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.
Are you tired of watching your tachometer bounce around like crazy? This should increase the search visibility to help people with a jumping or bouncing tachometer.
I'm not sure I've cured my jumpy tach needle yet (more testing needed), but so far I have tried:
1. Reflowing the solder of any solder joints that look suspicious
2. Replacing the screws that both (1) mount the tach and (2) provide the power/ground/signal electrical connections to the tach with shiny like-new screws I pulled out of a '95 Nissan Pathfinder (which has the same tach circuit board but some different components); this will help if surface corrosion of my stock screws (which do not look shiny and new) is interfering with good electrical contact between the instrument console's flexible circuit sheet and the screws or between the screws and the tach board hardware.
3. Replacing C1 (1uF 50V) even though it looked fine. I note that C1 is a power supply bypass capacitor (i.e. it connects between power and ground for the tach board to filter out noise that can easily cause chips like the AN8352UBK motor driver chip to glitch). I note that the Pathfinder's tach board had a 1000uF 16V capacitor for it's C1. I speculate based on the difference in C1 and other component values that the Pathfinder tach board may be using 12V power (while I measured ~7V power from my speedo board to my tach board), but I say the larger the bypass capacitor the better, so I replaced my 1uF C1 with a 100uF capacitor and I expect corresponding much greater than stock power supply noise reduction which may prevent motor driver chip/tach needle glitches. I note that the 1000uF capacitor C3 on our speedo boards is connected to the 5V output of IC2 and is not directly connected to the power supplied to the tach board, so C3 may not be providing noise reduction to tach board power.
Last edited by ALIENR2; 09-18-17 at 05:30 PM.
Reason: clarity of wording
Finding a replacement for TR4 is proving to have a similar result as the AN8352x/AN8363x components. However, this is what I found:
NEC semiconductors was absorbed into Renesas Electronics. I visited their website and discovered that the 2SB1217 transistor is obsolete with no replacement. They list it as "EOL" which I assume means "End of Life"? Here is the link to that page: 2SB1217 Transistor (TR4).
Additionally, I cross referenced TR4's characteristics on DigiKey. They listed 1 alternate component for TR4 however, it is obsolete and cannot be purchased. The one cross-referenced component is "KSB1149", made by ON Semiconductors (formerly Fairchild Semiconductors). The data sheet for KSB1149 is attached below.
Finally, I reached out to ON Semiconductors for any possible replacements for KSB1149. Since AN8352UBK and AN8363UBK were mentioned, I also sent an email to Panasonic regarding replacements for these two IC chips. I hope to have positive news when both companies reply to my inquiry.
I'm not sure I've cured my jumpy tach needle yet (more testing needed), but so far I have tried:
1. Reflowing the solder of any solder joints that look suspicious
2. Replacing the screws that both (1) mount the tach and (2) provide the power/ground/signal electrical connections to the tach with shiny like-new screws I pulled out of a '95 Nissan Pathfinder (which has the same tach circuit board but some different components); this will help if surface corrosion of my stock screws (which do not look shiny and new) is interfering with good electrical contact between the instrument console's flexible circuit sheet and the screws or between the screws and the tach board hardware.
3. Replacing C1 (1uF 50V) even though it looked fine. I note that C1 is a power supply bypass capacitor (i.e. it connects between power and ground for the tach board to filter out noise that can easily cause chips like the AN8352UBK motor driver chip to glitch). I note that the Pathfinder's tach board had a 1000uF 16V capacitor for it's C1. I speculate based on the difference in C1 and other component values that the Pathfinder tach board may be using 12V power (while I measured ~7V power from my speedo board to my tach board), but I say the larger the bypass capacitor the better, so I replaced my 1uF C1 with a 100uF capacitor and I expect corresponding much greater than stock power supply noise reduction which may prevent motor driver chip/tach needle glitches. I note that the 1000uF capacitor C3 on our speedo boards is connected to the 5V output of IC2 and is not directly connected to the power supplied to the tach board, so C3 may not be providing noise reduction to tach board power.
Jeff,
Have you de-soldered then re-soldered AN8352UBK on the tach board? One of the other members had success in fixing his tach with that process.
