EFI Live is neat in the ability it gives us to control tire and and gear changes. However..................

The most accurate way to check your speedometer is by driving at a constant speed though one of those sections of road where they have put up speedometer check signs and time yourself. I do not have anything like that around where I live right now.

This leaves a choice between GPS and math. I am running 315/75R16s and 4.10 gears. That is a 3.2% difference from the stock 245s and 3.73 gears. This amounts to a 2.1 mph difference at 70 mph, based on the math.

Today I checked it with my 12 channel GPS. Now many of you may think the GPS is super accurate and 100% reliable in determining ground speed. I am not so sure. I have been using a GPS while hunting since 1998. I am have also been a USAF pilot since 1983, and have used GPS while flying since 1998. I don't remember exactly when, but within a year or two after 1998 the Government took out the time error that was purposely put into the GPS system. The USAF always had the ability to remove this time error, but after the error was removed everyone had access to the same level of precision.

I know from repeated personal experience--after removal of the time error--that even when the GPS is indicating an estimated position error (EPE) of zero, that if you walk away from a point and come back to it a few hours later the GPS will only reliably get you to about 50 feet. Also, when flying, there is ground equipment (WAAS) required in order for an approach to have a fidelity of three meters which again shows the lack of precision of raw GPS. Further, the GPS will only give you a reliable ground speed when you are not driving through a turn.

All this makes me wonder just how accurate a GPS is at highway speeds, especially when it is not receiving a WAAS signal.

Mt GPS showed my speed at 70 MPH to be off about 1 mph, where my math shows it to be off by 2.1 mph. Yes, that is very close and the only way to resolve it is to find a speedo check area on the road somewhere. However, since I can't do that right now.............

Has anyone checked the MPH reading of their GPS against the actual time over the ground in a speedometer check area?

Not to get too much off here, but did you actually physically measure your tires while mounted with full weight on them (the 245s and the 315s) or are you just going by the advertised sizes of 245 and 315 coupled with the gears present for each?

Reason I ask is because they will be shorter than advertised (which I'm sure you are aware) and likely by varying amounts between the two. For example the 245 may be shorter than advertised by 1", whereas the 315 may be shorter by 2". Therefore, the difference won't just cancel out when comparing actual (measured) versus theoretical (advertised). This variation in sizes could potentially throw off a number early on in the calculations and then end up leaving your final answer (your calculated mph) off by a fair amount.

This may not be the case at all, just trying to offer up something to think about if you hadn't already.

Plus you are right, the GPS speed function is not going to be entirely accurate all the time but a blind squirrel finds a nut every once in a while. ;) :D The GPS should be able to measure your change in position fairly accurately though. So, if you can time how long it takes you to go some determined distance with the GPS, then you could get a pretty accurate MPH.


Good luck

If EFI Live was set up for it I would use loaded roll-out. In fact, I could still do that by dividing loaded rollout by pi. However, since I do not have the actual loaded roll-out figure for the stock 245 there is no point in using that figure for my new tires.

The stock setting in EFI LIve is based on a tire diameter of 29.53", which is an estimate to what loaded rollout is. You are right, that is about an inch compression from the nominal height of 30.47". The height of my rear 315s are 33.5"--from actual measurement where their nominal measurement is 34.6". My best guess is to use 33.5" and a 4.09 gear ratio (stock showed a 3.72).

To do better than that guess I need to know what my true MPH is. I then compare that to the mph on my speedo, calculate the percentage of error, then adjust the ratios in EFI Live to match.

That is why I am hoping someone has compared their GPS to a measured speedometer test area.

Even though the message below was sent to one of our LS1 customers, you can see just how crazy the whole system is (Paul wrote this).......

-- LS1 speedometer 101 --

Disclaimer: This info has been figured out the hard way, through trial and error and lots of PCM hacking. Some of it may not be 100% accurate as per GM's design specifications.

Caveat: These discussions assume a Corvette which has the VSS pickup fitted to the rear axle rather than the tailshaft as per the F-body and trucks. To make the EFILive speedo calculator function correctly for a C5, you MUST have the "Transaxle" checkbox checked. EFILive analyzes the VIN to determine if the checkbox should be checked or not - but you are free to override EFILive's guess if EFILive is wrong.

There are 3 main calibrations that effect the PCMs ability to determine the vehicle's speed: {H0101}, {H0102} and {H0104}.
There are another 2 calibrations that effect the PCM's ability to share the vehicle's speed with other modules: {H0105} and {H0106}.
The "other" calibration, {H0103} is not used by the PCM to determine vehicle speed. It appears to only be used when calculating the axle torque for various torque management duties.

