Standard Calibrations no longer work

[dwizle]

the standard calibrations don't work any more Map tps Bat voltage temp sensors
I can play with the calibration and bring Map/Baro and Bat V inline but Temp sensors
Is the The HC9S12C64 Processor going bad? or is F1 Polyfuse RXE050 going bad? 5.1v in 4.75 out

'86 944 na Porsche
MS2 Ver 3.0 ms2extra-pre3.4.3beta firmware Batch fire Injection, ICV PWM, VR trigger Single coil, 2 MPX4250A senors for Alt correction
Injectors: 4 Bosch 0280155868 flows 375 cc/min at a tested pressure of 3 bar, or 36 lbs/hr at 43.5 psi.
flows 342 cc/min at a tested pressure of 2.5bar, or 33 lbs/hr at 36.2594 psi.
Latency (dead time)@ 12volts=0.621 ms @ 13volts=0.506 ms @ 14volts=0.391 ms.
FUEL NOZZLE: 4-HOLE DISC - Ultimate Atomization High-impedance/high-resistance/14.0 Ohms
IAT Sensor : BWD WT3051 98 99 00 01 Mustang 4.6
CLT Sensor: Stock Bosch 944 sensor
Wide Band O2 Sensor Contoller: Spartan 2 OEM from 14Point7 http://www.14point7.com/products/spartan-2-oem
O2 Sensor: Bosch LSU 4.2 Wideband O2 Sensor
Idle Control Valve: Bosch 2 wire Idle Control Valve from a Volvo p/n GEGT7610-324

[142 guy]

Standard calibrations? Do you mean previous calibrations that worked and now the calibrations are no longer valid?

First observation is that the battery voltage measurement is taken from the sanctioned 12 volts on the MS V3 Board. Problems with the polyfuse F1 or the 5 v supply will not affect the Vbat calibration. A 0.35 volt drop across the polyfuse does seem a bit much. Did you determine that by measuring the voltage on each side of the polyfuse to ground? If so, I would re check by measuring with a DVM directly across the polyfuse. If you really do have 0.35v across the polyfuse, check the temperature of the LM293 and the polyfuse to see if they are really hot which might suggest a short circuit someplace.

My first inclination is that if you can still maintain serial communications with the processor and alter settings, its likely OK. My second inclination is that I would check your grounds on the MS. Measure the voltage from sanctioned ground on the MS V3 board to the vehicle chassis ground. It should measure 0 volts. The fact that both your Vbat and your other calibrations have a problem suggests that you might have some ground potential rise. If your LM293 and polyfuse are getting quite hot, it is possible that you have a short circuit somewhere that is drawing a lot of current through the regulator and polyfuse. This unplanned current flow through the ground could be causing a ground offset screwing up your calibrations. Check the grounds and Vdd points on the 40 pin socket for the daughtercard to make sure that they are OK. A problem with the ground connection for the daughtercard could cause ground offsets which would screw up your calibrations.

I fuse my MS board with a 2 amp ATO fuse which is a fast blow. If you are using a 2 amp fuse and you have good ground connections for the MS, normally the 2 amp fuse would not allow enough to flow through the MS power supply to create a significant offset between the MS circuit board and the vehicle chassis..

[dwizle]

Standard calibrations? Do you mean previous calibrations that worked and now the calibrations are no longer valid?

First observation is that the battery voltage measurement is taken from the sanctioned 12 volts on the MS V3 Board. Problems with the polyfuse F1 or the 5 v supply will not affect the Vbat calibration. A 0.35 volt drop across the polyfuse does seem a bit much. Did you determine that by measuring the voltage on each side of the polyfuse to ground? If so, I would re check by measuring with a DVM directly across the polyfuse. If you really do have 0.35v across the polyfuse, check the temperature of the LM293 and the polyfuse to see if they are really hot which might suggest a short circuit someplace.

