Bench testing GPIO for 4l60e

[miwyl]

Lance
I too would be interested in a stim board I have to test solutions for exhaust braking and tps load sensing ,methanol injection and so on
my project is with a 4bd1t isuzu diesel

thanks for this great technological endeavor

Mike

[tjabo]

Thad,

If 20 people committed to buying these boards, they would be about $14 each for bare boards (and less than $10 each if 50 were interested), then the BOM components would also be required (http://www.megamanual.com/mshftstimbom.htm).

Lance.

Sorry Lance, I guess I didn't have the "notify when a reply is posted" box checked, so I missed your reply until just now. I'd be happy to pay the $14 plus shipping for the board, or even more than that if a lower number was all you could get commitments for.

Thanks!

Thad

[Bernard Fife]

Thad,

Okay, that's two people indicating an interest in trans stims. We need a bunch more to keep these from costing a hundred dollars each, though. If we get somewhere over 12 (or so) people indicating serious interest, I will get them made.

BTW, the forum, emails aren't always arriving in people's mailboxes due to a server issue a few weeks ago. This is getting sorted out, but may take some time. In the meantime, it's best to check back here occasionally.

Lance.

[tjabo]

Thanks Lance!

(and no, I didn't get this notification either. Just in case, I'm going to go in and verify that I have my settings listed correctly)

Thad

EDIT: I don't see any big problems with my account, so I guess it must be the forum as you said Lance.

[mspinale]

Lance,

My GPIO build and install seems to be working correctly! The maiden voyage was cut short by an older Paxton fuel pump that could not says Aeromotive (who built it) tolerate Ethanol. I slowly lost fuel pressure as the noise from the pump increased

The 4L60e is shifting through all the gears and CAN link to MSII is working fine. There are some things that I am sure will need adjustment but all the critical things work!

One initial impression is around the 3800 stall converter. Wondering if the TCC can/should come in early for "easy" driving.

Thanks again for the help!

Marc

[Bernard Fife]

Marc,

Great, I'm glad it's working for you!

There are a number of TCC settings you can tune under 'http://www.msgpio.com/manuals/mshift/4L60E_MT.html#gt' hopefully adjusting these will help with your high stall converter.

Lance.

[jdrapersys]

I have begun bench testing my GPIO for 4l60e and so far I have not been able to get the indicator in TunerStudio for input2 to change value from "low" even though AMP6 is not grounded. I can get shifted into reverse and 1st gear (even though input2 is "low"), but I can not get manual up or down shift to have an effect. I suspect part of the problem is no input from a VSS. I thought that my tach readings coming over the CAN would give me a speedometer reading, but it is not to be or at least I haven't figured out how to get the speed calculated from the tach.

So how do I simulate a signal from a VSS. Looking at the schematic for the Trans Stim I see that input is still required from CWP L+,L-,R+ and R-. What is the CWP? Can I use the tach output from the X2 connector on the stim for MSII?

Oh and Lance, I'm new to this forum and don't know and can't figure out how to private message you, but I will remain interested in buying a transmission stim board unless I can find some other way to bench test. Is there anyone who has a used Trans Stim they would like to sell? If I build one I would be happy to pass it on to someone else when I'm done testing and perhaps fixing.

Best Regards,
John

[Bernard Fife]

John,

CWP is the crank wheel pulser program here: http://www.megamanual.com/router/crankwheelpulser.htm However, you can also use the stim's tach signal for the VSS, just use a jumper lead to connect the VSS input to the appropriate spot on your stim. For the 2.2 stim, you can jumper from the end of R4 furthest from the DB37 to the VSS input on Amp pin #2. (Be sure to turn off RPM checking (http://www.msgpio.com/manuals/mshift/V2tune.html#gr), or the controller won't shift most of the time when used on the stim like this where the rpm isn't tied to the vehicle speed.)

The VSS shouldn't matter for the Input1/2/3 status (otherwise the controller wouldn't be able to tell which gear you were in when you were stopped).

Input2 is on processor pin AD1, and this has to be jumpered at the 25x2 header to the GPI2 input circuit location (still on the 25x2 header). So that's the first thing I would check (that the jumper is in place as specified in the build manual).

If that is okay, I would check R22 to see if there is 5 Volts on this resistor (when the Amp pin #6 is not grounded there should be 5V on both sides of this resistor, when Amp#6 is grounded there will only be 5V on one side of the resistor, the other side will be near ground level voltage).

If you never get 5V on this resistor, check that it is well soldered.

