AA80E 8-speed

[Jim Blackwood]

That looks pretty good Marty. Hopefully it fits in with what we have already on the spread sheet.

Jim

[Jim Blackwood]

Marty, as far as I can tell your circuit should work fine provided the controller is programmed to accept it. I have no problem with it whatsoever.

We need to sort out the solenoid circuit assignments. I understand the transmission functions pretty well but the inner workings of the MegaShift, not so much. So I will attempt to make selections and hopefully Lance, you can make changes as needed to allow the MS to work properly.

We have nine shifting solenoids.
-Two are standard ON/off solenoids equipped with three way valves, SL and SR.
-Two are standard Linear (PWM) solenoids, SLU and SLT.
-Five are special High Flow Linear (PWM) solenoids which are physically much larger for higher flow rates and faster response times. These five directly control the four clutches and one brake band that shift the forward gears. The order of use in gear changes, from low to high is SL1, SL5 (for brake B1), SL3, SL4, and SL2. I would consider the lower gears more critical for timing and fine control, and would assign the drivers accordingly, absent other considerations.

The seven linear solenoids all have a resistance of 5.0 to 5.6 ohms and a pulse rate of 350 Hz. I do not know if there is any advantage to varying the pulse rate, Marty do you have any more information on that? In the absence of such maybe we should allow for varying the pulse rate if possible.

-SLT inversely controls line pressure. As I understand it, this gives overall control of shift characteristics while the individual solenoids SL1 through SL5 give finer individual shift control.
-SLU controls either the lock-up clutch OR the reverse brake B2, depending on the state of solenoid SR. Lock-up is available for 2nd through 8th gears. It would probably work in 1st as well but I doubt that would not be a good idea.

The standard solenoids have a resistance of 11 to 15 ohms:
-SL selects either the brake B2 or the TCC to be controlled by solenoid SLU.
-SR can be thought of as a trigger that initiates gear changes. On every gear change it diverts flow from the clutch apply control valve to the sequence control valve, and then back to the clutch apply control valve. I am not sure if this includes reverse or not, but suspect that it does.

So, referring to:
http://www.msgpio.com/manuals/mshift/spareport.html
It looks like we must use every available solenoid driver and then still need two on/off drivers if a speedometer output is enabled.

I have one immediate question. If PAD01 and PAD03 are used to drive PWM solenoids it says a voltage based manual shift position indicator must be used. Is that because we only have one input, or some other reason? Because we only need two inputs to use a switch based input, one for reverse and one for drive. Neutral would be the default. Seems like that would simplify things but I do not fully understand how this works. The fact that we need two more drivers for the switching solenoids is one more complication. I had hoped to use AD0 and AD3 for SL and SR but it looks like that won't work. So on this part of it Lance, I hope you have some ideas because I've got nothing.

I think everything else pretty much remains the same as in other applications, so on to the solenoid assignments. Please note, I am less than intimately familiar with the workings of the MegaShift so to some degree am flying blind here and anyone is very welcome to make helpful suggestions. I'm certain to overlook a few things.

SLT- Line pressure: Output 7
SLU- TCC/Rev: Output 3
SL1- Clutch C1: Output 2
SL5- Brake B1: Output 4
SL3- Clutch C3: Output 1
SL4- Clutch C4: Output 5
SL2- Clutch C2: Output 6

Which leaves Output 8 for speedometer drive, provided circuits can be found for the two relays below.

SL- TCC/Rev selector:
SR- Shift trigger:

That's about as far as I can take it guys. I hope you have some ideas.

Jim

[Bernard Fife]

If PAD01 and PAD03 are used to drive PWM solenoids it says a voltage based manual shift position indicator must be used. Is that because we only have one input, or some other reason?

Jim,

It is because those two outputs are normally assigned as digital inputs for determining the manual gear lever position. If you use them as solenoid outputs, they can't simultaneously be used as digital inputs.

Lance.

[Jim Blackwood]

Thanks Lance, I understand that part but aren't there four inputs for the shifter position, PRND? So if we take away P and N it seems we would still have R and D for signal high. With both inputs low wouldn't a default to neutral be workable?

That doesn't address the last two solenoids we need to drive but it gets us the simple switch arrangement on the shifter provided it works.

Jim

[Bernard Fife]

With both inputs low wouldn't a default to neutral be workable?

Jim.

Possibly. I have never seen an electronically controlled trans with just two digital inputs. That only gives you 4 positions, but these could be PNRD, and you can shift to other forward gears using the shift buttons, so it could work. The code doesn't allow it at the moment, though, it must use:
- the first three digital inputs,
- all four digital inputs, or
- the voltage based input.

However, it usually fairly easy to use a few resistors to make a voltage based input out of multiple digital inputs that connect to ground using a few resistors and a pull-up. See: http://www.msgpio.com/manuals/mshift/voltlever.html If there's is no way to make that work, and you make progress n the other issues, I will consider altering the V4 code to use just inputs 1 and 4 as a user option.

BTW, the pin that the 4th input connects to can only be used as an input (due to the microprocessor design) so it is not available to drive a solenoid.

Lance.

[Jim Blackwood]

Good to know. Can the 4th input be used as the voltage based one? If so that would leave one pin for the last 2 solenoids and if push comes to shove maybe we can use the speedo driver for the other one (meaning a GPS speedo or one that directly reads output shaft pulses would be required, or one with a shaft mounted pick-up). Not optimal maybe but at least a possible solution. I'll study it some more to see if I can suggest any other options but will this work? It would be nice to at least have a fall-back plan. Hopefully we can then improve on it.

There is no reason a voltage based switch won't work. We have the rotary switch and only need to wire in the precision resistors. Do we have to have a minimum of 3 positions/voltages/resistors? (RND, park is redundant) or can reverse or neutral be grounded? I figure the fewer resistors, the wider the gaps between them can be, making it more stable.

Do the solenoid assignments look OK to you?

Looks like we may be getting there. Plus I've got a samples request in at the company that makes the main transmission connector. That could be helpful for anyone else using this transmission.

Thanks,
Jim

[Bernard Fife]

Can the 4th input be used as the voltage based one?

Jim,

No, it is a digital only input (or it can be a 'maskable interrupt'). It would have to be an Analog-Digital Conversion (ADC) pin to do anything with measuring voltage.

Lance.

[Jim Blackwood]

Because switching solenoid SL is either ON in Reverse or Drive and off the rest of the time, looks to me like it can be controlled directly by the manual shifter position switch, something I think Lance hinted at.

SR still needs a driver if we are to have a speedo output. Still reviewing.

We need current ratings for:
SL/SR: (if about 1 amp direct 12v can be applied)
SL1-SL5:
SLT/SLU:

This will allow us to set the PWM percentages.

The trigger solenoid SR cycles ON/off with each shift. Does that mean it has to be assigned to a pin with the "Clutch" function?

Jim

[Jim Blackwood]

Been working on the board build a bit, just at the stage to hook it up and check the processor and of course now I have to come up with a usable serial cable. I jumpered the Tx/Rx to the AMP plug so I either have to put the connector together or rig some jumpers for testing and some sort of cable, but happily, with a new bios battery my tuning laptop is up and running so it should just take some updating and downloading to get that going and I can get back to work on the build. That should be illuminating. I'm already seeing the logic behind the standard circuit assignments with simple jumpers rather than jumper wires. We may want to move some things.

Also, I have been working on getting a contact in the Aisin transmission engineering department. Once I have a good connection I should be able to get the rest of the info we need.

Jim

[bondango]

Jim my board has just been released by customs, i collect this week so will have to touch base on the build up.

Marty