I'm having an issue with my speedometer reading while my car is idling. it reads much like the tach and settles down as RPM settles down. I found this thread regarding the issue but there is no solution in the end other than the suggestion to check grounds, which I have done.

A little background on my build is as follows. New engine with an adaptronic plug and play running single turbo, FFE triggger w/ stock sensor, and direct fire. I mapped and built my own engine harness for the car to fit my needs. This harness follows the pinout diagrams I have attached. I understand that this issue is most likely grounding and/or noise/ feedback but I'm at a loss as to how to solve this because my harness is wired as if it were OEM with the exception of the coil trigger wires running through the engine harness to the AC pump area. Grounds have also been checked more than once, with extra straps added.

I have been in contact with Ihor at IRP about the issue and he agreed that it seems like a feedback issue. He also said that he sees this with Rywire harnesses which leads me to believe the OEM harness has some extra shielding in it somewhere.

If anyone has any suggestions and/or a solutions it would be much appreciated.

Thanks,
Ryan

 

I have seen it with at least 1 Rywire harness but others had no issues, so I don't want to say they are all bad. I have also seen it on stock harnesses. Memory is a bit fuzzy as this was years ago but I remember at least one someone attempted some harness repairs and didn't shield the CAS wires. I also just remembered I had a similar issue with a car that had a poorly installed stereo. Not only did the speedometer move with rpm and the car stationary, but the odometer also counted miles. On the same car you could hear the feedback through the stereo even with the stereo turned off.

 

Ryan, thanks again for sharing your diagrams and charts. Saving me a bunch of time figuring out my harness.

When you built your new harness, did you use shielded wire for anything? On the oem harness, the knock sensor, Crank sensors, coil signals, and O2 sensors all use shielded wires. Of course your problem wouldn't be caused by the O2 or knock sensor wiring. But, the speedometer sensor wiring (unshielded) runs with the engine harness through the firewall. Could possibly be picking up interference from either the crank sensor or the coil signal wiring. I'm not sure which is more likely to cause the interference, but my bet is on the CAS. It makes sense because the speedo moves up with rpm.

I think the easiest way to check if this is the problem, would be to run a pair of wires from the ecu area down to the trans and the speedo sensor, bypassing the harness and see if the speedo works normally. Sucks that you would have to cut wires to do this.

 

I forgot to actually post the link to the other thread on the subject. Here it is https://www.rx7club.com/3rd-generati...eutral-908789/.

So the coil trigger wires are shielded in the stock system? I could've easily missed this because they run through the body harness with a Stock coil setup and I didn't think to check if they were shielded wires. That sounds like it may be my issue right there.

My CAS and Knock wires are shielded and the shielding around the wires is grounded exactly as an OEM engine harness. O2 sensor is an innovate MTX-L and runs separately from the engine harness. I can patch together a harness for my coils like the SBG kit that pulls from the stock igniter plug. If that solves the issue I'll pull the harness over the winter and shield my coil triggers.

 

Whoops, I went to grab a screen shot and it seems I made a mistake.





The SIGNAL wires from ecu to igniter are not shielded, only the wires from igniter to the coils themselves. Sorry for the confusion.

 

Damn, Thought that was going to be the ticket. Still probably not a bad idea to shield my trigger wires. Let's hope somebody chimes in with a solution.

Thanks,
Ryan

 

If it is in fact interference causing it, then it would still kind of point to this wiring because it is rpm changing.

I read the thread you linked to. I am guessing that you've checked for the engine harness ground that they are referring to as a ground for the cluster?

 

I have checked this ground and it is solid. It is the ground that mounts on the rear stud of the UIM. I am going to check my stock harness one more time to confirm that the three ground pins on the ECU, 4A, 4B, & 4C, are interconnected in the harness. I currently have wires from these pins running separately in the harness to three ground locations, front cover, front housing, and UIM. They are then linked together and the UIM ground has a strap connecting it to the body.

I had this same issue pop up on an old car that I converted a stock harness to a single turbo configuration. The issue solved itself in a matter of days though so I can't use that to help me here.

 

I Checked grounds and they are good although I may add a few more straps from engine to chassis. 4A, B, & C are connected in the OEM harness and in mine. I also tested the resistance of my speed sensor, this came in at 241.3 ohms which is in spec.