If you get 5V, then check the resistance from both sides of R22 directly to the Amp pin #6 (you can do this by counting the 6th pin coming up from the PCB into the connector from the copyright edge of the board). It should read ~330 Ohms on one side, and less than 1 Ohm on the other.

Make sure the diode in R30 is the right way around - the banded end should be furthest from the Ampseal end of the board.

Check the resistance across C16 with Amp pin #6 not grounded - it should be high (a thousand Ohms at least). Make sure nothing is installed in C17.

Finally, check the resistance from both sides of R22 to ground (use the middle pin of the voltage regulator as a ground point). If Amp pin #6 is not grounded, this should be 'infinite'.

Let us now what you find - we will help you get it sorted.

To PM me, click on my username, and then you should find a "send private message" link.

Lance.

[jdrapersys]

Very many thanks Lance,

It was pilot error on my part. I had all three zener diodes in R28, R30 and R36 facing the same way. So now all three inputs work, as does paddle shift down and brake sense. I do get a speedometer reading when I use a tach signal to AMP2, but very low for the RPMs. I had to go to over 16,000 to get to 40mph. Was that because I was in 1st gear? I thought the VSS was independent of tach in TunerStudio.

I can not, however, get paddle shift up to work. If I understand correctly, the first component in the VR2 section that should be connecting to AM15 is R60 as the Ra of the VR2 sub-circuit. I can get no zero or near zero resistance connection from AM15 to R60 or for that matter any location in the VR2 sub-circuit area. Nor can I get a near zero resistance connection from R60 to any of the AMP connectors. I do get what appear to be appropriate connections from the CPU to the components in the VR2 section.

I'm still posting to this thread because, I discovered, I have not been a member of the forum long enough to send private messages.

Very best,
John

[Bernard Fife]

John,

You should be getting very close to the number of posts required to PM. In any case, it's good to have these discussions out in the open, as many other people will likely have the same questions at some point.

The tach signal is a bit slow for the VSS. To get 60 mph with the default values, you need ~96000 pulses per minute, or about the equivalent of about 24000 rpm on a V8. You can change the number of teeth temporarily in the VSS setup to be 10 or so (from the default of 40), then you will be able to get much higher speeds in bench testing.

Normally the tach is related to the VSS by the gear ratios and number of VSS teeth when if gear (but not when in neutral or if the clutch is in). However, if you take both signals from the same source (like the stim) then they will be proportional at all times (even when they wouldn't be in a vehicle - like before/after a shift).

As you noted, the upshift button is on VR2, and this connects to processor port pin PT6 (and Amp pin #15).

For use with an upshift button, we basically need a circuit with a pull-up voltage, and pull-up resistor to limit the current flow when the button is closed (grounded). In the recommended 4L60E build configuration, there are:

- C31/R48 - 330 Ohm resistor, this is the pullup resistor connected to Vref (5V from the voltage regulator). There should be 5 Volts on both sides if the shift button is open, and 5V on one side and near ground on the other if the button is closed to ground the output pin (Amp pin #15). If there isn't 5V at any time while powered up, then check that the R48 end is well soldered. If there is 5V at all times, even when the circuit is grounded by the up shift button, then check the jumpers listed below.
- R60 - jumpered. If this jumper is missed, the switch button signal doesn't reach the circuit from the Ampseal pin #15. So make sure it is in place and well soldered.
- R49/Q15 - jumper. If this jumper is missed, the signal doesn't reach the 25x2 header or processor pin. Note that Q15 is the small transistor pad in the 'bottom left' corner of the VR2 circuit area (just the the right of VR1's R59). So check it carefully.
- R38 - 1.0K Ohm resistor (this is the resistor that leads to the processor pin). This must be in place, or jumpered, or the signal can't reach the 25x2 header or processor. Either side of this resistor should read 5V with the button 'open' and ~0V with the circuit grounded by the upshift button. This is the resistance you measure between the 25x2 header and the ampseal pin #15.
- D24 - 5.6 Volt Zener diode to prevent voltage spikes from reaching the processor. It is not essential, but will cause problems if it is defective or backwards.
- C30 - 0.1uF capacitor to smooth the spikes from the switch opening and closing. It is not essential. But check the resistance across it to make sure it is not shorted. It should measure at least several hundred Ohms across its two leads.

So I would check each of these very carefully against the instructions. It is not impossible that there are errors in the instructions, but they have been used many times, and I have just gone over them again on a new board and things look okay to me. So have a look and let us know what you find.

Lance.