V2.00 board errata

This forum is for general questions you have on assembling and troubleshooting your GPIO board. You can purchase a GPIO board at DIYAutoTune now!
Bernard Fife
Posts: 1696
Joined: Fri Apr 04, 2008 1:28 pm

Re: V2.00 board errata

Post by Bernard Fife »

All,

I found a problem in the build instructions and BOM for those targeting the 4L60E. You need to add 330 Ohm pull-up resistors in R22, R84, and R85. This isn't done if you are using a pot style manual gear lever sensor (ex. Ford), but it needs to be done for digital style inputs (like the 4L60E has). The change comes about because the pins were switched from digital inputs with internal pull-ups to analog-digital converters (with no internal pull-up - so an external pull-up is needed).

I have fixed the build guide and BOM. I will add a couple of more small changes in a few minutes, so watch this thread.

----> additions below <------

FOR ALL APPLICATIONS: Pull-ups should be added to the shift buttons as well. For the Upshift button on PT6/VR2, the circuit needs to be reconfigured a bit:

- remove R60 and jumper it with a snipped off lead,
- install a 1.0K Ohm resistor in R38,
- remove the jumper on C31,
- install a 330 Ohm resistor from the end of R48 nearest the heat sink, to the hole for C31 nearest the Ampseal connector.

For th Downshift button, install a 330 Ohm resistor in R25 (no other changes needed).

------> more additions <-----

ALL APLICATIONS: R27 (GPI2/4WD), R29 (GPI2/swB), R67 (EGT3/swC), and R70 (EGT4/swA) should not be installed, they should be jumpered instead. These are on digital inputs for ciurcuits that ground to pull low. The resistor in-line could affect the inputs at the processsor ports in some conditions, so they are best left out (EGT3 should be configured using the manual - http://www.megamanual.com/gpio/egt.htm - if it is not used for manual gear lever detertmination).

----> more <----

ALL APPLICATIONS: Install jumpers made from snipped off leads on R9, R10, R11, R12 (on the PWMx circuits) otherwise they won't be connected to ground.

I will add these to the manual as well.

Lance.
"Never wrestle with pigs. You both get dirty and the pig likes it." - George Bernard Shaw
Bernard Fife
Posts: 1696
Joined: Fri Apr 04, 2008 1:28 pm

Re: V2.00 board errata

Post by Bernard Fife »

Marc noticed that there was an error in the above - The copmponent labeled R38 should have been R48 (now corrected). Thanks Marc! (The build guide was correct.)
"Never wrestle with pigs. You both get dirty and the pig likes it." - George Bernard Shaw
Bernard Fife
Posts: 1696
Joined: Fri Apr 04, 2008 1:28 pm

Re: V2.00 board errata

Post by Bernard Fife »

Marc has caught another error. There are two tables listing the Amp connections in the MShift info. One showed swB as pin 6 (which is correct), the other listed swB as pin 5 (which is actually the 2/4WD switch). The wiring diagram was also incorrect as swA should go to pin 26, not pin 5. The build instructions are not affected, but if you used the defective table to wire your transmission, some wiring may need to be re-done (though you might have already noticed that both swB and 2/4WD were assigned to the same pin).

Lance.
"Never wrestle with pigs. You both get dirty and the pig likes it." - George Bernard Shaw
Bernard Fife
Posts: 1696
Joined: Fri Apr 04, 2008 1:28 pm

Re: V2.00 board errata

Post by Bernard Fife »

All,

Marc has pointed out some input circuits had inadequate protection. In most cases this was due to the removal of the 1.0K resistors inline with the processor pin necessitated by the internal pull-ups (see above) which cannot have significant resistance in the path or the pin will never go low. The 5.6 Volt Zener diodes still provide some 'spike' protection, but would not help for very long if the circuit was wired to 12 Volts, for example. In the most recent code (not posted), I have disabled the internal pull-ups, and restored the 1.0K resistors, which will now limit the current through the Zeners to well below their 1.0 Watt rating. Since the pull-up voltage is applied 'downstream' of the 1.0K resistors, the voltage at the pins is not affected.

The changes are:

- add 1.0K resistors to R27, R29, R38, R67, R70,
- add 330 Ohm external pull-up resistors in R21, R22,
- add 5.6 Volt Zener diodes in:
-- R28 (banded end towards Ampseal),
-- R30 (banded end away from Ampseal),
-- R36 (banded end towards Ampseal).
- move 5.6 Volt Zener diodes:
-- from C40 to C39 (banded end toward heat sink),
-- from C42 to C41 (banded end toward heat sink),
-- from C44 to C43 (banded end toward heat sink),
-- from C46 to C45 (banded end toward heat sink).

On VR2 (for the Up Shift button), it's easier just to start from scratch to describe this. You want:
- jumper R48 to C31 with a 330 Ohm resistor from the hole for C31 (the "1" in C31 is partly obscured by the silkscreen for R50) closest the Ampseal connector to the hole for R48 closest the heat sink. This is the pull up resistor,
- 1.0K Ohm resistor in R38,
- 5.6 Volt Zener diode in D24 (to match silkscreen),
- jumper empty location R49 (hole nearest the heat sink) to Q15 (to the pin nearest heatsink),
- jumper R60.
(Some of these are already in place if you followed the 'old' instructions.)

I have made these changes to the build guide: http://www.msgpio.com/manuals/mshift/assembly.html

Finally, the VR3 circuit is not currently used. It was reserved for the mode switch button (from auto to manual and back). However, this is not currently used in the code, since you can accomplish the same thing using just the shift buttons - push either one at any time to go into manual mode (and perform the shift), push both to go into automatic mode. So what we do with this input depends on what people want - we can certainly add a mode button, or we can reserve this for something like an input shaft speed sensor (like the 4L80E has).

Lance.
"Never wrestle with pigs. You both get dirty and the pig likes it." - George Bernard Shaw
Post Reply