=======================
{H0101} is defined as the number of pulses transmitted by the VSS per mile of travel. There should never be any other value in this calibration than the correct value. The correct value is calculated as follows:

H0101 = teeth * 63360/(pi*tire)
where:
teeth = number of teeth on the VSS pickup.
pi= 3.14159265
tire = tire diameter in inches.
(pi*tire) is the tire circumference)
63360 is the number of inches in a mile.

Sample calculation when tire diameter = 26 inches:
H0101 = 40 * 63360/(3.14159265*26) = 31027 pulses/mile

=======================
{H0102} is defined as the number of VSS pulses per transmission output shaft revolution. It contains the differential ratio encoded in its value. Using the pulses per mile and this calibration the PCM can calculate the differential ratio. It is also what EFILive uses to calculate the C5's "estimated diff ratio". It is calculated as follows:

H0102 = teeth/(true diff ratio)

Sample calculation when true diff ratio is 3.90
H0102 = 40/3.90 = 10.25641 pulses per TOS revolution

=======================
{H0103} is defined as the final drive ratio and is used by the PCM to calculate the axle torque multiplication effect of the final drive ratio. It plays no part in the PCM's speedo calculations. Note: EFILive uses this value when calculating {H0101} and {H0102} so make sure it is accurate in the speedo calculator. However, I recommend changing it back to 2.73 in the actual H0103 calibration - regardless of the actual diff ratio fitted to the vehicle. That way the PCM calculates a lesser torque value for the rear axles and potentially solves the WOT shift/redline issue.

=======================
{H0104} is defined as the number of transmission output shaft (TOS) revolutions per mile. This calibration is used to determine auto trans shift speeds. It is sometimes set wrong in a stick shift (i.e. the GM calibrators leave it at the default setting from the Delphi PCM factory). That does not matter because in a stick shift this value is ignored. It is calculated from {H0101} and {H0102} as follows:

H0104 = H0101/H0102

Sample calculation using values from previous samples above:
H0104 = 31027/10.25641 = 3025.1355 TOS revs/mile.

=======================
{H0105} defines the number of pulses output on J2, pin 50 per mile. Whatever module is hooked up to J2 pin 50 will receive this many pulses per mile.

=======================
{H0106} defines the number of pulses output on J2, pin 49 per mile. Whatever module is hooked up to J2 pin 49 will receive this many pulses per mile.

=======================
The following is purely speculation....

We have found that sometimes dealers fit different wheels/tires to a vehicle and then mess with the dashboard calibrations in order to make the speedo read correctly. I *think* what that means is the dashboard no longer expects 4000 pulses per mile but some other figure. You may need to mess with H0105 or H0106 to get the speedo needle reading correctly. I think the H0106 is the speed that is sent to the ABS system. Again you may need to play with these values to prevent the ABS from seeing a too large a difference in speed between the front and rear wheels.

I hope that helps clear up some of the speedo mystery. I don't think GM could have made it more convoluted if they tried. The whole speedo things smacks of being designed by a committee, with each person on the committee only interested in one particular speedo function. Then some poor programmer schmuck had to write the code to support such an awful design.

I thought that these sighns where good too imo
others dont
but when i had my gps it was bang on with the sighn
how do they measure the distance when putting the sighn in the ground i cant see them using a tape measure....lol

typically a roll wheel with a counter attached to a vehicle is used to place mile-marker signs if thats what you are referring to. I don't think they shoot them with survey equipment although they probably could since they are making so many other shots, might as well include those. I can't say thats how they do it though, at least not that I've seen.

C-ya

I thought that these sighns where good too imo
others dont
but when i had my gps it was bang on with the sighn
how do they measure the distance when putting the sighn in the ground i cant see them using a tape measure....lol

Seems the best option is to use your GPS and enable the WAAS, that is if you have a WAAS capable GPS. Next time I head to the next town, I will check my speed with the Speedometer Check Area on the hiway, and also with the GPS.

WAAS only works in close proximity to an airport that has the transmitting equipment. Maybe I could drive down the runway........... :D

Well, it just happens that my GPS is WAAS capable and I-15 or I-86 here both run next to the airport run ways. I will have to see what the difference is using the GPS by the airport and then out in the middle of no where. Guess it is time to take a road trip and see what happens and if I can recieve the WAAS.