My first inclination is that if you can still maintain serial communications with the processor and alter settings, its likely OK. My second inclination is that I would check your grounds on the MS. Measure the voltage from sanctioned ground on the MS V3 board to the vehicle chassis ground. It should measure 0 volts. The fact that both your Vbat and your other calibrations have a problem suggests that you might have some ground potential rise. If your LM293 and polyfuse are getting quite hot, it is possible that you have a short circuit somewhere that is drawing a lot of current through the regulator and polyfuse. This unplanned current flow through the ground could be causing a ground offset screwing up your calibrations. Check the grounds and Vdd points on the 40 pin socket for the daughtercard to make sure that they are OK. A problem with the ground connection for the daughtercard could cause ground offsets which would screw up your calibrations.

I fuse my MS board with a 2 amp ATO fuse which is a fast blow. If you are using a 2 amp fuse and you have good ground connections for the MS, normally the 2 amp fuse would not allow enough to flow through the MS power supply to create a significant offset between the MS circuit board and the vehicle chassis..

I mean the Calibration the come from the drop down window in TS.
I checked the voltage before and after the F1 5.01v on the out put leg of the LMnot sure of the part number I upgraded it to 1amp output since I was planing to ad a couple more 5v circuits
5.01v to the lm side of F1 and a current drop on the output. I am using a 550 watt computer power supply so it has regulated clean power going in...
when i san calibrations so for the Batt V TS says normally 0.0 and 29.7 gives accurate bat V that gives me a off the scale reading ...
I have 2 MPX4250 map sensors MPX4250 drop down gives off the scale reading played with the numbers to get them to read my densetity altitude 29.82 inhg the numbers are different for both sensors... So I know the numbers for my MAT/CLT must be way off but I can get the car to start n run and drive plus the O2 sensor must be off too..
The MS II gets power the same place where the porsche DME gets power plugs right into the factory harness...

I am thinking of going back to a later firmware to see what happens
But I will check the board grounds.. thanks for you insight

[jsmcortina]

A 0.35 volt drop across the polyfuse does seem a bit much.

Yes, that's too much and indicates a problem for sure.

OP - don't adjust the calibrations for the MAP sensors. Fix the problem and leave the correct original calibrations in place.
Check what voltage you see at TPS VREF. It needs to be 4.8 - 5.0V or something is badly wrong which will screw up all of your sensor readings.

What voltage regulator did you swap to? You must use a low dropout (LDO) type or you could run into problems during cold starting.

James

[dwizle]

think the Daughter board gave up the ghost
No more serial coms
HC9S12C64 Processor hot enough to burn my finger
LEDs instead of doing power up blink stay on for half a min or so

[jsmcortina]

Sounds like you've injected 12V into the processor. That is a sure fire way to break them.

James

[dwizle]

Sounds like you've injected 12V into the processor. That is a sure fire way to break them.

James

where would the 12v come from???
I used a MC7805CT from MOUSER
Linear Voltage Regulators 5V 1A Positive
Would replacing the MC9S12C64C fix it?

[jsmcortina]

where would the 12v come from???

Hard to guess. Measure the voltage at all pins and compare to the voltage diagram in the Hardware manual.

I used a MC7805CT from MOUSER

That's not LDO so could give problems with cold cranking.

Would replacing the MC9S12C64C fix it?

Until you fix the 12V injection you'll just keep breaking them. Plus, the chip needs to be loaded with the bootloader/monitor code over BDM. Best to send it off for repair.

James

[142 guy]

think the Daughter board gave up the ghost
No more serial coms
HC9S12C64 Processor hot enough to burn my finger
LEDs instead of doing power up blink stay on for half a min or so

Hot enough to burn your finger is probably a sign that the processor is damaged. However, on an off chance, power down, cool off and then power back up to see if it remains dead. However, before you do that, pull the daughter board out and power up the MS from whatever power supply you were using to check the voltages to the daughter card. No point in reheating the daughter card or toasting a second daughter card if the first is history.