How did the old screws look? Unless they are green like the Statue of Liberty they should conduct electricity well. If you have concern about the electrical contact between the screws and circuit boards then maybe a little dielectric grease would help ensure conductivity and reduce the effect of corrosion.
It will be interesting to see results from your testing since the capacitance value of C1 was increased. Although, I believe the 1uF cap would provide the necessary protection of the circuit.
Have you de-soldered then re-soldered AN8352UBK on the tach board? One of the other members had success in fixing his tach with that process.
How did the old screws look? Unless they are green like the Statue of Liberty they should conduct electricity well. If you have concern about the electrical contact between the screws and circuit boards then maybe a little dielectric grease would help ensure conductivity and reduce the effect of corrosion.
It will be interesting to see results from your testing since the capacitance value of C1 was increased. Although, I believe the 1uF cap would provide the necessary protection of the circuit.
Cheers,
George
I did not reflow the solder joints for my AN8352UBK; the solder joints looked good and the Skinny DIP leads are close together so I could easily make a mess trying.
Based on photos from other cars I don't think my old screws were unusually corroded and I agree that screw corrosion is not a likely cause of tach needle glitches. On the other hand, the new screws are nice and shiny and corrosion causes plenty of problems globally so it was worth a try.
My tach is working normally (though I haven't used it enough to be sure that the glitches are gone) with the larger C1. It's possible that degradation of C1 with age leads to reduced capacitance to a point where normal power supply noise causes AN8352UBK/tach needle glitches. I see no harm in using a larger C1, but a healthy 1uF is expected to be enough.
I got replies from ON Semiconductors and Panasonic regarding TR4, IC6, and IC1 (Tach Board). I don't have good news to report. Refer to my earlier post, Post #29. In short, there are no known replacements for TR4, IC6, or IC1. They must either be cannibalized from used circuit boards or purchase a new speedo or tach.
IC6 (Speedo Board) & IC1 (Tach Board):
My inquiry to Panasonic:
I am an end user attempting to find replacement IC chips with the following part numbers: AN8352UBK and AN8363UBK. Both ICs are 22pin DIPs. Are data sheets for these components available? Are there alternative ICs for these components?
------------------------------------------------------------------------------------------------------------
Response from Panasonic:
Looks like this part is really old and they don’t have anything that they can provide. I believe this is obsolete and there is no replacement.
I am an end user researching transistors that are found on an early 90s era automotive speedometer circuits. The part is labeled "B1217" and suggests its full designation "2SB1217". Based upon its characteristics, it cross referenced to KSB1149 on DigiKey(dot)com. Is this part available or is it replaced by another component?
I must caution members that these ebay components may be available on a short-term basis. They look to be exact replacements and you must buy them in the quantity available (5 or 10 pieces). These items originate from China so there could be problems/delays in shipping or clearing customs for US members. Please refer to the earlier post for a Mouser-sourced replacement of TR6.
After researching suitable replacements for TR5 (2SC458), I cross-referenced its specs on Mouser and found this: BC548 series transistor, made by ON Semiconductors. The differences between the original and this substitute are the larger power dissipation of 500mW as opposed to the original 200mW, and a higher Gain Bandwidth of 300MHz as opposed to the original 230MHz. The original also has a Transition Frequency (Ft) of 115MHz however, I did not see that specification on the substitute. Other differences include an increase in Collector Capacitance of 6pF (original has 4pF), and a Forward Current Transfer Ratio (Hfe) of 110 (original has a min of 60).
Despite these differences, I believe that the BC548 series is an acceptable replacement. Its data sheet is attached below for further review/discussion. Specifically, the proper replacement transistor from Mouser is 512-BC549BTF (Mouser P/N), BC549BTF (Manufacturer P/N).
Onto TR9’s replacement. Much like TR5, I cross referenced TR9’s (2SC1214) specs against Mouser’s database of transistors (filtered for specific ratings). There are 2 components that are close to the original specifications. The 2 components are 2N3416 or 2N3417, manufactured by Central Semiconductor Corp.
The difference in specifications are in Collector Power dissipation (Pc) at 625mW as opposed to the original 600mW; Collector cutoff current (Icbo) at 0.1uA as opposed to the original 0.5uA; Collector to Emitter saturation voltage (Vce(sat)) at 0.3V as opposed to the original 0.6V, Base to Emitter voltage (Vbe) at 0.85V as opposed to the original 0.64V, and DC Current transfer ratio (Hfe) 75 – 225 as opposed to the original 60 – 320.
