When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
I'm working on a daughterboard PCB replacement for IC6 on the speedo board, but it will be a few months before it's fully developed and tested. I had the first batch of prototype PCB's delivered today though. This will fit into space behind the speedo gauge face and the existing PCB, connected via 9 flying leads.
So we decided to replace a few caps (not all) and reflow joints on the speedo PCB. The needle now moves, but not as it should. And we now have a new problem...
The "Add coolant" warning buzzer (and light) is constantly on!
We have checked the coolant level sensor and cable - Its OK. I even tried grounding the sensor cable directly - still buzzing!
So:
- Water OK
- Sensor OK (not 100% sure, but bypassing it to GND should work, that's what the sensor does...)
- Cable from sensor to speedo OK
Is this prototype daughterboard of yours a replacement for all of IC6, or just parts of it? The PCB trace for the coolant level sensor seem to end up in IC6. I'm thinking there is something going on in the speedo PCB now messing up the coolant level logic. But Im not really sure where the signal coming out of the speedo to the F100 / CPU2 is...
The "coolant level sensor" seem to be coming in this way into TR4 and IC6.
Is the speedo PCB multiple layers, or just front/back? I measured continuity for the green arrow on IC6, but I don't know where this goes. And I'm not sure what TR4 does here.
ALIENR2's schematic with dots to highlight junction points
I modified ALIENR2's schematic to highlight the junction points with dots, labeled IC5, and added Pin 9 to IC6.
I have to admit that as simple as logic circuits go, the spaghetti that surrounds it often gets the better of me. I find it unusual to have all that circuit conditioning for IC4 Pin 3 and Pin 5. However, I am surprised to see relatively low resistance values to isolate Pins 1 & 3, and how the circuit uses +5V power as a reference voltage (probably 2V) to Pin 2 as a result of the R28 & R29 leg. I am equally surprised not to see any additional support circuitry around TR8. I'm not trying to dispute the circuitry just commenting on my deteriorating ability to recall electrical theory. Regardless, ALIENR2 (Jeff) has done a fine job at tracing out the schematic from the physical board and mapping out the ribbon connector!
Additionally, Jeff, you are correct about IC3. That is where the car's odometer is stored. It is possible to remove IC3 from a bad board and install it into a replacement speedo board. DaleClark discussed it here - Post #9 with his link in Post #10. I will caution anyone who may want to alter IC3's memory (or swap it entirely). Altering the Odometer, aka odometer fraud, is illegal*. Shall we not devolve this thread into a discussion about odometer alterations? However, it is important to understand how mileage is recorded and what pins trigger the storage register in IC3's non-volatile memory. According to Dale's post, R15 & R17 on the speedo board play an important role in converting a speedometer between metric (km) and imperial (mi) units.
* - Depends upon disclosure statements and/or local, state, and country laws.
I drew out the full schematic during a repair for a friend. Hopefully this helps some people.
I drew out the full schematic during a repair for a friend. Hopefully this helps some people.
I can't help but notice differences between the circuit originally posted by ALIENR2 then massaged by Gen2n3 and the relevant section of your diagram. The differences that immediately pop out are:
1. The connector pin numbers don’t match, e.g. the Gen2n3 TAM is at pin 6 you have it at pin 10;
2. The Gen2n3 TR8 collector is connected to connector pin 4 (Tach Signal), you have it connected to an unnamed pin 13
3. The Gen2n3 TR8 base connects directly to IC4 Out1, you have it connected to IC4 In1+
4. …. And I could go on ….
Reverse engineering a circuit board is difficult and prone to error. I have spent a bit of time in the past poking around IC4 and did not find anything to contradict the Gen2n3 diagram.
I can't help but notice differences between the circuit originally posted by ALIENR2 then massaged by Gen2n3 and the relevant section of your diagram. The differences that immediately pop out are:
1. The connector pin numbers don’t match, e.g. the Gen2n3 TAM is at pin 6 you have it at pin 10;
2. The Gen2n3 TR8 collector is connected to connector pin 4 (Tach Signal), you have it connected to an unnamed pin 13
3. The Gen2n3 TR8 base connects directly to IC4 Out1, you have it connected to IC4 In1+
4. …. And I could go on ….
Reverse engineering a circuit board is difficult and prone to error. I have spent a bit of time in the past poking around IC4 and did not find anything to contradict the Gen2n3 diagram.
Do you care to have another look at your diagram?
Are you referring to the J1 (?) connector on the left in the diagram? "J1"? Connector?
Are the pins here not reversed? TAM on PIN 10 is PIN 6 from the other end. PIN 9 (Coolant) is PIN 7 from the other end. This matches other information I can find.However, this post states the other way around...?
It also seem the C9 capacitor is not the correct rating. It should be 0.33uF 50V to my understanding (330nF).
