Peak and hold injector driver board

A forum for discussing the MegaSquirt related (but non-B&G) board development, assembly, installation, and testing.

Moderators: jsmcortina, muythaibxr

Post Reply
racingmini_mtl
Super MS/Extra'er
Posts: 9127
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: Peak and hold injector driver board

Post by racingmini_mtl »

The only thing I can think of are that either Q2 is toast and always conducting between pins 2 and 3 or something is grounding pin 2. The only things connected to pin 2 are the injector output and D2 (left side).

I think you said you've replaced Q2 before but if you have some extra, you could try to replace it again. Then inspect everything closely around pins 2 and 3 of Q2 on both sides of the board (when Q2 is not installed yet) and have a close look also at D2. I must say this is rather baffling. If you want, I could send you another board.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
linuxman51
Experienced MS/Extra'er
Posts: 213
Joined: Sun May 02, 2004 2:55 pm
Location: Montgomery, AL
Contact:

Re: Peak and hold injector driver board

Post by linuxman51 »

any potential problem with using a tip127fp? I'm going to depopulate #1, i suspect the board won't survive this time but we'll see. What things should I check for once its "blanked" (aside from obvious/visible shorts). Its almost like the circuit works in "reverse", as soon as power is applied to the output it triggers, until then i read 2.1M ohm reistance between the inj1 wire and ground (and I'm assuming thats megaohm as well)
12 years of ms-ing things up..
607whp volvo 740 (ms3pro)
1994 volvo 940 turbo L33+7875+ms3x
racingmini_mtl
Super MS/Extra'er
Posts: 9127
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: Peak and hold injector driver board

Post by racingmini_mtl »

Don't use the TIP127: this is a PNP transistor and you need a NPN.

If the driver is completely depopulated, you should take the schematic and check that you have continuity between the correct pads and not between the pads you shouldn't. Concentrate on the pads connected to Q2.

By the way, my offer for another board still stands and would be free of charge.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
linuxman51
Experienced MS/Extra'er
Posts: 213
Joined: Sun May 02, 2004 2:55 pm
Location: Montgomery, AL
Contact:

Re: Peak and hold injector driver board

Post by linuxman51 »

Ok you haven't updated the schematic to reflect the separate ground for pin 6 (took me a minute to remember that).

nothing appeared to be shorted (at least not via the continuity finder on my fluke). I re-attached everything, however I isolated the circuits for pins 2 and 3 on the fet from as much of the board as possible, and now I have simply achieved a non functioning circuit. I guess thats better than before. I don't know if I'm really in the mood to wait another three weeks before finishing my car, I'll have to decide if this is really the route I want to go.
12 years of ms-ing things up..
607whp volvo 740 (ms3pro)
1994 volvo 940 turbo L33+7875+ms3x
racingmini_mtl
Super MS/Extra'er
Posts: 9127
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: Peak and hold injector driver board

Post by racingmini_mtl »

You're right about the schematic. Here is the correct one if you need it: http://jbperf.com/p&h_board/v1_1/p&h_bo ... _schem.png.

I wish I could help you more but it's quite difficult to diagnose those types of failures remotely.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
linuxman51
Experienced MS/Extra'er
Posts: 213
Joined: Sun May 02, 2004 2:55 pm
Location: Montgomery, AL
Contact:

Re: Peak and hold injector driver board

Post by linuxman51 »

racingmini_mtl wrote:You're right about the schematic. Here is the correct one if you need it: http://jbperf.com/p&h_board/v1_1/p&h_bo ... _schem.png.

I wish I could help you more but it's quite difficult to diagnose those types of failures remotely.

Jean
yeah don't I know it (I do computer work for a living). its doubly frustrating because everything seems right yet it completely doesnt work. I reckon I'll take you up on your offer, let me know what it'll take to get it here sooner than standard.
12 years of ms-ing things up..
607whp volvo 740 (ms3pro)
1994 volvo 940 turbo L33+7875+ms3x
racingmini_mtl
Super MS/Extra'er
Posts: 9127
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: Peak and hold injector driver board

Post by racingmini_mtl »

PM sent.
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
racingmini_mtl
Super MS/Extra'er
Posts: 9127
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: Peak and hold injector driver board

Post by racingmini_mtl »

Complete kits for the V1.1 board are now available. Look here for details.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
linuxman51
Experienced MS/Extra'er
Posts: 213
Joined: Sun May 02, 2004 2:55 pm
Location: Montgomery, AL
Contact:

Re: Peak and hold injector driver board

Post by linuxman51 »

got the board today, just finished putting it together and hooking it up, it worked like a champ the first time out. Guess the last was just a defective pcb or some other component on that driver circuit. Thanks again Jean
12 years of ms-ing things up..
607whp volvo 740 (ms3pro)
1994 volvo 940 turbo L33+7875+ms3x
racingmini_mtl
Super MS/Extra'er
Posts: 9127
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: Peak and hold injector driver board

Post by racingmini_mtl »