The difference in specifications are a result of different test conditions. Please refer to the 2nd data sheet attached to this post. I propose that these differences are minimal and the 2N3416/3417 transistor should function within limits of the circuit. Refer to Mouser P/N: 610-2N3416, Manufacturer P/N: 2N3416 or Mouser P/N: 610-2N3417, Manufacturer P/N: 2N3417.
Note to any Member who replaces either of these components in the future: these are proposed replacements. Please provide feedback on your repairs and/or post any research to counter these proposed replacements.
In light of research for other components such as TR5 & TR9, I decided to examine a substitute transistor for TR7/TR8. The original transistor is marked as C144. Originally, the C144 part number cross-referenced to NTE123AP. My concern with this alternate part number lay in ease of purchase. Therefore, the specifications for NTE123AP were plugged into Mouser’s transistor filter. The result is a replacement component that can be easily located and purchased. The proposed replacement for TR7 and TR8 is 2N4401, manufactured by Central Semiconductors. The data sheet is attached for review.
Note to any Member who replaces this component in the future: this is a proposed replacement. Please provide feedback on your repairs and/or post any research to counter this proposed replacement.
The parts list thread will be cross-referenced with this post to help reduce confusion in the future.
This is what I propose:
TR7 & TR8: C144 (Original); Substitute P/N: NTE123AP
TR7 & TR8 (Proposed): Mouser P/N: 610-2N4401, Manufacturer P/N: 2N4401
Could you please describe what components were replaced? Did anyone else make repairs to the board before you? What soldering tools are you using for this job?
Additional photos will be needed, especially close-in shots of the solder joints on any component that you replaced or touched up. Use isopropyl alcohol and a cotton swab to clean up the areas surrounding the solder joints on both the top and bottom of the board.
This is what I see so far:
C3 - I need a better angle to see both top and bottom solder joints. The positive (+) lead looks ok but I cannot see the negative (-) lead.
C4 - there is too much solder on the one joint. Remove all of the old solder then reapply. A proper solder joint should look like a small mountain peak. There is also a black patch underneath the capacitor. What is it?
IC1 - What is that black patch on the top side of the board?
IC5 - Looks like a small black patch is there too. Did anyone do any repairs to it?
Digital Display - Did you remove the digital display from the speedo board? Closer photos of those solder joints is required. They look rough. It also looks like 1 trace is broken next to IC1 (adjacent to the capacitor below DA2). I cannot say for certain if the other traces along IC1 are broken. A close-in shot from multiple angles may be needed.
Trace next to C3 - Again, I cannot say with certainty if that trace is broken. A close-in shot from multiple angles may be needed.
Speedometer Solder Joints - A close-in photo from a couple of angles may be needed. It looks like those 4 joints (along with the Digital Display) have some scorch marks.
To help increase your soldering confidence, check out some YouTube videos on basic solder skills, especially when soldering solid state devices such as resistors, capacitors, transistors, etc... I have a link to one in a different thread.
A few helpful hints: Use electrical solder flux. The copper flux for plumbing is different and will not help in this application. Electrical solder must form a physical and electrical bond to the eyelet and component leg. Application of heat (via solder iron) is applied to the component leg and eyelet first then add solder to the joint.
Well to sum up my situation, the ODO doesnt work but the Speedo did work.
I Replaced the big capacitor (it was blown) and the smaller ones (10uf, 50v) just in case. My odometer still didnt work. So I looked all over, used the multimeter to check connections, all the chips have good connection to the board. The only thing i saw was that transistor being TR7 and Diode DA2 gave no readings, when check with the meter. Also the connections looked a bit corroded. When I used my Desolder gun to remove the solder to DA2 and DA1, and pulled them out, the leads pretty much crumbled off. I Replaced DA2 and TR6 and still didnt work.
I Finally gave the circuit board a final check (the top side) and found two areas where the pathways got broken up. I confirmed this using my multi meter and the meter wasnt able to show a connection. See pictures below.
I think I can get a wire soldered on the back side for the pathway where all the capacitors are. However for the other pathway from small chip to the larger chip, I can only solder on the back from the smaller chip and will have to solder the wire directly to the larger chip. I will post pictures when I am finished to show what i am talking about.
