pulled one fuel rail , ms unplugged from the harness
turned on fuel pump , and power to injectors no leaks at all


plug ms back into harness power it up fuel everywere with it powered on , power the ms off fuel stops spraying




and holy crap do those injectors spray a ton a fuel when wide open
\

 

perfect thank you



ms supports low z injectors no need for resitors except under some circumstances


btw my ms was originally an ms2 3.0 , ive downgraded it to an ms1 v3.0 though I do still have my ms2 cpu I can reinstall

 

what is getting me is soon as I kill power to the ms the injectors stop spraying , iirc if u blow the driver they will spray all the time

 

If the MOSFETS are good, then that would be true. That is, the MOSFET without a gate voltage would turn on the injector with the MS unpowered, but..........

It sounds like the MOSFETS are bad, so the source (injector side) is shorted to the drain (power side).

If you set the MS to hi-z settings then that could have done it. At low BPW, a low-z sees the pretty much only the 1.6ms "peak" time same as a high-z. If you ran it and the BPW was higher than that it could have burned out the drivers because it wasn't doing the current limit "hold" mode.

What do you have lying around for junk electronics? Old TVs have a fair amount of MOSFETs in them.

 

reinstall the ms2 cpu, and reflash with the latest firmware.

It's probably pooched and commanding it open.


-- Joe

 

It does this by software doing a PWM to the injector during the "hold" time. The resistors would have saved the FETs if you accidently set the MS to high-z with low-z installed.

With MS unpowered, put an ohm meter on the drain and source pins. That is the ground pin to the driver and the injector pin. This will tell if the FET is shorted / fried.

EDIT: Do the ohm test with the MS unplugged from the harness.

 

Lol alot of suggestions in aftermarket efi for megasquirt but i see more trouble shooting threads on that system than any other

 

It's because it supports just about any ignition type on the planet, and has a NUMBER of options that must be configured properly. It's complex, and very feature packed.

But it's not idiot proof. When someone asked me what ECM I recommend, I hardly recommend MS because most people won't figure it out and then bitch that "It sucks".

-- Joe

 

Looking at the schematic.........

If he had the .050 ohm sense resistors in there, then they should have turned off the MOSFETs via the 3904 BJT. It shows a peak of 14 amps (.7v/.050 = 14 amps).

Normally, peak and hold drivers put out about 2 amps peak and .5 amp hold per injector. Each driver handles up to 4 injectors. Not sure why they are driving them so hard.

Suppose you did by accident tune it to hi-z, then it would have had a current of 14v/(2.5/4) = 22amps. Max current for the MOSFET is 30amps. RDSon is .050 ohms so not much power dissipated. Seems kind of hard to destroy them unless you do something wicked to them.

 

i just got home i couldnt do anythign with the crappy internet i have down there , brought my ms with me along with a pigtail and 9v battery so i can do the diagniostics on it

this is a unit i bought from a member here used , and it wasnt put together well and was missing some key compnents , i still think it doesnt have the flyback circut installed or maybe thats what i did install a while back

either way i just need to go threw it step by step and see whats wrong

 

Ohm out the injector pin to the ground pin first at the connector with it unpowered. Should be "open" connection if MOSFET is good.

Doing PWM (low-z) without flyback ckt can/will fry it.

 

let me pull up the db 39 pinout and ill do that right now

 

ground pin to bank 2 injector pins shows resitance with the multimetter set to x100 scale it just maxes the meter

same as if i touch both multimeter leads together

ground pin to bank 1 shows around 2,500 ohms

this is with the ms unpowered

 

Sounds like they aren't shorted to me. When a MOSFET blows it is usually a dead-nuts short reading 0 ohms.

Wonder if it is a setting like anesthes thought. My next test I would power up and look at pin 1 on Q1 and Q5 and measure voltage. Should be around 0 volts to turn the MOSFET off.

You sure you had power (and ground) to the MS when it was in the car?

 

yes leds were on in the car , pin 1 is far left on q1/5 correct ill check that right now let me wire up this 9 volt battery

 

Yes, looking at the part so you can read the writing on it, pin 1 is on the left.