I haven't driven the car yet other than from one garage to the other(about 20 yards) and I wasn't looking at the speedo as I did so, stupid me. I know that many of the people that have this problem have it go away when they put the car in gear or start to engage the clutch which is baffling.

My next check is going to be testing the wires that run from the speed sensor to white connector for voltage or noise with the car on and unplugged from the speed sensor. There is also the possibility that I crossed a wire with one of the other sensor on the trans but the tachometer type function wouldn't be explained if that were the case.

Thanks,
Ryan

 

A little update on this but no solution.

I checked my wiring and found that I had the speed sensor connector pinned incorrectly. I fixed this and finished my intake piping then it was time to start the car. Issue was not solved, speed ran right up to 35mph. After warmup on this fresh build speed settled down to about 5mph. Just before I decided to shut it down I noticed that the speedometer would sit at 0 for 10-30 seconds and then go back to tachometer mode for a little. I would like to think this is a ground causing the issue but I'm still chasing ghosts.

 

I've been going through my wiring diagrams and the Factory diagrams and cant come up with any issues. In my research I stumbled across this post, https://www.rx7club.com/haltech-foru...orking-760375/, in the Haltech section which has relevance. It talks about grounding the black with blue wire going into the cluster to make their tach work and prevent the issue I am having. The problem with this fix is that it kills the fuel tank sender and coolant level buzzer which I don't have any desire to disconnect. I am convinced I am having a grounding issue somewhere with this car, causing the issue, but seeing as though I wired my harness exactly as stock while removing unnecessary wires I cant see an issue.

Here is a quote from the thread describing the issue I am having.

 

Did some more R&D this morning, I unplugged the speed sensor and hooked my DMM to the connector on the engine harness. With the key switched to ON I'm reading 1.7 volts, with the car idling I get 1.5 volts. Unfortunately I didn't get a look at the DMM when the Speedo decided to drop to 0. I then went to the harness connector that links to the dash and unplugged & probed it. Got the same 1.7-1.5 volts there. So if the voltage is not supposed to be on those lines my problem is occurring somewhere in the dash.

I assume that voltage is supposed to read 0 while not moving because the Speed sensor isn't generating. Can anyone confirm this?

 

When you are unplugging the connector at the trans, and at the connector in the dash, are you checking that voltage in the two wires that lead up to the cluster? The manual gives no check for that voltage because it is an internal circuit in the speedo CPU in the cluster.
From Service Highlights manual


The only checks they give are for the speedo sensor and the wiring leading from sensor up to the cluster. Basically to check that there are no breaks in the wiring, and that when you rotate the sensor gear by hand, you can see almost 5v generated. If the signal is making it up to the cluster, then by default the speedo is bad.
From Body Electrical Troubleshooting manual





Do you get the same speedo malfunction with the sensor unplugged at the trans ?

 

You are not going to like my next recommendation for troubleshooting steps....
I've been thinking your problem over as I work cause much of my work is mindless repetition.

 

If this involves cutting wires I won't be happy to do so. I do have a spare cluster from my old car but the speedo tended to do the same things after removing the twin turbo connectors from the engine harness. This whole thing may be a case of me not being as competent at wiring as I think I am.

With the speed sensor unplugged the tach functions normally, speedo sits at 0. I want to rule out the cluster as the heart of the issue because it worked without issue before the engine swap and the story is the same with the speed sensor. Both are original to the car and have 78k on them. I know things can just happen and break but For now I'm going to assume the speedo is good. I will be putting the car up on stands this afternoon and be checking the speed sensor as well as my wiring once again,

 

I don't think you need to cut or hack anything apart haha. But, complicated wiring problems can sometimes best be solved by using division. You want to "divide and conquer" so to speak. Isolating circuits and components until you back the problem into a corner.
I would say if your speedo sensor checks out good, which I think it will. Then, I would remove the cluster. Find the connector at the back of the cluster that has the two wires for the speedo sensor, and de-pin them from the connector, leaving them taped up. Hook up the cluster and check operation. If it's still there, then it's internal to the cluster. If it's gone, put the two wires back in the connector body.
Go to the connector down by the ecu which connects the engine harness to the dash harness. This should also have the same two wires for speedo sensor. De-pin them at this connector and check the operation. If problem is still there, you have some crossover or interference in the dash wiring somewhere. If problem is gone, then you have just isolated the problem to being within the engine wiring harness.
I still think it would be a good option to run 2 separate and temporary wires from the X-14 (?) connector down to the speedo sensor to remove the engine harness from the equation.