Which capacitors did you replace - are they C1 (10uF, 50V) C3 (1000uF, 6.3V), and C12 (10uF, 50V)?
You mentioned TR7 didn't work but then said TR6 was replaced. Which one was replaced - TR6 or TR7?
Before you start to solder jumper wires over broken solder runs (traces), let's first establish what is broken and what is ok. Repairing broken solder runs is no easy task. In fact, I normally recommend taking the board to a professional for such a repair.
What about the digital display, was it desoldered then resoldered?
What solder equipment are you using? For example, a 30W solder iron, a 45W de-solder iron, solder wick, solder flux (electrical), 60/40 rosin core solder in .032" diameter.
Cheers,
George
P.S. Clean up the board with cotton swabs and isopropyl alcohol before taking additional photos. This will eliminate years of dried solder flux and also help in examining solder joints up close.
Last edited by Gen2n3; 09-21-17 at 11:46 PM.
Reason: Added Post Script
Which capacitors did you replace - are they C1 (10uF, 50V) C3 (1000uF, 6.3V), and C12 (10uF, 50V)?
You mentioned TR7 didn't work but then said TR6 was replaced. Which one was replaced - TR6 or TR7?
Before you start to solder jumper wires over broken solder runs (traces), let's first establish what is broken and what is ok. Repairing broken solder runs is no easy task. In fact, I normally recommend taking the board to a professional for such a repair.
What about the digital display, was it desoldered then resoldered?
What solder equipment are you using? For example, a 30W solder iron, a 45W de-solder iron, solder wick, solder flux (electrical), 60/40 rosin core solder in .032" diameter.
Cheers,
George
P.S. Clean up the board with cotton swabs and isopropyl alcohol before taking additional photos. This will eliminate years of dried solder flux and also help in examining solder joints up close.
So I relooked over all the parts and I feel dumb for doing it and saying it but I replaced C1, 2, 4, 6, 11, 12, 13 with 50v 10uf capacitors, haha. Luckily I still had most of them from when I removed them. I tested all over them with my meter and they are still good. The only ones I need to order are C6 and C13 for 50v 1uf. I put all the original capacitors back where they were.
My bad about the TR6 and TR7. I replaced TR7 with NTE123AP, however I am not sure which is the correct way to mount it. Is the emitter side when looking at the flat side of the transmitter the right or left? Various pinouts show one or the other. So I replaced TR7, TR6 and DA2. TR6 and DA2 i do not have the part numbers, but if you refer to the thread for the part numbers i posted a link to an ebay ad that were similar replacements, or so it looked like. DA1 i could not find a replacement but according to my meter it seems good, when check both sides to the middle leads, so i reused it.
Yeah I removed the display with my desolder gun to check around it, all looks good and the joints on both side are connected with solder.
I actually connected the broken pathways on my own. Wasnt too bad, will post pictures later. I checked with my meter with the surrounding joints where I solder the wire on to make sure it wasnt touching anything else.
So as of now, after I put the original capacitors I still had back in, reconnected the broken paths, I am waiting for Capacitors for C6 and C13, 50v 1uf, but left in the 50v 10uf capacitors. I hooked it up to the car just to see it would work and still nothing, so now I guess those capacitors are probably the deciding factor? and maybe it I can figure out the correct way to mount C7 for the emitter side.
Edit: I am using a 25W solder iron, using rosin core solder, my desolder gun is rated at 100W. After I solder the joints, I go back over it with my desolder gun, to melt it a little more so the solder will flow through the joints.
Here are pictures. I actually found a 3rd broken pathway. So in total I had 3 bad paths from what I saw that got separated. This could be from me, when I cleaned, I use a tooth brush and qtips now with electrical cleaner. All are now hard wired in.
My Tools
Back view of my cluster
Front view of the cluster with the speedo/ODO removed
When using the desoler iron, this is what it looks like after it heated up the solder and sucked it out.
Picture of the paths hard wired in. I first could only find a small gauge wire with multiple wires in it, then later found one with a single wire. I will go back and glue the areas i hardwired to make sure no shorting occurs in the future, but for now just trying to get this to work.
3rd pathway I found broken, its the path going out from the right side of DA1.
