GPIO gets hot, VERY hot (FIXED)

'Microsquirt' Transmission Control development and support.

Moderators: jsmcortina, muythaibxr

Post Reply
Philip Lochner
Super MS/Extra'er
Posts: 1004
Joined: Thu Feb 16, 2006 6:18 am
Location: George, South Africa

GPIO gets hot, VERY hot (FIXED)

Post by Philip Lochner »

I've now done 4600km with my GPIO / TransControl 1.0.1 setup and shift behaviour is 99%. That missing 1% relates to a downshift into 2nd which I'll address later.

What worries me is that my GPIO gets very, very hot. So hot that I can not put my fingers for more than 1sec on the large output transistors mounted on the heat sink. But even then everything works as it should.

I just do not believe that this is how it should be, should it?

So today I investigated, and found two mistakes arising from having built my GPIO from MShift instructions and now running TransControl:
1) PWM4 was linked to PT1 (and not PM3 as per TranControl supplement);
2) VB1 was linked to PM4 (and not PT7 as per TransControl supplement).
Despite these errors, everything was working as it should - except that GPIO was getting very hot)

Having corrected the above, nothing seems to have changed.

It seems to me as if the 3-2 transistor(PWM4) and the LUF transistor (VB1) are the main culprits (their screws burn me most) as to generating the heat.

Any ideas?

PS: Wanted to upload msq and datalog but the site would not let me... Nothing happens when I click on "Browse"
Last edited by Philip Lochner on Sun Dec 27, 2015 10:56 am, edited 1 time in total.
Kind regards
Philip
'74 Jensen Interceptor 440ci (EFI'ed with MS2 and wasted spark + GM 4L60e GPIO controlled - both on Extra FW)
jsmcortina
Site Admin
Posts: 39585
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK
Contact:

Re: GPIO gets hot, VERY hot

Post by jsmcortina »

What transistors are you using there? They could be in linear mode somehow.

Do you have 1N4001 flyback diodes in place across the solenoids? It could be flyback energy dissipating in the transistors.

James
I can repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: http://www.msextra.com/forums/viewtopic ... 04&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
New users, please read the "Forum Help Page".
Philip Lochner
Super MS/Extra'er
Posts: 1004
Joined: Thu Feb 16, 2006 6:18 am
Location: George, South Africa

Re: GPIO gets hot, VERY hot

Post by Philip Lochner »

jsmcortina wrote:What transistors are you using there? They could be in linear mode somehow.

Do you have 1N4001 flyback diodes in place across the solenoids? It could be flyback energy dissipating in the transistors.

James
Thanks for the quick reply!

The PC (PWM3) and LUF (VB1) trannies have flyback diodes mounted on the solder side of the board. I mounted these not knowing if the solenoids have built-in diodes. (SOME of the solenoids in my tranmission definitely have built-in flyback diodes - not sure which have and which dont) Should I remove the the flyback diodes on the back of the GPIO board?

3-2 (PWM4) = TIP120
PC (PWM3) = 302 B18 EK106018 ??? (Flyback diode)
LUF (VB1) = TIP120 (Flyback diode)

BTW, the PC circuit also has the 5 Ohm resistor in circuit as per the Mshift instructions.
Kind regards
Philip
'74 Jensen Interceptor 440ci (EFI'ed with MS2 and wasted spark + GM 4L60e GPIO controlled - both on Extra FW)
jsmcortina
Site Admin
Posts: 39585
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK
Contact:

Re: GPIO gets hot, VERY hot

Post by jsmcortina »

What flyback diode and how are they wired?
What value base resistor are you using on the transistors?

James
I can repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: http://www.msextra.com/forums/viewtopic ... 04&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
New users, please read the "Forum Help Page".
Philip Lochner
Super MS/Extra'er
Posts: 1004
Joined: Thu Feb 16, 2006 6:18 am
Location: George, South Africa

Re: GPIO gets hot, VERY hot

Post by Philip Lochner »

jsmcortina wrote:What flyback diode and how are they wired?
What value base resistor are you using on the transistors?
1N4001 soldered onto pin 11 and 33 of the AMP connector pointing to the 12V feed on pin 1

PWM3 = 0 Ohm (Jumpered)
PWM4 = 1k Ohm
VB1 = 1k Om
Kind regards
Philip
'74 Jensen Interceptor 440ci (EFI'ed with MS2 and wasted spark + GM 4L60e GPIO controlled - both on Extra FW)
Philip Lochner
Super MS/Extra'er
Posts: 1004
Joined: Thu Feb 16, 2006 6:18 am
Location: George, South Africa