Find a friend that is a police officer and have him check your speed while maintaining a constant speed. Then compare your speedo to the speed detection device!

Dan

Talking to some fellow aviators today, they reminded me that GPS without WAAS is only guaranteed to a 10 meter fidelity. This means when the EPE is zero, it will be accurate to about 30 feet.

Talking to some fellow aviators today, they reminded me that GPS without WAAS is only guaranteed to a 10 meter fidelity. This means when the EPE is zero, it will be accurate to about 30 feet.

I have said that on several occasion here. But I am like Rodney Dangerfield, and I get no respect:exactly:
In another thread a few days ago, someone was talking about GPS for checking mileage, I stated it was not as accurate as doing a tire roll out measurement. Nether is mile marker post along the highway.

The problem with using tire rollout is in order to get it right, we need to have the tire rollout of the stock tire as well as knowing if the speedo was accurate with the stock tire. If we know the stock rollout and know the speedo was accuirate, then we can meaure the rollout of the new tire and adjust the reading by the percentage difference in size. If we don't know either of those two things we are back to estimating..........

Measure roll-out, I have a Garmon GPS with wass and I have used it to calibrate three trucks and it is dead on accurate using pi.

T;) NY

Tony,

So then you are saying the GPS, when you drive by/on a WAAS equipped airport, agrees exactly with your calculations when you use loaded roll-out? BTW, what loaded rollout do you use for a stock 245?

But I am like Rodney Dangerfield, and I get no respect:exactly:

I tell you, with my doctor, I don't get no respect. I told him, "I've swallowed a bottle of sleeping pills." He told me to have a few drinks and get some rest.

Love Rodney's work :)

Cheers,
Ross

Tony,

So then you are saying the GPS, when you drive by/on a WAAS equipped airport, agrees exactly with your calculations when you use loaded roll-out? BTW, what loaded rollout do you use for a stock 245?

I have only calibrated for 265's, 285's and 33-12.50's.

My father has Bridgestone 245's on stock rims, I will see if I can get a roll-out from his truck.

;)

I tell you, with my doctor, I don't get no respect. I told him, "I've swallowed a bottle of sleeping pills." He told me to have a few drinks and get some rest.

Love Rodney's work :)

Cheers,
Ross
:exactly:
He is great.

Get a stop watch, and run several 1 mile trips until you get the time and speed right:eek:
According to what I read some time back, the vehicle manufactures only have to be within something like 4% of being dead on.

By the way, does EFILive get the mph from the same source as the speedo? I ask because it reads exactly the same as the speedo.

Went by the airport today on the interstate and I did have WAAS enabled. My speedo that I adjusted the needle on when I put the new overlay in was within 1 MPH at 75mph. I set the speedo needle to be acurate at 65 with the GPS a few months back. GPS showed accuracy around 25' when WAAS was working. With out WAAS accuracy was around 35-45 feet when moving. My GPS is much more accurate when stationary. On the way back past the airport, I set the cruise at 65mph this time and it showed dead on with the GPS. I also timed the speedo on the interstate at one of the mile marker test strips they have. My time @ 75 was within 1/2 sec (faster) of what the sign says it should be. Seems close enough for me not to worry.

Thanks so much for taking the time to check that out!

If EFI Live was set up for it I would use loaded roll-out. In fact, I could still do that by dividing loaded rollout by pi. However, since I do not have the actual loaded roll-out figure for the stock 245 there is no point in using that figure for my new tires.

The stock setting in EFI LIve is based on a tire diameter of 29.53", which is an estimate to what loaded rollout is. You are right, that is about an inch compression from the nominal height of 30.47". The height of my rear 315s are 33.5"--from actual measurement where their nominal measurement is 34.6". My best guess is to use 33.5" and a 4.09 gear ratio (stock showed a 3.72).

To do better than that guess I need to know what my true MPH is. I then compare that to the mph on my speedo, calculate the percentage of error, then adjust the ratios in EFI Live to match.

That is why I am hoping someone has compared their GPS to a measured speedometer test area.

I would guess "loaded rollout" is going to vary with vehicle speed due to centrifugal force of the tires. Your tires are not going to run with the same footprint and diameter at 10mph as they would at 70mph. Correct?

Finally, a calibrated radar gun would probably be the best speed check, in conjunction with a level and straight highway.