After you measure the voltage, measure the voltage on the db39 injector pins to ground.

 

need a few mins my 9v battery is dead , going to wire it up to one of my server psus i use to run my bitcoin miners as they all put out 12v

 

0v on pin one of q1 or q5

q5 apears to be melted

checking u4 now

 

0v is mosfet off. Measure voltage at DB37 injector pin to ground.

sucks cause I have a pile of MOSFETs here. You have any junk TVs around?

 

u4 has 5 volts on pin 2 and 4

measureing coltage from inj pin to ground now

 

0v inj pin to ground

no junk tvs laying around

 

U4 is good, doing what is supposed to do, invert signal.

 

edit nm i thought u asked what does y4 do

so what ya thinking the mosfets bad?

 

can you measure voltage from Q1 and Q5 pins 2 to 3 (middle and right pins)?
put one prob on pin 2 and other probe on pin 3.

 

will do right now

 

0v between pins 2-3 on Q1 and Q5

 

dang, forgot you don't have a load/injector simulator on the inj pin.

Only way to to check Q1/Q5 powered up is have a load on the inj pin to power.

So far sounds like U4 is working and Q1/Q5 is shorted...........but ohm test said no.

 

You have any resistors you could connect between power and inj db37 pin for Q5 ckt (pin34/35)?
Or any injectors hanging around?

 

no nothing at home , i can get my hands on another ms unit but not till friday night , when were supposed to be loading the car on the trailer lol

im going to finish the test on u4 , and also check the resitors on that circut

 

going to run this test , worst comes to worse ill unsolder q1/5 and plug it back into the car and see if the injectors still hang open

First, check pins 2 and 4 on U4. If they're not getting 5 volts in the off state, turn the power off and ohm out R14 and R17, and make sure you have continuity to the CPU.

U4 is an inverting driver. If you don't see its output on pin 7 go high when pin 2 is grounded, or pin 5 go high when pin 4 is grounded, you've got a bad U4. Same if it's stuck high when it gets 5 volts on the input. Note that if you have a 9 volt battery, the points I've marked at "12V+" are going to be lower, more like 8 volts. You'll usually see somewhere around 12 volts on a power supply. If you're seeing more like 10 volts, don't panic; it's still working OK. Time to move on.

Check the input signal on pin 1 of Q1 and Q5. It will be a little lower than the voltage at U4, but not by much. If the voltage has disappeared, power it off and check the ohms on R15 or R20 as appropriate. Either R15 or R20 has failed, or the transistor in the current limiting circuit (Q14 or Q15) has shorted out and needs to be replaced.

If the signal is making it to Q1 or Q5, but the injector driver isn't working, first check for continuity from pin 2 of the FET to the DB37. If that checks out, and the injector driver is failing to turn on, replace Q1 (for injector 1) or Q5 (for injector 2). If the injector driver is stuck on, it could be same transistor, but it could also be the flyback circuit. To find out which one it is, desolder D5 (for injector 1) or D7 (for injector 2). If the output is still stuck on, it's the transistor. If the output goes off, it's the flyback circuit. The most likely culprit is Q3 for injector 1, or Q11 for injector 2.


for input signal on pin 1 what should i be looking for 5v pin 1 to ground?

 

wait i already measured that and showed 0 volts on pin 1
with no voltage on pin 1 leads me to belive i need to check out what i quated above

 

You are going to find u4 is good.
My best guess is Q1/Q5 is shorted even though on-state ohm test said no.

I really want to have you put a jumper wire from one of the 10k resistors sourced from vcc, but it is too difficult to explain like this. And risky to break something if done wrong.

If you come across a resistor, do the test I bet Q1/Q5 is all done.

 

You want to see 0v on pin1 of Q1/Q5. Which you saw so U4 is good.

It is either Q1/Q3 or TIP42...........but I thought you didn't have TIP42s installed.

