Technical question about speedometer adapters

yup, the speedometer shop guy suggested downloading a gps speedometer to my phone, said they have become quite accurate.
downloaded one that had good reviews and it damn near took over my phone, uninstalled it and haven't tried another.
is there one that isn't so invasive?

 

I use digihud, works pretty well and doesn't seem to have any crap on it. Might take a little configuring to get rid of the plethora of information you dont really need to see.

 

I spy Clark Brothers gear for speedo conversions. It is good. The info here may help Paul choose an appropriate ratio, and casing:

http://www.clarkbrothers.net/ratio_adapters.html
More detail here, starting at page 54:
http://www.clarkbrothers.net/pdf/MC794_CBI_Master3.pdf


Cheers,
Harv

 

Have you tested the speedo at a higher speed to see if the inaccuracy is linear [use a phone ???? GPS app]

A lot of speedos I have checked [that have minimal inaccuracy] can be corrected by moving the needle position in the speedo.
You remove the speedo from the car and hold the "drum" with the needle pointed at 30mph, Then twist the needle around to the 34mph position.

On my old 4x4 tow hack with oversize tyres ,I did this at the 80kmh position because that was the speed limit with a trailer [and the cops jumped on people speeding when towing a trailer]

 

@Paul , does the link in post #15 work?
The price is reasonable.

 

Be careful of GPS sourced speed information.

GPS works great if you are strictly traveling horizontally. For the most part planes can use GPS fairly accurately. However GPS cannot measure in the vertical. It can only 'see' you in a plan view(top down). Like looking at a map. It cannot tell if you are going up/down hill. This will affect your 'seen' speed vs your actual speed. Going up or down a steep grade the GPS may show you are doing ~30MPH but if the GPS thinks you are going '30' then you are most likely going faster than that.

Hypotenuse of the triangle. GPS can see your b base distance(speed), but not your a height(elevation) change as you travel the c hypotenuse.

Unless the new GPS systems have been programmed with good maps and known elevation changes in your area, which then it can calculate your actual speed, it may be no more accurate than the speed sign. If it uses Google Maps as an elevation reference, then it will probably be OK. But I know Google Maps is wrong in a few areas around me, so err with caution.

 

Mad Mike above suggested testing the speedo accuracy against something more accurate than the 'cop in a box'.

His method works.
Another method that also works is finding a longer stretch of road with mile markers.
Note odometer reading at your start point, corresponding to a milage marker. Drive to the ten mile marker and note the odometer reading (such as 10.8), then calculate the correction needed.

Both methods work.

 

What could be done is figure out what the speedometer gears were originally set up for in the vehicle the transmission came from.
Then do some math on the rearend and tire heights differences being used vs original application.
Should be close enough for government work.

 

lots of good information and advice here,
thanks all.
I put it all back together
downloaded digiHUD, thank you Jim
on a straight level stretch of road
holding 40 on the speedometer the ap said 46
holding 50 on the speedometer the ap said 57
holding 60 on the speedometer the ap said 69
that equates to 1.15, 1.14, 1.15 respectively
also looking closer at chassis obstructions the straight in line adapter won't work but I may be able to use an offset adapter.. 666F maybe..

 

Another thing to check...I always calibrate the odometer, then make sure the speedo is adjusted so that it reads correctly once it's getting the proper 1000 revolutions per mile. Did you find a stretch of road with mile markers? or use the digiHUD trip odometer to see if the odometer is reading accurately? It's pretty common for the speedo calibration to be off, as well as the drive gearing not giving it the proper 1000 revs/mile. Two things to fix.

 

I did not,
I only held speed for about 1/4 mile once for each of the three speeds tested.
no attention elsewhere.
'will have to look at all that tomorrow
if I can find time.

 

See the calibration page on my website for information, it’s too much to repost here.
www.speedoservice.com

 

GPS is 3 dimensional location data. It does compensate for altitude. But altitude effect is miniscule even at gradients. They also dont need any map data, they measure between their own data points.

GPS speed is extreemly accurate. The only issue is a small latency when doing tight turns but you would probably not even notice that as you wont be looking at the speedo at that point.

I have used a gps speedo that is an app on an old smart phone. I stick it on the dash with a magnet and remove it when its not needed. The smart phone does not even need to be on a network or have a sim card, just needs gps enabled and the app downloaded. If you also download google maps and set it to work offline and use downloaded maps you have a satnav as well.

 

Speedometer apps I've tried in the past have been grossly inaccurate. Moreover they've underread by a good 10-20%, which is on the worse side of useless. I uninstalled each after a day or so.

I've now downloaded DigiHUD: thanks for a credible recommendation! I'll test it when next I drive.

 

I have used DigiHUD for a few years now, and tested it against a few newer cars and found it to be quite accurate. I used it all the time on my 49 Buick until I got the speedometer rebuilt. It matched the newly rebuilt speedometer perfectly.

 

That's been my experience also.

 

I use an app "Speed Box". Seems pretty accurate to me.

 

My dash cam shows speed which is all most spot on with the speedo in my '39 p/u (thanks @e1956v) Only trouble is that there is a delay in the speeds it registers.

 

There is always lag time with GPS speedometer apps. especially below 30 mph, so in town it can be a pain but overall they are pretty good.

 

Would it be easier to just put different size tires on it and sell your old ones ?

 

cheapest way would be to just slow down, you kids and your hot rods

 

or get a Dymo labeler, and put new numbers on the face of the speedometer....

 

In the past i have put paint spots on the bezel. I found you only need to know 3 or 4 speeds in reality.

 

I’m installing a new set of retro look analog gauges in my 33. The speedo is GPS so solves all my problems.

 

or once you see what your true speed is compared to what your speedometer shows its easy to estimate.

if you can't remember your speed in ten mph increments do the math
in my case if I round the ratio to 1.15
at 30 on the speedometer it's 30+3+1.5

still though I want my gauges as accurate as is reasonably possible

 

I can make you up a 15% increase ratio adapter, so you don't have to play MATHLETE every time you drive.

 

Even if that were possible, the odometer would still be incorrect. And an odometer is a fairly useful gadget.

 

On its way today 8/14/23

 

5 years ago with the help of Speedometer Service I was able to correct my speedometer after a rear end gear change where there was no nylon gear available. It too was a very tight fit to the body trans hump. A few years later I found a stock Thunderbird 90* adapter and then added the correcting device I was using. I’m glad I had them make the new cable a bit longer.
There is an instrument company in Texas that lists many different parts for speedometer correction and ends. I preferred to use a HAMB member.

 

progress report:
I did as e1956v and others suggested and counted the 10ths on the odometer over a measured mile
the odometer rolled a little under .9 miles
I estimated approximately .87
1 ÷ 0.87 = 1.15
the same difference as what the phone ap showed at 40 and 60 mph
the needle moves smooth and steady above 10 mph so there shouldn't be any issues there,
I just need to spin the speedometer 15% faster to correct the reading.
I will give an update when the ratio adapter is installed.