Re: GPIO gets hot, VERY hot (FIXED)

Post by Philip Lochner »

Fixed the issue as follows:
1) Increased the value of the PC current limit resistor 3 fold from 4.7Ohm to 15Ohm (no perceptible impact on clutch pressures);
2) Increased the line pressures in 3rd and 4th to 99% at all values of TPS and made them as high as possible in 1st and 2nd (the higher the line pressure %, the lower the on-time of the PC solenoid, the less heat into the resistor)

So, in principle, one should tune for the highest line pressures you find acceptable, not the lowest. (I was tuning for the lowest)

Now that I realise how much heat the line pressure graphs cause at low values of line pressure, I might even try the original 4.7Ohm resistor just to see if it does not now run at acceptable heat levels.
Kind regards
Philip
'74 Jensen Interceptor 440ci (EFI'ed with MS2 and wasted spark + GM 4L60e GPIO controlled - both on Extra FW)
jsmcortina
Site Admin
Posts: 39585
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK
Contact:

Re: GPIO gets hot, VERY hot (FIXED)

Post by jsmcortina »

It's not so great that the transistor driving your pressure control solenoid is unknown!
I'm wondering if whatever it is, the Rds(on) is quite high causing resistive heating. When using the Microsquirt, one of the VND5N07 is used for the PC solenoid, this has an Rds(on) of 0.2 ohms.

James
I can repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: http://www.msextra.com/forums/viewtopic ... 04&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
New users, please read the "Forum Help Page".
Philip Lochner
Super MS/Extra'er
Posts: 1004
Joined: Thu Feb 16, 2006 6:18 am
Location: George, South Africa

Re: GPIO gets hot, VERY hot (FIXED)

Post by Philip Lochner »

jsmcortina wrote:It's not so great that the transistor driving your pressure control solenoid is unknown!
I'm wondering if whatever it is, the Rds(on) is quite high causing resistive heating. When using the Microsquirt, one of the VND5N07 is used for the PC solenoid, this has an Rds(on) of 0.2 ohms.

James
James, the heat I was getting comes exclusively from the serial resistor to be wired in series with the PC output as per the GPIO build instructions for MShift. Once I moved this resistor outside the GPIO box, the driver transistors became barely luke warm and stayed that way.

I had mounted this resistor on the solder side of the PCB, directly underneath the heat sink and it was the resistor heating up the heat sink, not the trannies themselves.

The transistor being used in Q3 for the PC circuit is IRLZ44ZPBF-ND.
Kind regards
Philip
'74 Jensen Interceptor 440ci (EFI'ed with MS2 and wasted spark + GM 4L60e GPIO controlled - both on Extra FW)
Raymond_B
Super MS/Extra'er
Posts: 1394
Joined: Thu Mar 06, 2014 2:17 pm
Location: Texas
Contact:

Re: GPIO gets hot, VERY hot (FIXED)

Post by Raymond_B »

Philip Lochner wrote:Fixed the issue as follows:
1) Increased the value of the PC current limit resistor 3 fold from 4.7Ohm to 15Ohm (no perceptible impact on clutch pressures);
2) Increased the line pressures in 3rd and 4th to 99% at all values of TPS and made them as high as possible in 1st and 2nd (the higher the line pressure %, the lower the on-time of the PC solenoid, the less heat into the resistor)

So, in principle, one should tune for the highest line pressures you find acceptable, not the lowest. (I was tuning for the lowest)

Now that I realise how much heat the line pressure graphs cause at low values of line pressure, I might even try the original 4.7Ohm resistor just to see if it does not now run at acceptable heat levels.
I'm no transmission pro, but are you sure it's good to run really high line pressures all the time? And I wonder about of fixing heat in the electronics with tuning parameters, that just doesn't seam right.
1995 Ford Lightning. Dart based 427 Windsor, Novi 2000, full sequential, E-85, etc. MS3X/v3.57
http://www.buildpics.org/
jsmcortina
Site Admin
Posts: 39585
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK
Contact:

Re: GPIO gets hot, VERY hot (FIXED)

Post by jsmcortina »

Is the power resistor really needed?

