All's well that ends acewell

[Airbear]

I've never lost the signal once it's started. Haven't been through any tunnels yet.

I do have the red/black wire connected to an always hot 12V and startup is quick sometimes but not always. Perhaps Australia has less satellites than North America, I dunno. All my riding is touring and almost always in the mountains, where overhanging trees seem to make a difference. When I stop to refuel it's typically under a roof canopy and I know it'll take a little while to start - quicker if I roll the bike out from under cover. I am used to it now and reset the trip meter before stopping the bike at a fuel pump. I switch the ignition on while putting on helmet and gloves and it'll usually be ready to go when I am ready to hit the start button.

The really nice thing about a GPS speedo is the reliable accuracy and smoothness of the needle.

[gspd]

First time since 1990 that my first ride of the year was NOT an absolute blast.
Just tried my new Speedhut electronic speedometer today for the first time.
This is not a GPS unit. It runs off a 3 wire pulse sensor that fits in the speedo cable hole in the transmission.
I won't get into the calibration procedures here but they are very simple.

VERY DISAPPOINTED! to say the least.

The speedo needle lags waaay waay way behind when accelerating from any speed, or from a dead stop. It takes as much as 30 seconds to catch up and indicate approximately the speed you're actually going.
When decelerating, the needle takes about as long to react, and never drop below 50-60 km/h until coming to a full stop. Once fully stopped for about 2 seconds, the needle abruptly drops to 0.

I've eliminated any possible wiring issue(s), by hot-wiring it on the bench and running it with a drill.
It reacts the the same as on the bike.

The sensor (pulse generator) is from Wunderlich and is supposed to work 'most' electronic speedos.
The Speedhut speedo is supposed to work with 'most' pulse generators.
Trying to fool the electronics with extreme slow or fast calibration figures made no difference.

I'm going to call SPEEDHUT Monday and see if they have any suggestions or a refund policy..

In the meantime I'll reinstall my "No Longer Available" BMW Motometer speedo that works perfectly, in spite of its built-in, but comfortably consistent, inaccuracy at certain speeds. If I could, I would buy a new one just to have a spare, but NLA is NLA.

[Rob Frankham]

I would suggest that there's clearly something wrong with that unit. They need to replace or refund. If happens to all of them, it's most certainly not fit for purpose.

Rob

[Rob]

I had a Honda Civic with a factory digital speedometer. I gave the car to a friend in exchange for the 2002 Civic she had. I attempted to replicate the digital speedometer with a dedicated smartphone (cheepie) with a digital speedometer app on it. Not nearly as satisfying as the factory unit.

I had my Garmin GPSr mounted to the R65 a long time ago. I'm pretty sure there was some lag, there, too.
30 seconds is crazy, though. I'll have to install an app and compare it to my current Civic with the sunroof open.

[gspd]

Talked to Speedhut.
They think that the pulse sensor I am using is not compatible with their unit, not enough pulses per mile (PPM's)
I don't want to install a rim magnet/fork leg pickup (and its associated wiring) which would be a possible option.
I am returning the speedometer to them for a refund.

[Rob Frankham]

Maybe I'm getting this wrong but that doesn't seem to make much sense. Surely, the speedo drive turns faster than the wheel... in fact a quick mental arithmetic calculation (read questionable) suggests it rotates around eight times faster than the wheel (dependant on final drive ratio) so it seems fairly obvious that a speedo drive related sensor will produce a signal with around eight times the pulses per mile than a signal derived from a fork/sensor. Of course that assumes that each system only produces one pulse per rotation but the alternative would appear to be sticking substantial numbers of magnets to your wheel.

Rob

[gspd]

Maybe I'm getting this wrong but that doesn't seem to make much sense. Surely, the speedo drive turns faster than the wheel... in fact a quick mental arithmetic calculation (read questionable) suggests it rotates around eight times faster than the wheel (dependant on final drive ratio) so it seems fairly obvious that a speedo drive related sensor will produce a signal with around eight times the pulses per mile than a signal derived from a fork/sensor. Of course that assumes that each system only produces one pulse per rotation but the alternative would appear to be sticking substantial numbers of magnets to your wheel.

It doesn't really make sense to me either but who am I to argue with Speedhut.
Speedhut instructions say:
Compatible with all factory speed senders or transmissions.
Programmable from 1000 - 200,000 pulses per mile
Low pulse count per mile will result in a slow response from the programmable speedometer. Our programmable speedometers can read from 1K to 300K pulses per mile. Most transmissions will output at least 10K or more pulses per mile. However, if your transmission is outputting around 1-5K pulses per mile you might notice a slight delay in speedometer responsiveness.
Slight delay!!! my ass. The speedo is unusable as is.
My OE speedo has a w715 ratio indicated on the face plate.
I think this means 715 cable rotations (or sender unit pulses) per kilometer.
That converts to over 1000 pulses per mile, so I'm in that mystical 1-5K range.
I don't understand how the speedo can know how many pulses per mile are being generated when bench running with a drill without knowing the hypothetical wheel size being used.
The unit is programable and I tried fooling it into seeing my wheel as half its actual size and then as double its actual size.
That did not affect the lag or correct the needle stuck at 50-60 kmh when barely moving and only dropping to zero after completely stopping (the drill) for a few seconds.

In comparison, when using my OE speedo and cable on a drill, the needle action is perfectly smooth and linear and instantly mimics my trigger finger movements.

I'm just happy that Speedhut is giving me a refund.