And yes it is a good idea to double check ALL the grounds you've routed throughout the engine harness and check the polarity of the wires for the speedo sensor although I don't really think it would make a difference being an AC generated signal but you never know with these cars.

 

I just went through the "divide and conquer" approach. Started at the speed sensor. That checked out so I moved on to the connector just past the firewall. I first checked for continuity in the engine harness and that was good. I then moved onto the dash harness side of the harness and got the same 1.5-1.7 volts as yesterday there. Next step would be to the connector on the cluster.

I got the cluster loose and just by moving it around I could get the voltage to fluctuate between .8 & 1.5 volts. Unplugging the connector that the speed sensor wires run through dropped the voltage to 0 as you would expect. I did not get a chance to de-pin the two wires and see if the voltage is coming from the cluster itself or picking it up from nearby pins. This will be done tomorrow. and I fear the results are that something went sideways in the cluster for no apparent reason. This cluster has never been removed, only pulled out a little to get the HVAC controls out.

Andre, Can you probe your engine harness speed sensor connector and let me know if any voltage shows in your circuit with the key switched "on"? I assume this will be 0 but need to knock that variable out of the equation.

In other news, I need a new gauge cluster shroud because mine exploded into 10 pieces and took a chunk out of my finger.

 

The plot thickens. Looking at the back of this cluster and gauge shroud it's clear that I;m not the first person to be in there. The heads of all the Philips screws that hold the ribbon board on all show signs that they've been removed and the little cover that protects the ribbon board to circuit board connection is missing. At this point I am willing to put the blame on the cluster but am hoping to track the issue down inside the cluster after some research.

 

I'm afraid to say that I'm also worried that the cluster has "gone bad" but maybe this will be added to the list of other known cluster problems due to OLD AGE SYNDROME

I'm really sorry to hear about your cluster hood mine also broke a couple bits last time it was out.
Wish I could help you sort your issue by comparing to my car but here is my situation:




Did you say you have a spare cluster to swap in for testing?
I have a speedo unit (bare) that you could borrow and install in your cluster. It's original to my car, has no ODO display. But everything else was working when it was removed.

 

I do have a spare cluster but if I remember correctly, something was goofy with the speedometer in the car it came from. I don't think the speedometer functioned as a tach but it sat at about 20mph. From looking at the back of my ribbon board I see one part that may be an issue. There is a kink in an edge of the board where the missing black piece should be protecting. This kink hit the upper most ground line. Ill be doing some more research on this to find exactly what that ground is for and working from there.

 

Im having this same issue...... ill try to get a video of it mine does the same thing when im idling my speedometer jumps to about 20mph but its intermitent. When i accelerate and reach the speed its at idling at is just acclerates like normal.

 

The 1.5 Volts DC is definitely coming from the cluster. The variable here is whether or not that is actually supposed to be there. I assume it should not be there since the speed sensor should be generating an AC signal in those two wires that run to the speedo. If not I would know to be looking at the the circuit board for a component that is bleeding voltage.

Another question is, what signal actually drives the Tach? I would think its an ignition coil wire but I cant find exactly what it should be from to see if the signal wire or circuit is disturbed in any way. I know the Tach signal goes into the speedo from somewhere and then back onto the ribbon board to the tach. Somewhere along the line the Speedo is grabbing Tach input since it goes up and down with RPM and I think this would either be from a bad wire/connection or a bad ground.

 

The tachometer signal comes direct from the ECU terminal 2B.

 

I checked this pin and the speed input to the ECU for continuity to the cluster and it checked out. I pulled the speedometer out of the cluster and see no signs of a malfunction on the board such as a blown or leaky capacitor. This leads me back to whether or not that 1.5V DC is supposed to be there in the speed sensor lines.

I reassembled my cluster and am going to install tomorrow with the addition of a ground wire on the Tach to see if it is that simple. If that doesn't do it I will sub in a working speedo from Andre if that offer is still on the table. Last idea I have would be to re-route my coil trigger wires back through the body harness as if the car were stock because my routing through the engine harness is possibly creating noise.

 

Sure no problem, just let me know how it's going, I will send just the speedo unit by itself, probably can ship cheap in a flat rate little box.