James
I can repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: http://www.msextra.com/forums/viewtopic ... 04&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
New users, please read the "Forum Help Page".
Philip Lochner
Super MS/Extra'er
Posts: 1004
Joined: Thu Feb 16, 2006 6:18 am
Location: George, South Africa

Re: GPIO gets hot, VERY hot (FIXED)

Post by Philip Lochner »

Raymond_B wrote:I'm no transmission pro, but are you sure it's good to run really high line pressures all the time? And I wonder about of fixing heat in the electronics with tuning parameters, that just doesn't seam right.
I agree. I was tuning for the lowest pressures I could get away with as doing so would surely provide the best chances for not springing oil leaks or bursting pipes etc. Also of course to get the smoothest shift action. But to get lower pressures requires higher duty cycles on the pressure control solenoid which creates more heat into this resistor. So, I am forced to choose between two evils... high pressures or lots of heat dissipation into this resistor.

I thank God that I did not mount this resistor in the wiring loom as this would surely have destroyed the transmission wiring harness or possibly even caused my car to burn.
Kind regards
Philip
'74 Jensen Interceptor 440ci (EFI'ed with MS2 and wasted spark + GM 4L60e GPIO controlled - both on Extra FW)
Philip Lochner
Super MS/Extra'er
Posts: 1004
Joined: Thu Feb 16, 2006 6:18 am
Location: George, South Africa

Re: GPIO gets hot, VERY hot (FIXED)

Post by Philip Lochner »

jsmcortina wrote:Is the power resistor really needed?

James
I did remove it briefly but could not really test the outcome as doing so, I believe, resulted in so much RMI that it corrupted the VSS signal, and hence I could not drive the car. My VSS setup is that the VSS sensor output goes to a TP500 and the 12V square wave TP500 output then spilts to GPIO and the speedo. I could thus see by the speedo behaviour (and on TS screen) that even the TP500 was affected by the RMI. This prompted me to re-install the resistor but with higher values to reduce the current (and hence heat) in them.

With Microsquirt not having this current limit resistor, it is clear that its not really needed for control of the PC circuit (provided duty cycles are kept low enough) but it seems needed for RMI reasons (at least in the case of GPIO) I've noticed that others also have VSS issues. Now I'm wondering if their VSS issues are not compliments of PC RMI....

The PC solenoid happens to be the one with the lowest resistance (4-5Ohm) of all the solenoids in the 4L60e. I do have an inductance meter so I could measure the solenoid inductances as well. What are the chances that PC also has the lowest inductance...?
Kind regards
Philip
'74 Jensen Interceptor 440ci (EFI'ed with MS2 and wasted spark + GM 4L60e GPIO controlled - both on Extra FW)
Philip Lochner
Super MS/Extra'er
Posts: 1004
Joined: Thu Feb 16, 2006 6:18 am
Location: George, South Africa

Re: GPIO gets hot, VERY hot (FIXED)

Post by Philip Lochner »

Philip Lochner wrote:Fixed the issue as follows:
1) Increased the value of the PC current limit resistor 3 fold from 4.7Ohm to 15Ohm (no perceptible impact on clutch pressures);
I believe I HAVE found an effect of having 15Ohm rather than 4.7Ohm. When shifting into 1st from N, with 1% line pressure commanded, the engagement into 1st is notably more "assertive" confirming that physical line pressures are actually higher than when a 4.7Ohm resistor is fitted despite the 1% being commanded. I'm rather certain that shifts into 2nd,3rd and 4th are also more aggressive but not so much to be an issue.

The rather hard shift into 1st (from N) is not acceptable though, so I've decided to rather try two 15Ohm / 5W resistors in parallel, giving an effective 7.5Ohm resistor of 10W. Will report on this as soon as I've driven the car again.
Kind regards
Philip
'74 Jensen Interceptor 440ci (EFI'ed with MS2 and wasted spark + GM 4L60e GPIO controlled - both on Extra FW)
Matt Cramer
Super MS/Extra'er
Posts: 17499
Joined: Thu Apr 16, 2009 8:08 pm

Re: GPIO gets hot, VERY hot (FIXED)

Post by Matt Cramer »

jsmcortina wrote:Is the power resistor really needed?

James
Probably not with the PRC driver transistor we use in our current MegaShift kits; it is rated at 50 amps.
Matt Cramer -1966 Dodge Dart slant six running on MS3X
jsmcortina
Site Admin
Posts: 39585
Joined: Mon May 03, 2004 1:34 am
Location: Birmingham, UK
Contact:

Re: GPIO gets hot, VERY hot (FIXED)

Post by jsmcortina »

If the document linked is correct, then the maximum resistive current a standard pressure control solenoid should take is maybe 4.5A.

