VSS help please

This forum is for general questions you have on assembling and troubleshooting your GPIO board. You can purchase a GPIO board at DIYAutoTune now!
Bernard Fife
Posts: 1696
Joined: Fri Apr 04, 2008 1:28 pm

Re: VSS help please

Post by Bernard Fife » Mon Jan 12, 2015 11:03 am

Great, thanks. I will have a look.
"Never wrestle with pigs. You both get dirty and the pig likes it." - George Bernard Shaw

Bernard Fife
Posts: 1696
Joined: Fri Apr 04, 2008 1:28 pm

Re: VSS help please

Post by Bernard Fife » Mon Jan 12, 2015 11:53 am

Addicted,

Yep, something is definitely wrong. I suspect an underflow/overflow issue, because the speedo is wrong for the input. For example, with an input of 501 Hz I would expect to see 60 mph with your setup. However, I see 60 mph with an input signal of 118 Hz (and VSS_per is reporting correctly at 8.48 msec). I will sort this out and post back here.

Lance.
"Never wrestle with pigs. You both get dirty and the pig likes it." - George Bernard Shaw

Bernard Fife
Posts: 1696
Joined: Fri Apr 04, 2008 1:28 pm

Re: VSS help please

Post by Bernard Fife » Mon Jan 12, 2015 12:01 pm

Addicted,

Sorry, I got tripped up on the VSS location again.

It should be:

tire diam = 23" = 877 rev/mile.

With 8 teeth, this is 877 * 8 = 7016 pulse/mile.

At 60 mph that is 7016 pulse/min or 7016/60 = 116 pulse/sec or 116 Hz.

So that part works as planned.

The output shaft rpm *should* be 877 rev/minute. {edit} Actually it shuld be 877 rpm multiplied by the final (and intermediate) gearing. In MShift terms, the output shaft speed is the speed of the shaft before the final drive ratio, NOT the speed of the half-shaft or wheel. So ignore the rest of this text, and go to the following posts {/edit}. But os_rpm is not displayed as anything close to that value (as you said), and I know why now. The output shaft rm is calculated from the speed, tire size, and critically from the axle ratio. If the location is on the wheel, then the axle ratio should come into it. So I will look into that next and post here when it is sorted out.

Thanks for spotting this and bringing it up!

Lance.
"Never wrestle with pigs. You both get dirty and the pig likes it." - George Bernard Shaw

Bernard Fife
Posts: 1696
Joined: Fri Apr 04, 2008 1:28 pm

Re: VSS help please

Post by Bernard Fife » Mon Jan 12, 2015 12:19 pm

Addicted,

Okay, this *should* work:
Monitor_5099a.abs.s19
(81.47 KiB) Downloaded 140 times
GPIO_MShift_5099.ini
(276.24 KiB) Downloaded 138 times
The real fix for the issue you reported is in the INI, but the S19 has a fix for the CANbus users (to generate the output shaft rpm properly).

Lance.
"Never wrestle with pigs. You both get dirty and the pig likes it." - George Bernard Shaw

Addicted
Posts: 188
Joined: Mon Oct 14, 2013 7:14 am
Location: Winchester, England

Re: VSS help please

Post by Addicted » Tue Jan 13, 2015 11:26 am

Sorry Lance, it didn't work.. :(

The relationship between speed and hz has remained the same at 7.2

8.6mph @ 1400rpm in First gear = 62hz

I need to go back through my notes though as we disagree on what the hz should be at a given speed.. ;)

edit: ignore that, I think we agree on the hz/mph ratio.. on a previous page you calculated 60mph = 116hz.. although I think I got 118hz but hey whats 2hz? :D

Also I referenced the wrong december datalog in my previous reply that indicated a ratio of ~1:1, I have another december datalog that showed the correct ratio closer to ~2:1 (I created a custom field that simply devides Hz by MPH in MLV)

8.6mph @ 1400rpm in First gear is correct, I just think it should be approx 17.2hz
Attachments
2015-01-13_18.21.32.msl
(788.43 KiB) Downloaded 131 times

Bernard Fife
Posts: 1696
Joined: Fri Apr 04, 2008 1:28 pm

Re: VSS help please

Post by Bernard Fife » Tue Jan 13, 2015 1:41 pm

Addicted,

If we agree that 118 Hz is 60 mph, then 8.6 mph ought to be 16.91 Hz (118*8.6/60).