Check the voltage on the 40 pin socket to make sure that you are getting 5 v on pins 1, 20 and 31. If you have the S12C to js9 jumper installed for an IAC you should get whatever your power supply voltage is on pin 16. Check to make sure that your ground continuity is good on pins 2, 19 and 32. With the daughter card out, you could also check for voltages to ground on all the other pins. Depending on whether the MS is on the bench or in the car and what is hooked up to it, you will see some voltages on the analog inputs. None of those voltage should be above 5 volts.

If your voltmeter can be set up to measure DC current, with it set to measure current connect the meter into the DC power supply to the Megasquirt. With my MS sitting on the bench and connected up to a lab power supply running at 12 volts supplied through a JimStim, the not spectacularly accurate meter on the power supply says that the MS is drawing a little over 0.1 amp. If you're MS is drawing much more current than that, you likely have a problem.

You also stated that you replaced the MS voltage regulator on the MS so that you could power up some extra 5 volt circuits. I don't know what you mean by extra circuits; but, that seems like an extraordinarily bad idea if you are using the 5 volt supply to power up things outside of the MS box. If you have other things that require 5 volts, fabricate a separate 5 volt supply for that stuff.

[142 guy]

One little tidbit of information with respect to the Vbatt calibration in TunerStudio. On the MS2 V 3.0 board the 12 v supply is measured using a voltage divider circuit. The design values for the two resistors in the divider circuit is 10 k and 50 k ohms. The default calibration works with these resistance values. I have tow V 3.0 boards from DIYAutotune and both of them came with what looks like a 51 k resistor and both resistors have a 5% tolerance. The aggregate error in my voltage divider circuit was a nudge over 5% so don't be surprised if your calibration is out. With the 51 k 5% resistor your worst case error could exceed 10%.

If I had been paying attention on assembly, I would have substituted suitable 1% resistors from Digikey.

[dwizle]

think the Daughter board gave up the ghost
No more serial coms
HC9S12C64 Processor hot enough to burn my finger
LEDs instead of doing power up blink stay on for half a min or so

Hot enough to burn your finger is probably a sign that the processor is damaged. However, on an off chance, power down, cool off and then power back up to see if it remains dead. However, before you do that, pull the daughter board out and power up the MS from whatever power supply you were using to check the voltages to the daughter card. No point in reheating the daughter card or toasting a second daughter card if the first is history.

Check the voltage on the 40 pin socket to make sure that you are getting 5 v on pins 1, 20 and 31. If you have the S12C to js9 jumper installed for an IAC you should get whatever your power supply voltage is on pin 16. Check to make sure that your ground continuity is good on pins 2, 19 and 32. With the daughter card out, you could also check for voltages to ground on all the other pins. Depending on whether the MS is on the bench or in the car and what is hooked up to it, you will see some voltages on the analog inputs. None of those voltage should be above 5 volts.

If your voltmeter can be set up to measure DC current, with it set to measure current connect the meter into the DC power supply to the Megasquirt. With my MS sitting on the bench and connected up to a lab power supply running at 12 volts supplied through a JimStim, the not spectacularly accurate meter on the power supply says that the MS is drawing a little over 0.1 amp. If you're MS is drawing much more current than that, you likely have a problem.

You also stated that you replaced the MS voltage regulator on the MS so that you could power up some extra 5 volt circuits. I don't know what you mean by extra circuits; but, that seems like an extraordinarily bad idea if you are using the 5 volt supply to power up things outside of the MS box. If you have other things that require 5 volts, fabricate a separate 5 volt supply for that stuff.

By extra circuits I mean a extra map sensor and eventually sequential injector drivers

[142 guy]

By extra circuits I mean a extra map sensor and eventually sequential injector drivers

Extra injector drivers do not materially increase the load on the voltage regulator. If your second map sensor is an MPX 4250, their supply current is typically 0.007 amps so I don't think that is going to be an issue either.