[ChrisArnold]

I just installed the moates 8-way switch, and after reading up one of cgrey's old posts, I datalogged TUNEPOS. I tried the QH in both mode 1 & 5, but in both cases the switch wasn't working properly. When the switch is in physical position 1, which I assume is fully counterclockwise, it reads the tune loaded into tune4. When the physical switch is in positions 2 & 3, the QH reads tune3. Physical position 4 is back to tune 4, and 5-8 switch back and forth but always read only tunes 3 & 4.

To ensure I wasn't mistaken or getting confused about anything, I cleared all other tunes than 3 & 4 from the QH. It really is only switching back and forth between tunes 3 & 4. I further verified the only difference in the tunes, which is an extra 6* of timing everywhere, by eatching the dashboard and seeing the commanded timing jump.

So, has anyone experienced this? Is it just a bad switch or is something else going on?

[Jman]

I don't have experience with this, but I think you could test out the switch with an ohm meter. I would unplug it from the Quarterhorse just to be safe. You can ohm out the end of the connector so as to be testing the rotary switch and the cable at the same time. If that test doesn't give good results, try testing just the rotary switch with no cable connected.

The link to the Moates website where I found this is:

https://support.moates.net/quarterhorse/

Thank you
J.W.S.

[cgrey8]

I know there is an issue in the QH that is very intermittent where the tune you select (hardware or software mode) doesn't serve the "correct" tune. However MOST of the time for me, it was Tune 8 that got erroneously served...not always but far more common when it happens.

I suspected this was happening and for GUFx, I added both a downloadable parameter that includes the Tune Position Number (named Z_TUNEPOS). Then I added a dataloggable value (TUNEPOS) to the payloads that essentially relays to the QH what value it is reading from that Tune Position Parameter memory offset.

Before there was "BE support" for this, I wrote each tune position with the Z_TUNEPOS parameter set equal to the tune position. This way as I datalogged, I could flip between the tunes and SEE which tune was actually being served to the EEC.

Clint eventually added automatic support to populate the tune position value to the PID Z_TUNEPOS so anybody else that wants to add that to their strategy/definition file, BE will automatically update that PID with the tune position value similar to how it updates the checksum value just before it writes the tune.

I never could figure out exactly what it is that causes the QH to go sideways and misbehave. But it seems the only way to get it OUT of this state is to write to each tune position, back to back, until the QH sorts itself. Clint added a hot-key into BE for me to use to trigger a "write to all positions" for me to test this. Sometimes 1 attempt was all it took, but OFTEN it took 2-4 attempts before the problem got corrected.

This hotkey is not a feature that's very useful under normal conditions since it writes the currently loaded tune to all positions with the only difference being the Z_TUNEPOS parameter value. So I don't know that it ever got documented or added to a UI. I'm not even sure it is still there. If it is, maybe Clint can chime in as to what, exactly the hotkey is. If I had to guess, it's something like CTRL-ALT-A, but I could be wrong on that. It's been MANY years since I've messed with that.

[ChrisArnold]

Yeah, cgrey, that's the thread I read... Wanna say it was like 2012. Anyways, they're only $30 for a switch and wire. So, I ordered one. Got it in, today. Installed the new switch and wire. The functionality is the same. So, it's either the QH hardware or the software somehow. The tunepos from positions 1 - 8 goes 4,3,3,4,4,3,3,4.

I guess I wasted $$$ on the new switch, but at least I like the newer black knob, in my black interior, more than the billet aluminum knob.

I'll check out how the solder looks around the cable port, and if I don't see anything there, then I'll just live with have 2 tunes, because it's all I really wanted: one for 87, and the other for 93 octane.

[cgrey8]

The problem I was experiencing had nothing to do with hardware failures. It was all in the firmware (or more correctly stated, rigidware) in the QH's CPLD chip.