When I feed an 17 Hz signal to my bench set-up, with your recent msq, the speedo shows 8.2 mph. At 40 Hz I get 20 mph as expected.

One thing you might try is setting the input masking value to zero to disable it. This can help avoid things like triggering only on every second or third pulse.

Lance.
"Never wrestle with pigs. You both get dirty and the pig likes it." - George Bernard Shaw

Addicted
Posts: 188
Joined: Mon Oct 14, 2013 7:14 am
Location: Winchester, England

Re: VSS help please

Post by Addicted » Tue Jan 13, 2015 2:26 pm

Lance, it can't be false triggering on the input because then it wouldn't be getting the speed correct in relation to the engine RPM would it, but it does..

First gear is 2.714:1, intermediate gear is 1.159:1 and final drive is 3.700:1 that gives an overall of 11.638:1

1400 rpm / 11.638 = 120.3 output shaft rotations per minute, with a 23" diameter tyre thats gives 8.23MPH

120.3 x 8 = 962.4 pulses per minute, 962.4 / 60 = 16Hz..

We obviously need to take into account that we'll round up numbers slightly differently in calculations, but we're clearly in the same ballpark..

If you feed 17Hz into your bench setup and the speedo shows ~8.2mph what is CAN1.VssHz showing in a datalog?

Bernard Fife
Posts: 1696
Joined: Fri Apr 04, 2008 1:28 pm

Re: VSS help please

Post by Bernard Fife » Tue Jan 13, 2015 2:47 pm

Addicted,

With an input signal of 17 Hz (approx, my signal generator isn't great at slow speeds - that's why there's no decimal point), and a speedo of 8.2 mph, the VSS Frequency gauge shows 16 Hz (and VSS period = 62.01 msec) on my bench with your MSQ. The datalog gives VSSHz as 16.0. That all looks correct to me.

Lance.
"Never wrestle with pigs. You both get dirty and the pig likes it." - George Bernard Shaw

Addicted
Posts: 188
Joined: Mon Oct 14, 2013 7:14 am
Location: Winchester, England

Re: VSS help please

Post by Addicted » Wed Jan 14, 2015 1:05 am

Well I'm at a loss then. The software appears to be reading the signal from VSS sensor correctly as it displays the correct speed/rpm throughout the datalog, its only that one CAN1.vsshz field than is incorrect.

What is CAN1.vsshz used for?

Bernard Fife
Posts: 1696
Joined: Fri Apr 04, 2008 1:28 pm

Re: VSS help please

Post by Bernard Fife » Wed Jan 14, 2015 5:36 am

Addicted,

VSSHz is not used for anything at all. It is just there for informational purposes. It doesn't even come from the controller, it is calculated by your PC based on a formula in the INI. The VSS period (VSS_per in the INI), on the other hand *is* sent from the controller, and VSS_per the actual measurement that is used in the code to calculate the current speed.

The last INI I sent wasn't quite right though. It adjusted the VSS Hz, but I think it also buggered up the os_rpm when the sensor was mounted on the wheel (like yours). I think the attached INI is better (it fixes this and also calculates the VSS Hz solely from the VSS period rather than back-calculating it from the speed, gearing, etc.).
GPIO_MShift_5099.ini
(276.46 KiB) Downloaded 143 times
This INI also adds Wheel RPM to the gauges and the equivalent wRPM to the datalog (to distinguish it from output shaft rpm).

With this INI, on my bench and with a 16 Hz input to VSS (VR1) and with your MSQ, I get on the TS gauges:
Speed = 8.2 mph,
VSS Freq. = 16.1 Hz (~ 2 rev/sec or ~120 rpm at the wheel),
VSS period = 62.165 msec,
OS RPM = 514 (= 120 wheel rpm * (3.700 * 1.159)) since the output shaft (in MShift terms) is the output from the transmission's shiftable gears to the final (and intermediate) fixed gearing - this is different than the wheel rpm, of course - and this may be where we are having semantic issues). The engine rpm would be 514 times whatever the current internal gear ratio is (plus any converter slippage). If you were in first and it had a ratio of 2.714:1, with no converter slippage the engine rpm would be 1395 rpm, and
Wheel RPM = 119.8

Lance.
"Never wrestle with pigs. You both get dirty and the pig likes it." - George Bernard Shaw

Post Reply