James
I can repair or upgrade Megasquirts in UK. http://www.jamesmurrayengineering.co.uk

My Success story: http://www.msextra.com/forums/viewtopic ... 04&t=34277
MSEXTRA documentation at: http://www.msextra.com/doc/index.html
New users, please read the "Forum Help Page".
Philip Lochner
Super MS/Extra'er
Posts: 1004
Joined: Thu Feb 16, 2006 6:18 am
Location: George, South Africa

Re: GPIO gets hot, VERY hot (FIXED)

Post by Philip Lochner »

Philip Lochner wrote:
I believe I HAVE found an effect of having 15Ohm rather than 4.7Ohm. When shifting into 1st from N, with 1% line pressure commanded, the engagement into 1st is notably more "assertive" confirming that physical line pressures are actually higher than when a 4.7Ohm resistor is fitted despite the 1% being commanded. I'm rather certain that shifts into 2nd,3rd and 4th are also more aggressive but not so much to be an issue.

The rather hard shift into 1st (from N) is not acceptable though, so I've decided to rather try two 15Ohm / 5W resistors in parallel, giving an effective 7.5Ohm resistor of 10W. Will report on this as soon as I've driven the car again.
With the 7.5Ohm I now have a gentle Neutral - Drive (1st) engagement. Did not drive much today but it seems GPIO now only gets luke warm.
Kind regards
Philip
'74 Jensen Interceptor 440ci (EFI'ed with MS2 and wasted spark + GM 4L60e GPIO controlled - both on Extra FW)
mill383
Helpful MS/Extra'er
Posts: 134
Joined: Thu Nov 26, 2009 5:02 pm
Location: Indianapolis, IN

Re: GPIO gets hot, VERY hot (FIXED)

Post by mill383 »

I was wondering the same thing, about needing the 4.7ohm 5W resistor as called out in the MegaShift documentation. TCU documentation is silent on this interface.

I have a MSGPIO with MegaShift SW on a 4L60E functional in a Camaro. I am working on putting a microSquirt V3 with TCU SW in a different car, am just in the early stages of bench testing things. In the Camaro, I have the 4.7ohm power resistor mounted externally from the MSGPIO hardware, but have never checked its temperature.

Something to remember with the PC circuit is that it is a fail-safe design, meaning if power is lost to the PC solenoid, it goes to full line pressure. For minimum line pressure, the Solenoid needs 1 amp of current through its coil. The Service manual says, "The applied current can vary from 0.1 to 1.1 amp." So during low TPS settings, putting around, the PC current will be high, near the 1 amp limit. Both MSShift and TCU software invert this relationship in the tuning SW. i.e, when the desired Line Pressure curve is near 100% (firm shifting), actual current to the PC solenoid is near zero amps. Conversely, when putting around, PC current needs to be around 1 amp. If you have the 4.7 ohm resistor in series with the solenoid, then in the tuning SW you will have a low % Line pressure command.

Of interesting note, the lowest point in the TCU's baseline command line pressure curve is 60%. I suspect this 60% PWM duty cycle (inverse would be 40% PWM to the sol) is to keep the current to the PC solenoid in-check. If you were to reduce this 60% to 10% or 0%, you could potentially over-current the PC solenoid. Some quick math, 1amp through the sol, that is about 8.5 Voltage drop across the coil. If you were to connect a solid 13.5V across the coil, then current through it would be around 1.7 amps.

So I am a bit surprised that the current TCU .ini file allows you to command values lower than 60%, without some kind of potential over-current warning. Or include a statement in the help for that table why going below 60% is not advisable. Maybe it's not so much an over-current concern (the PC solenoid is cooled in trans fluid), but a matter of controlability. Going beyond 1 amp, there will be no change in line pressure.

Bottom line is you need to drop about 5V at 1 amp (5W) when the PC sol is in full-soft setting. This can be done with a 4.7 ohm power resistor (as documented in the MSShift manual) or by PWMing the driver FETT to keep the current limited to 1 amp. I suspect this is what is happening with the implementation of TCU SW with microsquirt HW. I suspect INJ2 FETT might get a bit toasty. I'll be monitoring PC solenoid current with an ampmeter when I get it wired up and INJ2 FETT temp.

Dave
Dave
1958 Cushman scooter Microsquirt'ed and turbo-ed
1994 Camaro MS3Pro and GPIO MShift
1996 Buick Roadmaster wagon, MS3Pro and uV3 TCU
Post Reply