Persistent Speedo issue - sorted!
Ever since I built my car I’ve had intermittent problems with my Speedo. The problem has been going on for nearly two years and I’ve had a few gos at fixing it but I fixed it yesterday.
As some may know I chose Autometer Arctic White dials which I bought direct from a dealer in the US. The dials are well made and seem to have a good reputation in the US.
Since my SVA the Speedo developed an intermittent issue where the needle would swing wildly above the real speed.
My first thought was that the long twisted pair of leads was picking up noise from the car as the sender is non-active and only a switch. I tried grounding one lead near the dashboard to a known good earth. That didn’t work.
I next soldered the leads to the BMW sender unit to avoid connection issues as the connector looked a bit old and corroded. This didn’t fix it.
My next thought was that there was insufficient current through the sender reed switch to create the signal for the Speedo. So I fixed a 1k5ohm resister from the other lead to the +12v supply. This should give me about 100mA ‘switched’ current. I convinced myself this would cure it, but no joy.
I then presumed the BMW sender was faulty as these are known to fail. A brand new unit direct from local BMW spares department also didn’t cure the problem.
On my recent trip up to Stoneleigh my Speedo was going crazy and I had to resort to the Satnav or rev counter (in top gear) to check my speed.
After a serious bit of pondering I decided it was one of three problems. The worst option was the need to buy a new speedometer which cost about £120. The other two were an intermittent connection or that old idea of ‘noise’ pick up. I checked all the connections and remade one joint that looked a bit dodgy - no improvement.
I thought back to my days when building electronic amplifiers and radios and remembered that all circuits had 100nF caps liberally applied across supply lines. I looked around my garage and saw an old telephone lurking on a top shelf. Deciding that this looked a likely source of a suitable capacitor, I unscrewed the back and chose one of the largest Green Mylar Caps I could see. I couldn’t decipher its value but it had a 47 in it so it is probably 470nF or 47nF. With the soldering iron warmed up I quickly had it out of the phone and a couple of short leads soldered on. I stuffed the other ends of the leads into the connector behind the dashboard and bingo! the needle was calm and slowly moved up and down with the wheel speed. I then permanently fixed the capacitor and road testing the car this morning confirmed it was the problem.
So my original hunch was correct, I just didn’t find the right solution first time.
(I’ve now noticed that the radiator fan causes interference on the radio when it cuts in! Aargh...)
|
Linear Mode