Checking connections and making sure the surrounding connections are not touching
Wow! I’m surprised that you replaced all those capacitors with 10uF, 50V capacitance values. The broken traces may be the result of over-current caused by using the same value of capacitors, or damaged by heat transfer with soldering irons. May I suggest that you replace all the capacitors (C1, C2, C4, C11, C12, and C13) with new ones? Please make sure that you order the proper values! New components will ensure their tolerance values are within specifications and should last longer than the originals that were reinstalled. May I also recommend following the parts list for these components here: https://www.rx7club.com/3rd-generati.../#post12217993.
Thank you for identifying TR7 as the replacement transistor. The best way to ensure proper alignment of TR7’s pins is to refer back to its data sheet. Photos are also a good indicator. Luckily for you, photos of a working board exist here: https://www.rx7club.com/3rd-generati...nents-1112010/. Yes, the same thread also acts as a parts locater in addition to a parts list. Basically, the overall shape of TR7 should help identify the proper location of its Emitter. It is shaped like a trapezoid and the outline of TR7 on the circuit board is such that the shorter length is the “top” and the longer length is the “bottom” of TR7. There is also an “E” stenciled on the left side of the board. That identifies the Emitter leg. Again, refer to the data sheet for TR7 and it will have a diagram that identifies the Emitter leg.
As for TR6, DA1, and DA2, go through the above thread link for part numbers. I recall some commentary about DA1 and DA2. TR6 was one of those ebay links that you so kindly posted on the same thread. If you don’t want to use ebay as a source then there is an SMD component that supersedes the original transistor. An adapter must be used with the SMD. Again, the information you need is found in the parts list thread. Please take advantage of it.
Thanks for posting the information about your solder equipment. Here are my observations based upon your photos. My intent is increase your awareness and help boost your confidence; not to offend or embarrass you. Furthermore, I do not know your skill level for soldering so please bear with me as I may describe things in simplest terms. Again, I do not wish to offend you.
The 25W soldering iron is low for this application. A low wattage solder iron is good for narrow gauge wire repair around 20 to 24AWG (gauge). On a circuit board, a low wattage solder iron is less efficient at keeping the solder molten long enough to penetrate through the eyelets (some people refer to them as solder cups, or cups). That equates to a longer time of heat applied to the leads of the component and the eyelet. As such, the longer heat is applied to the board then the surrounding area can be easily affected. It can char the board around the eyelet and potentially damage the thin copper trace sandwiched between the board layers. I refer to this as heat transfer damage. This type of damage could also happen during solder removal, which I’ll address shortly. For example, the 4 eyelets that the speedometer face is soldered to circuit board shows evidence of heat transfer damage. Char marks and bubbled up solder traces are clearly visible here. Another example of heat transfer damage can be seen along the digital display solder joints (back side of the board). The charring is obvious here however, it is mostly contained around each eyelet. Although there are at least 3 traces along the bottom row that show some bubbling. The 3 traces may not be seriously damaged but they would need to be examined under a microscope (or magnifying lens). On the other hand, each solder joint on the back side of the digital display is shiny, evenly filled, and forms a mountain peak. The same credit applies to the solder joints on the top side of the circuit board’s digital display. Well done! I must caution you to examine each of those joints again because of shading in one photo – the joints cannot be easily identified. Furthermore, look at the 8th solder joint on the lower row (it is actually Pin 16). The joint is misshaped, looks grey and dull, and is flat. That could be a cold solder joint. Helpful hint: do not re-flow a (suspected) cold solder joint. Remove the old solder (from the underside of the board) then apply fresh solder to the joint.
This is a good time to ask, do you use electrical solder flux? It comes in paste or liquid form. It acts like a catalyst for solder to flow through joints. Think of it as a penetrating oil like WD-40 or PB Blaster. Instead of loosening the rust bonds between nuts and bolts, the fluid seeps in between the component lead and the eyelet. Flux SIGNIFICANTLY reduces the amount of time heat is applied to a solder joint and will cause the solder to wick in between the joint to form a solid electrical and mechanical bond. If you hold a wire straight up and down with the tip pointed up then and apply flux to it then the solder would flow up the length of the tip. Just remember that plumbing flux is for copper pipes and is formulated differently than electrical flux. Flux *stress* could offset heat transfer damage with low wattage solder irons.
