How to make the stock tachometer respond faster

The background

Ever since I acquired my miata in '95, I love the responsive nature on every aspect of the car. However, the more performance mods I put in to make it more responsive, the more I noticed that my stock NA tachometer was too laggy. Please note that this trick probably don't apply to NB tachs which I found to respond faster than NA tachs. And for the NA tachs, this mod will not fix its inaccuracy where it usually shows 7500rpm with actual 7250rpm. I investigated the tach and came up with this little trick to improve its response.

How it works

The tach uses a chip to convert the frequency of ignition pulses from the ignitor to corresponding electric current signals to the solenoids driving the tach pointer. The chip includes a capacitor as part of its external circuit to determine how fast the electric current signals to the solenoids should change. This damping idea is to avoid jumping pointer at idle or while engine is taking up driveline slacks. However, I found that the factory damping is too much for such a responsive car. By reducing the stock 16V 4.7u capacitance, the tach will respond faster to any change in rpms.

Finding the new part

I tried with a few difference capacitance and settled with 1u (one mu). Anything smaller leads to a jumpy tach at idle. I found that when I lift or gas suddenly at high rpm and the engine is taking up driveline slacks, I get some jumping on the tach with the 1u cap but I can certainly live with that. If you don't like that jumping, maybe you'll like 2.2u better. Try sourcing the cap from a electronics hobbist friend as the capacitances are very common. Failing that a 1u or 2.2u cap should be cheap and easy to find in your local electronics part store or Radio Shack. It's so cheap that often times you have to buy 2 or even 10 in a package. Capacitor higher than 1u usually comes as a electrolyte one. One-mu or lower is common as electrolyte or tantalum. The type doesn't matter. A non-polar cap should work fine too but for the same capacitance, a non-polar cap is usually bigger in size and more expensive (talking maybe a buck or two difference). The voltage rating has to be 16V or higher. A higher voltage rating cap is usually bigger in size but doesn't affect performance.

Getting to the tach PCB

Remove instrument clustor as you would when exchanging gauge faces. Unbolt the tach from the clustor main PCB from the back. Ground yourself first by touching momentarily on the metal chassis (if you're in the car) or some big metal pieces nearby. Some people are prone to carry more statics than others and fry electronics more often. The tach is a delicate piece of equipment so use extra cautions especially on the PCB. If you're good at navigating fingers thru' tight space and/or use a pair of needle-nose pliers/tweezers, you may be able to remove/replace the cap without unbolting the gauge face. I have 14+ years of electronics experience and have relatively small and steady hands plus I used a pair of tweezers so I was able to do it without removing the gauge face.

Locating the cap on PCB

It's been almost 5 years since my mod so I can't recall precisely, but according to J_Man on the forum thread, it's easy to locate the cap. I took the following pics a couple of years ago. Note that the light blue cap is the new one I put in so your original cap will look different. However, the location should be identical.

Removing the old cap

Since the tach circuit involves delicate components, please use extra cautions when removing and replacing the cap. If you're not familiar with soldering small electronics components, I'd highly recommend an electronics hobbist friend for the job. Use a small soldering iron designed for electronics. Do not use a high-power gun. A pair of fine needle-nose pliers or tweezers designed for electronics hobbist is recommended. They help as heat sink and at precisely locating the leads. They also help leverage the old cap out when desoldering. I don't remember if there was room to stick my tweezers below the cap and clamp on the leads on the component side. If not then clamp the cap body itself. In the meantime, desolder on the other side of the PCB and pull the cap with the tweezers. You might find it useful to lever the tweezers on some hard but not fragile object on the PCB. Again I can't remember exactly what I did.

Putting in the new cap

Polarity is important with a polar cap. Although it's highly unlikely that you'll blow the cap or the low current circuit with a reversed cap, the reversed polarity will render the capacitance useless and it's like putting in a very low capacitance cap. So if you find the tach jumping rigorously at idle, chances are you have the polarity reversed. For a electrolyte cap (cylindrical in shape), most often the negative side is identified with a high-contrast strip with a few "-"'s on it and possibly with arrow signs pointing to the lead as well. A tantalum cap (ellipsoidal in shape) often has a small "+" side near the corresponding lead. If the component is new, the positive lead is most often (if not always) longer than the negative lead.

Once you figured the polarity, clip the leads to about 1/4". Insert and solder the component in. Use tweezers on the leads on the component side if possible. Use as little heat as possible when soldering. Again, I'd highly recommend an electronics hobbist friend if you don't feel confident not damaging the tach.

Testing

Replace tach onto cluster and hook back cluster for testing. You may want to leave the pod off until after confirming everything works fine. Notice any pointer vibration at idle. Rev the engine to check the operation of the tach. If possible, roll the car in 1st gear at about 2000rpm and suddenly lift all the way and see how much jumping on the tach while the engine is taking up driveline slacks. If not happy with the capacitance, redo with a different one. Once happy with the capacitance, replace the rest and have a test drive. Don't take eyes off the road and onto the tach for too long as I'm sure you'll be tempted to do so after the mod. :D Drive safely.

...and more
While the forum thread of hacking the stock tach has emerged recently, it brought my attention to share the other trick I did long time ago (about 4 years).

For those who run aftermarket ECUs that can show real time (and most likely with higher resolution) water temperature from your ECU, this is a trick you might consider doing, especially if you don't have oil temp gauge already but don't want to add a "non-stock"-looking gauge in the car.

Before I did my engine rebuilt in '00, I had always wondered if the factory water temp sensor would work as an oil temp sensor. I figured that I don't need 2 water temp readers in my car so I decided to convert my factory water temp reader to oil temp reader.

So when I had the oil pan off while rebuilding the engine, I drilled and tapped a hole (1/4" NPT IIRC) at the bottom facing the transmission and put a brand new factory water temp sensor in there. I think I bought it at about $25US regular price from the dealer back then so you can most likely get it cheaper.

I extended the factory connector originally hooking up to the back of the head to the bottom of the oil pan using a long wire and a pair of matching male-female connectors. IOW, I can easily unplug the factory water temp connector from my extension and hook it back up to the factory water temp sensor.

Since my '90 (probably applies to all NA and even NB) runs 3 water temp sensors from the factory, one for the gauge, one for factory ECU, and one for the fan, I can theoretically cheap out and just move the factory water temp sensor to the oil pan and plug the hole in the head. However, I didn't know whether I'd need the factory water temp reading later on and I was also thinking maybe a toggle switch to choose between showing water/oil temp would be cool so I left the factory water temp sensor in there and bought a new one. If you have a spare sensor from a spare head, then it's a no-brainer decision.

Without linearizing the factory water temp gauge, you probably won't see much difference at normal operating condition when oil temp is usually close to water temp. However, the first thing you'll notice on cold start is that the new oil temp gauge starts to climb so much later than water temp would that you would think that it's not working. You'll notice how much longer it takes to warm up oil than water in your engine. Some day I'll "linearize" my factory water temp gauge and it'll become even more informative than the original "idiot" water temp gauge. ;)