So im fixing an early period avenger coupe with a 2.5L 6g72 engine, auto tranny. had a stuck shut thermostat and needed a new idle air control unit. did all that, got it running again BUT the speedo doesnt read. So i pull the cluster and pop out the speedo to bench test it. the needle deflects when I jump 12v directly across the stepper motor terminals on the individual gauge. so does it work? what commonly takes a dump? is it the output sensor or the cluster typically? i know i only tested the stepper motor and the actual speed signal has to go through many other circuits before it gets fed to the stepper...but the stepper is good at least lol. will obd1 tell me vehicle speed if i hook it up and drive with it. that way it will tell me if my sensors are reading and its the gauge not responding/wiring. also, where does the computer get its speed reading from? is it a single sensor like the output sensor or is it somewhere else and or a combination of sensors. i assume it goes to the pcm first then the pcm feeds a signal to the speedo. if anyone has any info on the gauge cluster pinouts and stuff like that to put me in the right direction i am looking for the info. the cluster is made by nippon-denso. i know the usual case with these things is to just junk it and pull a different one out of the junk yard but i hate doing that if one stupid gauge is the prob and it makes you scrap all the good ones too. it seems there are 4 connections to the speedo gauge via the mounting screws, three are attached to the traces and one is not. i can talk about it all day but im sure someone has been through this already. i'd like to repair the single speedo gauge if the sensors are good which i need to find out first i guess but if its the gauge i'd like to repair it by itself and i'd need some circuit details which i have no reference to really start besides prodding around with an ohm meter.

sorry for not posting in the old avenger subdivision, admin can move the thread right?

 

there are speed sensors in the tranny , Output speed sensor will cause no speedo

 

Changed the input and output sensor on the tranny. The speedometer reads accurately up to 40mph but will not respond to speeds faster than 40. I hooked a volt meter up to the speedo signal pin and grounded the nagative probe to the metal plate where the cluster mounts to read the signal output. the meter was set to scale 2v AC or less and as I drove around it put out mV values in scale increments of the miles per hour, like .025V is 25mph, .008V is 8mph, etc. the meter was displaying predictable and consistent values UP UNTIL 40mph(.040V) then it would not break past .040V when i was going well past 40mph...so even with new input and output sensors on the tranny. the signal is incorrect. the gauge is doing its job but is not being fed the right signal. I talked to a dodge technician at a dealership and he said that on avengers without a dedicated speedo sender attached to an extension housing of the final drive(pre-95 models have this setup). then the Right Front ABS Sensor on the wheel hub is the dedicated sensor for the newer setups. So I'm in the process of changing that ABS sensor. it is required to remove the passenger side quarter panel to access the connector anchored to the frame, there is a bolt holding the abs sensor on which for me; snapped off. Untill I drill out the rest of the bolt and re-thread the hole. It wont make this fix easy.

 

My understanding is that the signal is pulses so voltage means little to nothing if it reads to forty , I'd guess the speedo might have an issue (try swapping cluster) I've seen three or four other cars with similar troubles be clusters

I found NOTHING in the manual mentioning the ABS sensor being died into the speedo

 

Exactly, i couldnt find any wiring diagrams or anything suggesting the source of the speedo signal, but a dodge technician told me over the phone the right front ABS sensor (a magnetic pickup) provides the signal. i'm guessing its one of those things they want to keep a mystery so you are required to take it to a dealer for a regular milking. so im in the process of removing and replacing the ABS Sensor, as with most ABS sensors the damn thing is stuck in the mounting slot on the steering knuckle, the center magnet the coil is wrapped around is made of a metal that is harder than my hacksaw blade. i'm probably gonna tear out the knuckle today, blast it with some fire, smack it with a hammer, or both. better yet while the abs sensor is disconnected i could slap on the wheel again and take it for a drive to see if i get the same weird signal output. Im doing this for my gf's friend, money is involved and i want to save her the trouble of buying unnecessary parts, then again im gonna have to swap parts to see where this signal comes from. i hope this is good info for this forum. half the stuff i know comes from forums.

 

The SIGNAL is from the OUTPUT SPEED sensor I KNOW that , if it reads till 40 that suggests that the speed sensor is OK you have a cluster issue

 

The problem has been resolved as follows: I bought another Speedo cluster off ebay for $24 with a cracked lens. swapped the good lens off my other one to have a fully functional cluster. I plopped it in, took it for a spin and it reads accurate all the way with better response to acceleration. WOW you mean after all that heartache it was just a bad cluster!? i could have just replaced the cluster and then palm-faced myself for the rest of the time i spend on this WITHOUT even needing to touch an ABS sensor, which if you've been following along i destroyed a perfectly good one because some dodge mechanic told me and swore up and down that was it. Since i had no clear basis on how this car's speedo sensor setup worked (had no dedicated sender on my avenger model). Dodge doesnt want you to know that stuff anyway and it is not in a hayne's manual. i started troubleshooting from scratch. but there were three assumptions that i made which turned this into a nightmare. (1) The dodge mechanic is always right. NOT! do not believe everything you hear unless you get multiple cases of the same information from other places too. I took this one guys word over the phone because i esteemed him an almighty Dodge Technician...dont do that because i ended up destroying a $200 sensor and needing to dismantle a steering knuckle to do so. (2) Do not assume the nature of the speedo signal as simple linear voltage output from measuring it with a simple Voltmeter. The signal is a complex digital pulse waveform which can be detected using a 2V AC scale on a voltmeter BUT the meter is only measuring the average voltage of the signal and that is fairly irrelevant to diagnosis unless you are getting no voltage at all. even if you see some sort of intelligent voltage signal, that doesnt suggest that reading is the actual mechanism. to see the mechanism you would have to observe the actual waveform with an oscilloscope. (3) Do not take the owners word that the speedo cluster worked fine before and therefore cannot be the culprit. this was not my car, i was fixing it for my girlfriends friend. she told me the speedometer worked fine before. i said "are you sure" and i kept saying "are you sure". there was a point where i accepted her word as being intelligent and knowledgeable and disregarded the cluster as a priority. BIG NO-NO! Its a sad fact of our society that most chicks dont know a damn thing about cars and you cant take their word seriously about most behavior details and i would be tickled pink to see more ladies step up to the plate, cuz thats hot! this chick just bought this car, and she bought it with a bad IAC which idled the car around 1500 rpm...and tried driving it a few times. i can understand why she might have thought the speedometer was working because it measures up to 40 somewhat, but if you go past 40 the needle drops. When youre this girl trying to drive this car that idles twice what it should, youre going to be scared to go faster than 40 and see the speedo is actually broken. That being said when it comes to troubleshooting speedo or gauge problems, unless you can pinpoint a direct sensor to that gauge it is always worth your time to get another cluster FIRST. that way you have 2 references for measuring the behavior of the sensors, the sensors that work will also work for the other gauge or not and you can say "well this gauge is responding correctly therefore it must not be the sensor", throw out the bad cluster and call it a day.