Glad to hear it's working fine!
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
linuxman51
Experienced MS/Extra'er
Posts: 213
Joined: Sun May 02, 2004 2:55 pm
Location: Montgomery, AL
Contact:

Re: Peak and hold injector driver board

Post by linuxman51 »

:yeah!: car cranked up on the first try. now I've gotta sort out a bizzare tach signal issue (roving between 0 and 1200 at idle of about 1000 before settling down, seems to be related to the tps). This is awesome. Thanks jean!
12 years of ms-ing things up..
607whp volvo 740 (ms3pro)
1994 volvo 940 turbo L33+7875+ms3x
linuxman51
Experienced MS/Extra'er
Posts: 213
Joined: Sun May 02, 2004 2:55 pm
Location: Montgomery, AL
Contact:

Re: Peak and hold injector driver board

Post by linuxman51 »

:RTFM:
Car ran friday, won't run now, board tested fine on the bench after the fact, I just have a quick technical question:
what would cause the injectors to not fire with ms1/hi res, but work fine with ms2-e? This is on the stim at a low rpm value. ms2 chip on the board, everything works like a champ. Ms1 chip on the board everything but the injectors works like a champ (this was a cause for great frustration while the whole deal was hooked up to the car, as I have my latest tuned map for ms1/hi res and the dang thing wouldn't crank). the only electrical thing to change since friday was a charged battery and the alternator exciter wire (oh and the power steering/waterpump belt but I doubt this has had an impact on anything). I'll be making sure the grounds for the board didn't come loose this evening, it'd be nice if thats all that happened :oops:
12 years of ms-ing things up..
607whp volvo 740 (ms3pro)
1994 volvo 940 turbo L33+7875+ms3x
racingmini_mtl
Super MS/Extra'er
Posts: 9127
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: Peak and hold injector driver board

Post by racingmini_mtl »

There is no reason why it would work with MS2 and not MS1. There is no direct connection between the CPU and the p&h board.

You should have a look at the signal coming out of the CPU with either an oscilloscope if you have one or a LED tester and then the signal going to the p&h board. Also, make sure the TPS pot on the stim is not too high so that you're in flood clear.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
linuxman51
Experienced MS/Extra'er
Posts: 213
Joined: Sun May 02, 2004 2:55 pm
Location: Montgomery, AL
Contact:

Re: Peak and hold injector driver board

Post by linuxman51 »

racingmini_mtl wrote:There is no reason why it would work with MS2 and not MS1. There is no direct connection between the CPU and the p&h board.

You should have a look at the signal coming out of the CPU with either an oscilloscope if you have one or a LED tester and then the signal going to the p&h board. Also, make sure the TPS pot on the stim is not too high so that you're in flood clear.

Jean
oh man if I did that I don't think I can come back and post again :lol: I'll check it at lunch and make sure, the rpm for the ms1 chip was fairly low, that could be it. I noticed while I was testing it with the ms2 chip, until I hooked megatune up and realized the tach was pegged at something like 26,000 rpms only one injector bank worked, somewhat sporadically.
I didn't think there was a fundamental difference between the two chips as far as basic injector triggering was concerned, but I wanted to double check lest I spend days chasing something that is known to not work 8)
12 years of ms-ing things up..
607whp volvo 740 (ms3pro)
1994 volvo 940 turbo L33+7875+ms3x
linuxman51
Experienced MS/Extra'er
Posts: 213
Joined: Sun May 02, 2004 2:55 pm
Location: Montgomery, AL
Contact:

Re: Peak and hold injector driver board

Post by linuxman51 »

:RTFM:
car doens't crank without outside interference. Here's the skinny after purchasing a set of noid lights (injector circuit test lights), I get zero pulsewidth while cranking. none. nada, the null set. As soon as the car hits off everything starts working as per normal (a helping of ether in the intake will get the car cranked up without issue), car idles fine.

So... what might be off in the config thats causing a no blink at the plug while MT is reading cranking pw's of 4.5? I get the same consistent 230-250 rpms whilst cranking. Code is HR10d with a known good map (i.e. car never failed to crank and run with my old v2.2 setup).


*edit*:
With the cranking rpm set to 100, the car will start up every single time. set the cranking rpm higher than 200 and it will flat out not crank up. It seems like the pulse while cranking with the cranking rpm set higher (I've previously had it set around 400, which has seemed fine :shrug:) is very weak to non existent?
12 years of ms-ing things up..
607whp volvo 740 (ms3pro)
1994 volvo 940 turbo L33+7875+ms3x
linuxman51
Experienced MS/Extra'er
Posts: 213
Joined: Sun May 02, 2004 2:55 pm
Location: Montgomery, AL
Contact:

Re: Peak and hold injector driver board

Post by linuxman51 »

Ok I had to work late last night, played with the car briefly this morning (i'm quite sure my neighbors love me now, there's no downpipe on the turbo at the moment and you can hear the exhaust valves closing) and I've drilled it down to this:

The car flat out will not run if the injectors are set up to fire simultaneously. I can crank it up with alternating and cranking set to 100, rev it up to 3k, make the change in megatune, and it instantly dies. So this would be why it doesn't start with the correct cranking rpms.