 

r15 showing about 230 ohms if i have this meter set right

i modded this ms once and i think i added q9

or i added Q9/12

taking a picture of the board now

 

here u can see were it looks liek q5 got hot and melted

 

and here u can see all the stuff not installed

 

re ohmed ground pin to injector output pins and have restance on both , that means they are shorted correct

 

Yes, Q5 looks ugly.
I would expect a higher resistance on the Q1/Q5 inj pins to gnd.
Although, it could be TIP42 causing the low resistance. Q3/Q11 could be blown.
Only way to tell is to snip one pin on the package and measure, then if it is good put a small jumper wire.
When mosfets look like Q5 they are usually all done. Is that the ckt that measured 2500 ohms?

That is, cut only one pin above solder joint and then resolder if part is good. That why you don't risk wrecking board removing entire part.

 

not sure which pin to snip or desolder
going to try this ^^^^^^


right now im trying to figure out what i will need to change to run my friends ms2 v3.0 in the car for the weekend , im pretty sure my wiring stays the same but i need to see if he has the right jumpers inplace for ignition though im sure he does he built his for gm hei

edit yes im pretty sure Q5 was the one that measured 2500 ohms

 

Yes, you can do the diode de-solder, but you don't have the resistor/load I was talking about to see if Q1/Q5 is on once you go to test.

 

Yes, you can do the diode de-solder, but you don't have the resistor/load I was talking about to see if Q1/Q5 is on once you go to test.

By cutting Q5 pin 3 you can ohm to pin 2 to see if Q5 is shorted. This is with board not powered.

The diode method is valid if you have the MS with injectors installed or a load resistor.

 

If you have the time, doing the de-solder of diode and testing in the car is easiest.

Gotta give you credit, stacking hay in the morning, got the smarts to make your own turbo setup, and the smarts to poke down into an MS.

 

im going to desolder the diodes and wait till i get to the shop tonight or in the morning to see if the injectors still hang open

im about 5 mins away from buying an ems pro infact i already have one in my shopping cart and prolly would have bought it already if i could get it in my hands before friday

ill renmove the diodes and see what happens , as for now ill just plan on installing my buddies ms into the car for the weekend

 

Don't do it, you can figure this out. It is Q1/Q5 and maybe Q3/Q11 if MS was blinking LEDs properly (it didn't forget it is an MS controller).

Probably $20 worth of parts to just replace all at Mouser.com

 

lol im not but the thought has crossed my mind , we got thunderstorms atm so ill do the diode test in the morning and see whats up

 

That was a fun read.

In all seriousness, once you determine if you have a hardware problem or not, please re-install the MS2 cpu and flash it with the latest 'extra' firmware.


-- Joe

 

why did you take out the ms2 cpu? i ran mine controlling fuel only, but i left the unit complete....

also never had any issues with my injectors holding open due to the ms2's "flawed design"

 

its a hardware issue i already knwo that since the injectors do not stay grounded when i unplug or power off the ecu

and this reply is for both of u

i am 100x more familiar with the ms1 code then i am the ms2 code , when i first got this unit years ago i was having some issues which looked like a cpu issue , since i know the ms1 extra code better and the cpu is alot cheaper i bought an ms1 cpu to replace the suspect ms2 processor.

after replacing the the processor the issue was still there and later sorted out , i never reverted back to the ms2 cpu

since im going to more then likley run my buddies ms2 unit in the car , when i repair the injector driver on mine i am going to reinstall the ms2 cpu

i have done numerous ms1/extra v2.2 and v3.0 installs , this car was the first time using ms2 in the beginning

 

ive got the diodes cut on one leg so i can test , just waiting for the rain to stop as i dont feel like getting soaked riding the 4 wheeler over

besides that issue im going to take a look and make a list of everything else i need to finish the car before friday night if its just to much im prolly going to not goto the race

the reason being is i will have to tune the car on the track on saturday at the expense of around 400$ for making the trip and if i only get 1-2 runs ill be nowere near having the car tuned enough to make a good pass to keep the internet ppl happy sorry not worth it to me those ppl can hold off another 30 days till the next event when the car is in a better state if thats what i decide today

again it all depends on how much work i actually have left as i really do wanna go but im not risking hurting the car or beating myself up over the next 3-4 days to make it happen

 

Diodes d5/7 removed and Injectors still stuck open so the.transistors r fried if I am reading the trouble shooting guide correctly