I'm also getting a bit confused reading this schema. The trace for coolant level does not make sense to me, the coolant goes straight to IC5 in this schema, that does not match my tracing.
Are you referring to the J1 (?) connector on the left in the diagram?
Yes that one.
Originally Posted by pasen
Are the pins here not reversed? TAM on PIN 10 is PIN 6 from the other end. PIN 9 (Coolant) is PIN 7 from the other end.
That would explain the J1 terminal number mismatch. The speed sensor terminals 8P & 8P1 also line up when reversed.
It appears to be carsonsargod's first post and I don't want to go too hard on him. We want people on the forum to share their hard-won knowledge and an undertaking on this scale is one I would never attempt.
As a first draft it is bound to have errors and it will take more than a few iterations before it starts to become reliable so, in the mean-time approach with a lot of caution.
That would explain the J1 terminal number mismatch. The speed sensor terminals 8P & 8P1 also line up when reversed.
It appears to be carsonsargod's first post and I don't want to go too hard on him. We want people on the forum to share their hard-won knowledge and an undertaking on this scale is one I would never attempt.
As a first draft it is bound to have errors and it will take more than a few iterations before it starts to become reliable so, in the mean-time approach with a lot of caution.
Yeah, but as far as I can see, for my needs on the coolant level warning the trace seem wrong. I measured continuity to IC6, not IC5. Dont see a trace to IC6 for Coolant in the drawing. But a drawing would be really helpful
I can't help but notice differences between the circuit originally posted by ALIENR2 then massaged by Gen2n3 and the relevant section of your diagram. The differences that immediately pop out are:
1. The connector pin numbers don’t match, e.g. the Gen2n3 TAM is at pin 6 you have it at pin 10;
2. The Gen2n3 TR8 collector is connected to connector pin 4 (Tach Signal), you have it connected to an unnamed pin 13
3. The Gen2n3 TR8 base connects directly to IC4 Out1, you have it connected to IC4 In1+
4. …. And I could go on ….
Reverse engineering a circuit board is difficult and prone to error. I have spent a bit of time in the past poking around IC4 and did not find anything to contradict the Gen2n3 diagram.
Do you care to have another look at your diagram?
Thanks for taking a look at the schematic I sent in. I am not an RX7 owner or even an automobile enthusiast, I'm an electrical engineer who did a repair for a friend. Of course I realize the difficulty or reverse engineering circuitry, although this PCB was not too difficult other than the annoying white solder mask. That doesn't mean I didn't make any mistakes. Unfortunately I no longer have the PCB in my possession for verification of any of the discrepancies that you noticed, it has been returned to the owner, or I would gladly double check and verify the issues you pointed out. I have done my best to review what I can via my own photographs and online content.
1. Pin 10 vs pin 6 on the TAM is clearly explained based on where you start counting the pins, which are clearly labeled on the PCB. It is possible I numbered them in reverse.
2. I show the collector of TR8 on Pin 12, not 13, which would also be Pin 4 if reversed. Your evidence is starting to pile up against me. I have relabeled the pins in the reverse order which agrees with the consensus on this thread.
3. I reviewed TR8 and found that the earlier versions of the schematic agree with your point here, as does the logical function of the circuit. Correction applied.
4. I am open to hear any further comments. I certainly don't want to mislead anyone, I would rather not post anything at all than do that!
I am reviewing the other comments but will post this for now.
In Pasen's image the pin numbers are clearly labeled, pin 1 being on the inside of the board and pin 15 close to the edge. I have a photo of the revese side of this board which clearly show pin 1 going to the anode of D2, pin 14 to the ground plane, and pin 12 to the collector (middle pin) of TR8. Can someone please point me to the alternative pin numbering source that everyone but me finds so obviously? Thanks!
Last edited by carsonsargod; Jun 1, 2026 at 10:35 AM.
Based on your May 25th image, I can see the pin numbers and they appear to be correct to what I have. I am not sure where the reverse pinning came from. I may be wrong about this but I'd really like someone to show me. In your photo I see pins 1 and 15 marked clearly, 15 being closes to the edge of the board. I have a photo from which I can see the traces and they seem correct to my schematic...so I'm a bit unsure where this other pinning originated.
I think you are correct on C9. I can't verify this because I no longer have the PCB, but I checked my parts order from when I recapped the board and I did order a 0.33uF 50V. Thanks for pointing this out.
In Pasen's image the pin numbers are clearly labeled, pin 1 being on the inside of the board and pin 15 close to the edge. I have a photo of the revese side of this board which clearly show pin 1 going to the anode of D2, pin 14 to the ground plane, and pin 12 to the collector (middle pin) of TR8. Can someone please point me to the alternative pin numbering source that everyone but me finds so obviously? Thanks!
This is not something that I had given any attention to previously and I don’t have a cluster at hand to check anything. Mine is back in the car, now working faultlessly.
Looking at photos from the Flexible Printed Circuit connector side of the board, pins 1, 2, 14 & 15 are clearly marked and are the reverse of those in ALIENR2’s post so it seems your pin numberings are the ones we should be using..
Maybe add a note to the effect: Pin Numbers are as viewed from the Flexible Printed Circuit connector side of the board. Other schematics use different number schemes.
@carsonsargod Did you change the format of the schematics? I can barely read the V4 document of yours, text is a mess for some reason.
When you say pin label, you mean these? I did not notice this earlier! Pin label
This would confirm this previous post is incorrect (reversed) and what got me confused.
And the below image seem to have the correct pin order then (15 near the edge), seen in post #158 and matching the order but the labels/names don't match up to your new V4 at all (but your old..). I don't know who is right or wrong here to be honest, but I believe your initial schematics was not reversed.
This is not something that I had given any attention to previously and I don’t have a cluster at hand to check anything. Mine is back in the car, now working faultlessly.
Looking at photos from the Flexible Printed Circuit connector side of the board, pins 1, 2, 14 & 15 are clearly marked and are the reverse of those in ALIENR2’s post so it seems your pin numberings are the ones we should be using..
Maybe add a note to the effect: Pin Numbers are as viewed from the Flexible Printed Circuit connector side of the board. Other schematics use different number schemes.
Done. Attached V5 with pins reverted to how I had them originally as numbered on the PCB. Fixed your note on the IC4 (#3). Also correct C9 to 0.33uF. If any other corrections are needed I will happily do my best to make them. Certainly don't want to mislead anyone, the entire point was to share and make other people's lives easier
@carsonsargod Did you change the format of the schematics? I can barely read the V4 document of yours, text is a mess for some reason.
When you say pin label, you mean these? I did not notice this earlier! Pin label
This would confirm this previous post is incorrect (reversed) and what got me confused.
And the below image seem to have the correct pin order then (15 near the edge), seen in post #158 and matching the order but the labels/names don't match up to your new V4 at all (but your old..). I don't know who is right or wrong here to be honest, but I believe your initial schematics was not reversed.
Yes, a lot of confusion over the numbering of the flat cable connector. Previous documentation on here has the numbering reversed from that actually marked on the board.
Yes, first production board has been assembled and fitted. I can't believe it all fitted and worked perfectly! I was fully expecting some level of re-design, even after all the development effort. Rather than repeat everything here, this is a link to the UK based RX-7 FD forum that I use to update FD owners on my work on repairing FD clusters; https://fdoc.co.uk/forum/threads/das...-repairs.7601/
However, here are a couple of photos, one showing the IC6 motherboard in place. The biggest challenge is to carefully remove the failed IC6 chip without damaging the PCB and then soldering in 9 wires to connect to the motherboard. My professional vacuum desoldering tool did the job admirably though! The blue variable trimmer resistor sets the zero position. The existing variable trimmer resistor on the Mazda PCB remains unchanged in its function to calibrate the speedo.
I can assure you 100% that the sole function of IC6 is to drive the speedo gauge motor. It has absolutely no connection to the coolant level warning
I am not selling the daughterboard, only providing a service to install it in clusters where IC6 has failed. I repaired a cluster I was sent from Sweden recently, but the sender was unaware of the complexities of shipping items in/out of the EU to the UK and ended up paying more in customs duty than the repair cost itself .
Any idea on why the new schematics is so "pixelated" for me? Or is it just me?
V5 V1/old
Sorry about that. I don't know what I did the first time that was different, but it's definitely not what I had to do this time to get it super high-res again. Sorry for the delayed response.
I can assure you 100% that the sole function of IC6 is to drive the speedo gauge motor. It has absolutely no connection to the coolant level warning
I am not selling the daughterboard, only providing a service to install it in clusters where IC6 has failed. I repaired a cluster I was sent from Sweden recently, but the sender was unaware of the complexities of shipping items in/out of the EU to the UK and ended up paying more in customs duty than the repair cost itself .
Well, that is the exact reason I would rather not send stuff back and forth to the UK. Have bought a new/tested cluster now. The old probably have IC6 issues as well, but no point in trying to fix it in the current state. Coolant level warning needs to be sorted. But after I re-flowed solder joints and replaced C3, C4 and C6, the speedo needle moves unpredictable and having its own life. Is this the sign of a failed IC6?
@carsonsargod Is the BUZ output in your diagram? I cant seem to find it. "BUZ" Buzzer output?
This is the coolant level sensor input to IC5 according to the drawing. When the sensor is grounded, all should be fine, according to the serivce manual.
How does this sensor circuit work? Can someone more knowledgeable explain? What happens when the sensor no longer grounds via the coolant?
Schematic
.