Moving along, take a few minutes to watch the video in this link: https://www.rx7club.com/3rd-generati.../#post12200510. It offers basic solder techniques for any solder project. Although, the presenter did not use flux. If you are looking into upgrading your solder iron, check out Sears (because RadioShack went belly up) for an inexpensive 45W solder iron. This would be a “medium” load iron that would quickly dispatch solder jobs on most circuit boards. While there, you may even find some flux!
Afterwards, examine the jumper wires installed on your board. The gauge wire is too big and the length of exposed wire is too long for each solder joint. There is a good potential of wire failure and short circuits from the exposed wire to surrounding solder joints/traces. The jumper point next to C2 looks like a cold solder joint. Again, the links above will offer great advise about jumper wires. As a reminder, clean the solder joints of a component with isopropyl alcohol after it is replaced. This is essential when flux is used because flux can deteriorate a circuit board. Do you see all that brown stuff on your circuit board? Those stains were caused by residual flux when the board was first manufactured.
The de-soldering iron you have is rather substantial. A 100W iron is definitely overkill in this application. I noticed that it has an adjustment switch on it. What setting do you use to extract solder? I would recommend a setting equivalent to 45W. It is possible that when the old solder was removed from the circuit board that heat transfer damage may have occurred. I also applaud your candor when using the de-solder iron to reflow some eyelets. However, please stop this technique of touching up solder joints. The more times a solder joint is heat cycled in a short time span then the greater the chance for damage. The eyelet itself could be damaged (lifted from the board), a trace could break, and/or burn a hole through the circuit board. The amount of time it should take to create or remove a solder joint is approx 1 second. The longer the heat is applied to a circuit board, the greater chance for damage.
I know that your intention was good when removing the digital display to inspect the components underneath it. However, its removal was not necessary. The potential to cause damage to its eyelets, long legs, and surrounding traces increased significantly because of different soldering/de-soldering iron temperatures, the time spent transferring solder to each joint, or warping the long lead legs. In addition, the original solder joints were removed and the possibility of creating cold solder joints increased. That may cause a ghost problem for you to troubleshoot. Believe me when I say that we have a hard enough time replacing capacitors and transistors! I don’t recall seeing a part number for the display itself. Then again, did you see any specific part number on its backside or anywhere else?
Ok, this post is long enough! You must be tired reading all of this. I hope this sheds more light on your project. I’d like to hear that all your hard work paid off and that your speedo/odo comes back to life!
You are welcome. Don't abandon all hope if the new caps do not fix the problem. Which caps will you replace - only C6 and C13 or all the capacitors (C1, C2, C4, C11, C12, and C13) that were previously removed? Have you verified the proper position of TR7? Based upon the photo and its data sheet, it looks like the transistor is correctly installed.
I have some other recommendations for you to try:
1. Replace the jumper wires with smaller gauge wire.
2. Shape the jumper wire around components, never go over them. That includes SMDs. Tape may be used to hold the wire in place but avoid using it at the ends.
3. Examine solder joints for solder bridges (when solder from one joint makes contact (or touches) another solder joint.
4. When a component is de-soldered then removed, take a close-in photo of the eyelets (front and back sides). The zoom function works wonders here.
5. Clean the area of the component with isopropyl alcohol after it is removed from the board and after the new one is installed.
Cheers,
George
P.S. I may have not expressed it before (the jumper link discusses it) however, jumper wires are considered components and should therefore be soldered on the component side of the circuit board.
Last edited by Gen2n3; 09-24-17 at 09:20 PM.
Reason: Added Post Script
Wow! That was a good write up on DaleClark's thread. I'm glad to see that you got a replacement. What did you do with the old USDM Speedo board and JDM odo chip & (speedo) faceplate?
Thanks, hope maybe it may help clear up some stuff for other people when doing the same modification or dumb it down to make it easier.
I still have all the parts from both clusters, just boxed it up and stored them. Kind of sucks to have to spend $200 plus and buy the whole cluster when I just need the Speed/odometer assembly but worth it to me.
My next area to tackle on is the Cruise Control. I havent looked into it yet, but I saw that DaleClark had a thread about that, so I will definitely look into that.
09-16-17, 09:55 PM
