Dash2 speeds

[osborni]

I have a Dash 2 and DL1.

Dash2 firmware is 9 1 9

A Cheery speed sensor wired into the Dash2 for road speeds.

Symptom:
above ~99 mph, the speed readout goes flaky then to 0. It returns to normal when the speeds drop below 99 again.

I worked perfectly a month ago.

I did change the brake cooling ducts to which the speed sensor is mounted on. I didn't measure the gap from the sensor to the bolt head, but it was visually the same.

The oddball thing is that the speed sensor data is perfect when logged on the DL1 and viewed in analysis.

Any suggestions as to why it's flaking out on the Dash2? Are they super sensitive to air gap?

It's one pulse per rev with about a 2 meter circumference wheel (245/40/17) at it seams to happen around 20-23 hz or so, which is well inside the capability of the dash and the speed sensor.

From the other thread, I think I'm having GPS issues too, so couldn't switch to a GPS feed on the Dash2....

[Support]

Hi,

It might be the gap, so it's worth checking this - alternatively it's likely that although the frequency is well withing the spec of the DASH2, the pulse size is on the limit for what it will accept. Very short pulses are filtered out to stop it getting false triggers. To address this either:

- use a long target for the pickup
- use the same size target on a small diameter rotating "thing".

Thanks,

Support

[osborni]

OK, I'll see what I can do.

Is there a hardware spec for the minimum pulse length in milli/micro seconds?

[osborni]

I did some math and at ~130 mph, given the size of the target (~3/8" bolt head) and the radius that it's mounted at (2.4"), the target exposure time to the hall effect sensor is 0.00086 seconds, which at least for the sensor's spec sheet of 10x10-6 ( 0.000010 ) seconds for rise and fall, should be enough.

I'll check the airgap this weekend. Given that it worked before I changed the part out that the sensor mounts to I suspect that I just missed by 0.5mm or so on the mounting of it. Though I guess fabricating a longer target wouldn't hurt too much.

[pault]

There are three potential issues. First as Support mentioned above, the minimum pulse width. You have calculated 0.86 mS. According to the knowledge base, you only need 0.25mS. So this should be okay.

The other issue is how these sensors work. As you move the sensor closer/farther from the bolt head. It will change the output. This is why it is important to get the sensor at the right distance from the bolt head. The farther away the sensor is from the bolt, the shorter the output pulse will become, until at some point it doesn't even detect the bolt head.

Additionally the sensor output drive type makes a difference. A lot of sensors will only drive its output one way and uses a resistor for the other direction. To the layman, this means the value of the external pull-up resistor will effect the output. A smaller value will narrow your pulse, a larger value will stretch it. This effect is likely quite small though.

So I'm guessing that you just need to move your sensor a little closer to the bolt head to make this work.

Paul