I am not sure the centrifugal force on an LT tire is significant at 70-80 mph. However, I can't think of any convenient way to measure it at 80 mph to verify for sure....... :D

I am not sure the centrifugal force on an LT tire is significant at 70-80 mph. However, I can't think of any convenient way to measure it at 80 mph to verify for sure....... :D

It might not be significant, but I'm sure it's a factor. Just how much I wouldn't know. Running your rear tires rock hard would minimize the low speed - high speed differences, but I'm not a fan of doing that, especially in these trucks. Just food for thought.....

Anyone own a radar gun? :lol:

Hey Blaine, who says that you have to be near an airport for WAAS to work, I get the correction all of the time on my Garmin, I see the correction on the satellite display. According the Garmin, there are only 25 WAAS stations in the country so they won't be focused near an airport specifically or there are a lot of airports without it.

This is where I got my info. If this isn't correct or is just marketing I would like to know.
http://www.garmin.com/aboutGPS/waas.html

HMMM. Their definition of WAAS seems more like what the US did in the late around Y2K to allow general users to have the position correction the US Military always had. I remember my 1998 model GPS going from 26 foot EPE and bring me to within 300 feet of an object to getting EPEs as good as "Zero" and getting within 50 feet of an object.

From my time as a USAF pilot--had my "fini-flight" Dec 2004, I remember being told WAAS was specialized equipement being installed at certain airports to allow three-meter accuracy for precision aircraft approaches.

Now you have me wondering if the three-meter accuracy is called something other than WAAS.....................

Clinton turned off Selective Availability in early 2000. WAAS was debuted later in that year. Perhaps you are thinking of DGPS?

Here is some good stuff.
http://www.gpsinformation.org/dale/dgps.htm
http://en.wikipedia.org/wiki/GPS

I can't believe you made me look stuff back up in my AF Pubs. Here is the section on WAAS in AFI 11-217 Vol.

6.12.4.1. Wide Area Augmentation System (WAAS). WAAS augments the basic GPS signal for IFR use from takeoff through Category I precision approach. This system improves the accuracy, availability, and integrity provided by GPS, thereby improving capacity and safety.

6.12.4.1.1. System Description. Unlike traditional ground-based navigation aids, the WAAS covers a more extensive service area. Wide-area ground reference stations (WRS) are linked to form a U.S. WAAS network. These precisely surveyed ground reference stations receive signals from GPS satellites and any errors in the signals are then determined. Each station in the network relays the data to a wide-area master station (WMS) where correction information for specific geographical areas is computed. A correction message is prepared and uplinked to a geostationary satellite (GEO) via a ground uplink station (GUS). The current WAAS site installation consists of 25 WRSs, 2 WMSs, 4 GUSs, and the required terrestrial communications to support the WAAS network. The message is then broadcast on the same frequency as GPS (L1, 1575.42 MHz) to WAAS receivers within the broadcast coverage area of the WAAS. The WAAS broadcast message improves the GPS 95 percent signal accuracy from 100 meters to approximately 7 meters.

What is this, a pilot forum now? :) Okay, so I'll admit I'm one too. I've studied some special stuff on GPS and how we use it to guide bombs. I think as far as GPS velocity is concerned, you need to remember that the estimate of probable error is a prediction of how closely the GPS can figure out your actual physical position vs. where the gps plots you in its theoretical space. The satellite constellation is spherical, but the planet is not. To shorten this up, while your EPE may not give you a great position in relation to reality, the error is fairly constant over a brief period of time (like while you drive down the road for a mile or two). If your GPS computed position is 10m west of reality at mile marker 15, it'll still be 10m W at mile marker 16, so the speed indicated should still be right on the money.



If you really want to know how accurate your particular device is, check the manufacturer’s specs. Garmin lists speed accuracy on most, if not all models. There is a lot of trick math the GPS has to do in order to filter out crap signals, including using Doppler shift just like a radar gun does. For example, the Etrex/Legend line read:
Specifications:
- Position Error: <15 meters 95% typical.
- Acquisition Times: Warm ~15 sec, Cold ~45 seconds, Auto Locate ~2 minutes (longer in non clear-sky situations.)
- Speed Accuracy specification 0.1 knot RMS steady state
- Physical Size: 4.4"H x 2.0"W x 1.2"D (11.2 x 5.1 x 3.0 cm)

Blah blah blah. Speed accuracy of .1 knots RMS steady state. Speed IS NOT instantaneously indicated. (see tricky math part above). 1 knot = 1.15077945 mph,so .1 knot should give you an accurate STEADY STATE speed within .115 MPH.


Hopefully being accurate to within a tenth of a MPH is good enough for y'all :eek: