348 Speedo Burned Internals

I can see that the link from diode to the ground (long red lines on the pic) is browned which suggests it was carrying high current to the ground. Also, there appears to be a small crack in the green stripe on the "after" picture.

It is quite likely that the small transistor (BCX19) has also failed as its collector is directly linked to the base of the big blown transistor. With regard to the big IC, it will be very difficult to de-solder. An easier way is to cut its legs first and de-solder them one by one or leave them in and solder the new IC onto the cut legs. On the other hand, you may consider leaving the old big IC in as the problem (short) seems to be in the power supply part of the circuit so there might have been no damage to the IC (but it is a risk).
Image Unavailable, Please Login
Image Unavailable, Please Login

 

The "diode" could actually be a Transient Voltage Suppressor (TVS). The table below suggests that but it does not show the combination blue-green (or yellow?). For the first stripe blue, it shows that it is TGL 41 series TVS (see the attached datasheet). These are available for various breakdown voltages; how do we find out what voltage is the one in the Speedo?

If indeed it is a TVS, then I would guess it would be some 16 to 20 volt to protect the circuit from voltage spikes like the ones produced by disengaging relays.
Image Unavailable, Please Login
Image Unavailable, Please Login

 

Wade, I'll post more numbers later but I'm on my phone so going to focus on the obvious right now. Comparing your voltages to mine does indeed reveal an anomaly. With the key on, the red/black wire in my car shows battery voltage minus a bit (the resistance of the windings in the sensor?) with the sensor connected, but only .01 volt with the sensor disconnected. Without being able to see where that wire goes I'm purely guessing since both of our speedos are not working, but my WAG is that the problem is in your car. I base that on the fact that the green/white wire has +12VDC applied per Miltonian's diagram, so I can't see why the other sensor wire should be showing voltage when the sensor is disconnected. Again, that's a WAG but I wouldn't connect up the new speedo until we can find out more.

Another bit of news coming out of our voltage tests is that the white/yellow wire seeming supplies +5VDC with the key on. At least that's what my tests show, whether the sensor is connected or not. Your voltage is a tad lower but your battery appears to be pretty low. No idea where the regulated supply might be located though. Sure wish we had the diagrams from the repair manual, because as of right now I'm not sure how we resolve the disparity in voltage numbers on the red/black wire.

More later...

 

Below are the various measurements I took from my '91 US spec car. Note that the wire I identified as W/Y is in the same place and seems to deliver the same five volts as the Y/W wire in Wade's car, but, being pedantic, I call mine W/Y because the yellow color is almost imperceptible.

348 Speedometer Harness Measurements:

Wire color code:
G/W: Green with white tracer
Blk: Black
Y/R: Yellow with red tracer
R/B: Red with black tracer
W/Y: White with yellow tracer

Resistance between wires at speedo connector with speedo, sensor, and battery disconnected:

G/W-Blk: 42.5 ohms
G/W-Y/R: 46.8 ohms
G/W-Red/Blk: infinite
G/W-W/Y: infinite
Blk-Y/R: 5.0 ohms
Blk-R/B: infinite
Blk-W/Y: infinite
R/B-W/Y: infinite

Resistance of wires at speedo connector to ground with speedo, sensor, and battery disconnected:

G/W: 43.1 ohms
Blk: .3 ohms
Y/R: 5.5 ohms
R/B: infinite
W/Y: infinite

Voltage at speedo connector with speedo and sensor disconnected, battery connected, key off:

G/W: 0 volts
Blk: 0 volts
Y/R: 0 volts
R/B: 0 volts
W/Y: 0 volts

Voltage at speedo connector: speedo and sensor disconnected, battery and key on:

G/W: 12.5 volts
Blk: 0 volts
Y/R: .04 volts
R/B: .01 volts
W/Y: 5.0 volts

Resistance between wires at speedo connector with speedo and battery disconnected, sensor connected:

G/W-Blk: 40.9 ohms
G/W-Y/R: 45.9 ohms
G/W-Red/Blk: 5.5 megaohms
G/W-W/Y: infinite
Blk-Y/R: 5.5 ohms
Blk-R/B: infinite
Blk-W/Y: infinite
R/B-W/Y: infinite

Resistance of wires at speedo connector to ground with speedo and battery disconnected, sensor connected:

G/W: 40.6 ohms
Blk: .2 ohms
Y/R: 5.3 ohms
R/B: 4.15 megaohms
W/Y: infinite

Voltage at speedo connector with speedo disconnected, battery and sensor connected, key off:

G/W: 0 volts
Blk: 0 volts
Y/R: 0 volts
R/B: 0 volts
W/Y: 0 volts

Voltage at speedo connector: speedo disconnected, battery and sensor connected, key on:

G/W: 12.83 volts
Blk: 0 volts
Y/R: .04 volts
R/B: .12.3 volts
W/Y: 5.0 volts

My supposition of the purposes of the five wires at the speedometer connector:
G/W: from sensor, +12VDC
Blk: ground
Y/R: speedometer backlight positive voltage source (does not go to speedometer board)
R/B: second lead from speed sensor
W/Y: +5VDC source for speedo board electronics

Resistance across original 348 speedo sensor leads: 3.76 megaohms

Resistance across new 348 original style speedo sensor (PN 116987) leads: 5 megaohms

 

The sensor is not a simple induction type (with just a coil) but an electronic type with a couple of transistors inside, thus your Megaohm resistance measurement.

 

Thanks, Miroljub. I'd seen that written in another post but didn't have any idea if it was true or not. Perhaps that helps explain why +12VDC is applied to one side of the sender.

 

We really, really, really need a shareable copy of section 5 of the wiring harness.

Forgot to mention that in a previous test I'd checked continuity from the sensor connector to the speedo connector and found it to be okay. Need to check it again though, with an eye toward measuring any resistance, although I can't imagine what that would tell us.

Also forgot that the red sensor wire trace apparently changes color at some point, so perhaps there is another component inline between the sensor and speedo. I've been thinking that the red sensor wire in your car is somehow contacting a hot wire inside the harness (I've seen this before when a positive wire shorts, gets hot enough to melt insulation, and winds up with conductivity between itself and other wires in the same loom) but maybe the positive voltage is being introduced by another component, or wherever the wire changes color?

Or maybe it's my car that's screwed up, instead of yours. After all, my speedo doesn't work either. But I can't see why voltage would be supplied to that sensor wire when the key is off. Besides, your car is toasting speedo boards and mine isn't, so seems to me there's reason to believe that there's an issue with your R/B wire.

Like I said, we really, really need a copy of section 5. Guess I'll try to bump that post later. Surely there are people with a hard copy of the manual. Just need to find one who is willing to donate a few minutes of their time to the cause... Wish I had more to go on before leaping to the assumption that my speedometer is bad.

 

Wade-- one more long shot thought: the +5VDC at the white and yellow wire has to be coming from someplace. Perhaps the red sensor wire passes through whatever is regulating that 5 volts? You might check the wiring on the back of the footwell fuse panel (and anyplace else you can think of) to see if you can find a Y/W wire and then see if there are any R/B or R/W wires going to the same component. Like I said, it's a long shot, but without a schematic that's about all we have.

If you have replacement components on hand I'd just do as I suggest in my PM and repair the board, then test it without the R/B wire connected. If it doesn't blow run a temporary replacement for the R/B wire and see what happens... That doesn't do anything to find where the mystery voltage is coming from but it would at least help confirm the theory that the voltage on the R/B wire is what's causing the smoke to escape from your speedo board.

 

Think they'd take $4500 even?

 

The voltages you read on the red/black wire even when the speedo and the sensor are disconnected is probably just some "ghost voltage" feedback from the ECU as, I believe, the the signal from the speed sensor also goes to the ECU. A way to verify this is to check with the ECU-s disconnected.

 

I'm guessing that the G/W wire picks up +12VDC from the fuse box, perhaps from a separate wire, since there are two going to the connector for the speedometer. A logical assumption would be that one G/W wire goes to the sensor and the other goes to the fuse box... which is probably completely wrong since logic doesn't seem to be applicable here.

What I'm curious about is where the +5VDC delivered by the Y/W wire originates. Maybe there's a module in the fuse box that serves as a +5 volt regulator, and the other speedometer wires might also go to/thru that module. It's a long shot, but in any case that +5VDC has to be coming from someplace.

 

Probably from the ECU.

 

Wade, referring to your post (which has just disappeard) with a picture showing the underside of the circuit board and your suggested connection of the red/black wire to the MELF. Did you not accidentally switch the orientation? Perhaps the red/black goes to the pin as shown below.
Image Unavailable, Please Login

 

The function of the big IC pins should be as on the pic. This is according to some typical circuits of the chip.
Image Unavailable, Please Login

 

Miroljub, do you have any suggestions for the next step in attempting to diagnose my own problem? As a reminder, the speedo needle started bouncing one night as I neared the house. The next time I started the car the speedo & odo were both turning while the car was stationary (showing ~40 mph, but erratic). I replaced the sender with a new original style sender, after which both the speedo & odo worked perfectly for one trip around the block. They have not worked since. I have confirmed that there is a clean square wave getting to the G/W & R/B wires at connector on the back of the speedo. I have confirmed continuity between the connector on the back of the speedo and the pins on the speedo board. I have replaced the topside electrolytic capacitors as preventative maintenance. No components on the board look or smell burned. I cannot see any bad solder joints on the board. I'm now stumped...

Cheers,

///Mike

 

In your case, could be the big IC (cheap and easy to get). However, we have not established what the smaller 8 pin IC is. I couldn't find any datasheet using the numbers seen on the chip.