may be fried. Look for/at U14 & U15.

RBob.

Dmm

Got it. Tcc pin a7 tracked to U15 Delco 274 Z 60904. U14 is 16125773 by Allegro Microsystems(?). Interestingly that 6965 has 274 H 60904. It's interchangeable?
Besides 8746 i took one 6965 with the same problem with tcc and one serviceable 6965 as a reference sample. All of them look healthy. No pcb detachment or gap or something melted.
Could't get datasheet for 60904 but found it for similar DS3668 ( pdf ) and roughly understood how it works. Check it without desoldering from plate apparently impossible because i can't emulate data bus on 9,10,15,16 pins. So I tried to find a current substitute from the range available in my city. For example L293 ( pdf ) but it's not really similar. I'm stuck.

U15 pins:
1 - A4 EGR
6+8 - A7 TCC
3 - A5 SES connect with 5(gnd) through film 100F capacitor
9,10 - goes to 16-pin Delco 43538
15,16 - goes to 40-pin Delco 23263
2+7&11+14 - through 3 diodes respectively connect to 2,7&11,14 U14 pin (???)
4,5,12,13 - gnd

U14 pins:
1 - C2 smog
3 - C1 smog
6 - A3 CCP
8 - A2 not used
10 - goes to 40-pin Delco 23263
9,15,16 - goes to 40-pin black chip
2,7,11,14 = U15
4,5,12,13 - gnd

Pinout shows that the DS3668 looks like 60904.

8746 our patient

U15 60904
U14 16125773

6965 slightly roasted by myself, no tcc lockup too, but everything else is fine...

U15 60904

U14 16125773

And healthy 6965

U15 60904

U14 16125773

No difference, truly? Hmm

Dmm

I could sacrifice one 6965 for 8746, but I'm not sure that the 16125773 compatible with 60904. And this capacitor. And big chance to finish off 8746.

RBob

iTrader: (1)

The DS3668 is what we use to replace failed quad-drivers.

You can do some in car testing. Solder a wire to pin 9, when high (+5 V) that commands that driver to turn on.

Solder another wire to pin 8, should go low when pin 9 is high (TCC enable). Otherwise should sit at about vehicle voltage.

Be sure to attach the upper case half to the metal heat sink of the ECM. Even if upside down with just those two screws.

RBob.

Dmm

Here we go. Added one extra wire.

Results:

Ignition off
9 to gnd = 0v
8 to gnd = 0v
OBD pin F to ECM pin A7 = 0v

Ignition on
9 to gnd = around +0.05v
8 to gnd = 0v
F to A7 = 0v

In drive
9 to gnd = depending on the rpm +0.033-0.077v
8 to gnd = 0v, sometimes reading around +0.01v
F to A7 = 0v

TCC lockup
9 to gnd = around +5.49v
8 to gnd = around +0.014-0.018v
F to A7 = +13.6v = vehicle voltage

Ok, control +5v applied on pin 9, the path for +13.6v opens, but the burnt transistor cuts off the voltage to the minimum, right?

Dmm

And it's not clear what to do with the capacitor. In 6965 he is absent. If i replace 60904 with 3668, do i need to solder capacitor back?

Dmm

While waiting for delivery of ds3668 from the mouser bought another 1228746.
And what would you think? Tcc doesn't work either. Winald displays the active tcc flag, but the rpms still float.

After that, I assumed that the problem might be in bin:
-changed the load limit tables to match the actual tps.
-made a min of temperature for the tcc enable
-checked affecting 4 switches polarity and man/auto trans flag
-removed the flag tcc unlock in dfco
No changes. The tcc lock flag is present, but there is no physical lock.

Either I am very lucky and bought two non-working ecms, or I am stupid and cannot notice a slight difference in the control of the tcc between 8746 and 6965. And the problem is absolutely not in the quad driver.

According to the alldata, only three control wires go to 4l60: + 12V, solenoid ground control and 4th gear sensor. They are conventionally the same in caprice and camaro. Solenoids are the same too.
Three wires also go to the ECM: A7 = tcc sol ground, C7 = 4th gear switch and A10 = vss buffer. Now I'm looking for differences in the vss buffer camaro and caprice. The speed is displayed correctly in winaldl but there may be a bad effect on the tcc.

I ran out of ideas, so requesting RBob... please...

Dmm

So, during this time I managed to receive and replace the quad driver, it did not help, the lockup did not work. Then I installed 16197427, a much more pleasant ecm compared to 8746. But the lockup still did not appear.
Last month, I serviced the engine, changed 187 heads to grinded 193, cleaned the transmission connector and ground points on the block and body. Then rode on 7427, rides very vigorously, but still there was no lockup. Today I put back 8746 with a soldered quad driver and, oh miracle!!, the lockup began to turn on.
What is that?! The connector was purged and cleaned last year just after the lock disappeared.