calderp

I have a 1986 Toyota SR5 that's experienced a serious power lost. The misfire happens through acceleration and causes the engine to shake. It started kind of all of a sudden.
I brought it to a mechanic who replaced all the fuel injectors (two of which had burnt) and concluded that the misfire was a fuel deficiency caused by one of the fuel injectors not getting a signal to boost fuel flow.
We tried replacing the fuel injector resistor (28515-26030) but it didnt solve the problem. Somewhere between the ECU and the fuel injector there's a problem.
Anybody know whats up? or Where we should look next?
I'm having a hard time finding anyone with a diagnostics computer that will give me a reading a truck this old.

AKHeathen

i learned from someone on here today that where the wires to the injectors are pressed together ?(basically spliced) it can fail, and apparently common. check that, then do an spark plug wire disconnect on all cylinders and check if it changes the idle. if it doesn't it is a dead cylinder.

Terrys87

iTrader: (1)

Here is a thread that will probably cover what you are dealing with>>> https://www.yotatech.com/forums/f116...firing-290833/

Also the 85-87 injector connectors are a poor design. Call Alex at http://connectorsfast.com/web/ for new connectors, he will know what you need.

wallytoo

iTrader: (1)

check your fuel pump output. the pumps fail regularly. i just replaced mine for the 4th time in less than 3 years. fwiw, i purchased supposed "original" equipment denso pumps 3 times, and a "high quality" aftermarket pump once. one pump lasted 22K miles, one only lasted 3.5K miles, the other one is in between. the 4th i won't know until it fails, which i expect within one year. at least toyota expected this - they provide a nice access panel under the rear passenger seat to get at it in the 4runner. replaced the 3rd with the 4th yesterday, with a full tank of fuel. that was the one that only went 3.5K. oh, what a feeling. toyota.

scope103

Did the mechanic say which one? If not, that's what a "noid" light is for. Commercial noid lights cost what they do because they have fancy universal connectors, but if you're willing to back-probe the injector connector a plain-ole 12v (integrated resistor) LED from Radio Shack will work (remember that these are polarized).

Once you figure out which one is not getting power, as AKHeathen suggests you should just carefully untape the harness until you find the break (incidentally, all four injectors are wired together, and fire at the same time).

Don't worry about that. There is no signal in the data stream which will identify the mis-firing cylinder. ("Modern" vehicles use a multi-tooth "gear" to determine the crank position and speed exactly; if the crank doesn't speed up when the plug fires, it concludes the cylinder mis-fired.) You can use the spark-wire-disconnect method (AKHeathen) to identify the suspect cylinder, but since lots of things can cause a dead cylinder I'd follow it with a noid light before I started harness surgery.

wyoming9

iTrader: (1)

Wallytoo

you must have really bad luck

Been driving Toyota`s for the last 15 years never had a fuel pump fail.

Went through 5 ECM units I guess that evens out

RJR

One of my "back burner to-be-done" projects is to connect my Hantek USB scope to a couple of the distributor signals, collect some data in the laptop, and see if it's possible to duplicate this function. Might be handy sometime. Don't hold your breath, though.

wallytoo

iTrader: (1)

i'm sure.

notably, there were two pumps in the tank of my '95 legacy (to straddle the rear driveshaft), and both still work @ 298,770.

AKHeathen

are you referring to misfire cylinder identification? to do this you need at least one crank angle identifier. in distributored engines like mitsu (mopar) it's done with a beam inturupt plate, but similar to pulse rings and there is always one extended open or close section to identify "start" and pulse counts to the immediate speed change. i suppose you could use a double inductor (timing light clip) with one on #1 and the other on coil with tach input, then program a adruino module to sense the speed change with pulse timing and a selection of cylinder numbers with a switch cancelling ic counter pairs. would be a handy tool for older engines.

scope103

AKHeather - You may be over-thinking it. I think RJR just wants to do all his timing off the NE signal (an inductor clip on one cylinder would help identify "which" cylinder is the one in question, but wouldn't add anything more). The problem is that you get NE only 3 times per revolution (2 times in a 22re) as contrasted with 24 or so with a crank position sensor. It may be possible with some clever analysis of the timing.

But like all good "back-burner" projects WHO KNOWS what you'll find!

RJR

Yeah, scope, that's what I was thinking of. According to the docs I have, on the 3vze there are two signals from the distributor. Ne appears to output 6 cycles per crank revolution, and G1(or G2) once per crank revolution. I figure if I just digitize those signals at a 1Mhz rate or so (1 degree = 83uS at 2000rpm) and grab a couple of engine cycles, then the information should be there in the data record. Whether one can detect a misfire with only 6 cycles per revolution is a question, but with enough averaging perhaps so.

The Hantek USB scope is a lousy scope, but as a simple digitizer it's almost perfect for this application. Adequate sample rate and resolution, and plenty of memory depth.

Like I say, though, don't hold your breath.

scope103

I was thinking about this on my bicycle ride today. The left cam sprocket has the "flat" side pointing forward. You could easily paint black and white segments on that rim, then aim an LED/phototransistor (expertly mounted in the timing cover) at the rim. You should be able to get a pretty good signal with 24 (15°) (or more) transitions per cam revolution (or roughly 4 transitions per plug firing). You could easily include a "missing tooth" described by AKHeathen to zero everything. The problem I see is that the rubber timing belt might have enough compliance to damp out the "push" you're looking for as a signal of the cylinder firing.

There's really only one way to find out; so I AM holding my breath.

wallytoo

iTrader: (1)

replaced the fuel pump again, because fuel pressure was down to 22psi, and would't go higher. got tired of pricey pumps failing, so tried a cheap one.

worked for about 15 seconds before it locked up. good thing i bought 2. put the 2nd one in. truck has been running for 10 minutes. pressure holding @ 30, and jumps to 40 when i pull the vacuum line from the FPR. we'll see how long this one lasts.

6 pumps in less than 50k miles.

wally

wyoming9

iTrader: (1)

Your not in the habit of trying to run the truck only with the fuel level below 1/4 tank all the time??

Very seldom if ever to my tanks get below 1/2 tank.

Odin

That's a good point. I've seen many who suggest that pumps can fail early because they are not submerged in gas which helps to cool them. Others suggest that they may be partially ingesting air at that point too, unless you're on flat ground that is. Neither are good things.

wallytoo

iTrader: (1)

right. a brand new pump failing in less than 15 seconds has anything to do with fuel level.

Terrys87

iTrader: (1)

That is just a bad pump out of the box. I go into panic mode if my tank level gets less then 1/8th but generally fill up at a 1/4 tank. Check your Return Line. I see several of those lines clogged and feel it just makes the pump work that much harder when could lead to early failure.

wallytoo

iTrader: (1)

both the liquid return line and vapor return are clear w/no obstuctions. the FPR works correctly, the pulse damper works too.