So I ask you, what would cause this? I'm out of ideas. I've changed the ground location for the daughter card, hooked the 12v supply up to a 30amp circuit, the ms has its own dedicated 20amp circuit, and I'm getting absolutely no noise in my datalogs while cranking or idling, in fact its probably smoother than its ever been (guess most of the new harness is good :yeah!: ).

Its not a big deal i guess, as I can get the car to start, i've always run it in alternating mode, and its my race toy so hard starts aren't a problem as long as it lays down the fat pipe on the track, but I don't want to run into a problem where it gets higher up in the rpms and bails out due to the closeness of injector events.
12 years of ms-ing things up..
607whp volvo 740 (ms3pro)
1994 volvo 940 turbo L33+7875+ms3x
linuxman51
Experienced MS/Extra'er
Posts: 213
Joined: Sun May 02, 2004 2:55 pm
Location: Montgomery, AL
Contact:

Re: Peak and hold injector driver board

Post by linuxman51 »

:yeah!:
Jean walked me through a couple of tests & we ultimately determined that somewhere along the way the IDXI on the v3 board bailed out (intermittently) I replaced this and everything works as it should.
12 years of ms-ing things up..
607whp volvo 740 (ms3pro)
1994 volvo 940 turbo L33+7875+ms3x
deluded
MS/Extra Newbie
Posts: 5
Joined: Tue Jul 08, 2008 7:47 pm

Re: Peak and hold injector driver board

Post by deluded »

after spending untold hours reading the data sheet for the lm1949 and studying the circuit + captured real world wave forms i am confused/concerned about the behavior of the wave form under two circumstances.
1) when battery voltage is less than 4 X injector ohms
2) when injector ohms >3.38 (assuming best average battery voltage of 13.5 volts)

either situation seems to make it impossible to reach the peak trigger current and the circuit relies on the timeout to drop back to the hold current. specifically, the very popular bosch 160# injectors are 5 +/- .1 ohm which RC engineering rates for 2.5 ohm peak and 1 amp hold current (2.5:1 ratio rather than the internally set 4:1 ratio that the lm1949 uses). i directly measured the 5 ohm value from new sets of both the blue tops and white/grey tops; i realize many web sites list a little lower theoretical value. using the above battery voltage the best case injector current should be 2.7 amps which could never reach the 4 amp peak trigger. i confirmed this with a scope in a running car. i'm waiting for a .2 ohm sensing resistor to test and see if these injectors will work with a 2 amp/.5 amp peak/hold though i doubt that .5 amp will be sufficient to hold the injector open.

i have prototyped a daughter board for an oem ecm using suggestions on this thread including the NID9N05CL-D mosfet and the shortened timeout (works great but for the above issue). i also utilized the 4 lead 13fr100 discovered after researching the BOM sensing resistor alternative. this forum has provided invaluable information that i appreciate greatly. thank you for considering this question.
racingmini_mtl
Super MS/Extra'er
Posts: 9127
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: Peak and hold injector driver board

Post by racingmini_mtl »

You're correct that the LM1949 is designed for injectors needing a 4:1 peak/hold ratio. And the BOM for my board has been designed for injector needing a 4A/1A peak/hold current. If you need a different ratio, you'll have to use something other than the LM1949 or have a 4:1 ratio and tweak the parts to have a consistent behaviour.

As you say, when the battery goes to less than 4 times the injector impedance, the circuit relies on the timer. However, even though the current never reaches 4A the injector should still open after some time unless battery voltage is very low or the injector impedance is too high. Low impedance injectors requiring 4A/1A are well under 3 Ohm impedance so your two concerns would not be an issue. And the timer component values can be changed to accommodate some borderline cases.

In your case, you will need to see if changing the sense resistor as you're planning will give satisfactory results. Using a 0.2 resistor might be a bit high because you'll never reach your desired peak nor hold value. You could try using a 0.15 resistor which would give 2.67A/0.67A peak/hold values or a 0.13 resistor which would give 3.08A/0.77A peak/hold values. I'm not sure that the 0.13 Ohm resistor I found in my quick search would be correct as a current sense resistor but the 0.15 is (13fr150).

By the way, there were low impedance injectors which required 2A/0.5A peak/hold and therefore would use the 0.2 Ohm resistor with the LM1949. So it is possible that your injectors will be fine with this.

So you are using a mosfet instead of the Darlington? Do you use an external clamp diode or do you use the internal one? And since these are surface mount, how do you deal with the heat? I'd be interested to see more so if you could either PM me or email me at boards at jbperf.com.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
RicerX1
MS/Extra Newbie
Posts: 2
Joined: Fri Aug 08, 2008 12:59 am

Re: Peak and hold injector driver board

Post by RicerX1 »

deluded - Any luck with changing the sense resistors for the bosch 160# injectors?
I'm about to do a similar setup.
Post Reply