Dodgy data - please help

[PaulDuggan]

I recently bought a used model D with a PM4. I'm told it was bough reconditioned from C2 direct. Seems well looked after and some parts (like the chain) are clearly almost new. Since I bought it I've been using it with a Polar sensor and ErgData with the C2 cable, and sometimes everything is OK; other times it's not... I see the following symptoms:
Wildly erratic stroke rates. While pulling mid 20s it can show <10 or >50 and varies rapidly.
500m split times showing 1:30ish when I know I'm pulling 2:20ish. When I pull very hard the split times look more reasonable (i.e. when I feel like I'm pulling 1:45-1:50 that's pretty much what it shows). Goes back to 1:30 when I *slow down* again. Stroke rate still nuts.
Every so often I see a screen on the PM4 telling me that the battery power is 'low', but at the same time 100%! Other times it initially shows 96% before counting up to 100%. The PM4 is running from brand new Duracell non-rechargeable cells.
The Info page tells me that the PM4 is configured for a Model D/E.
F/w version is 318, h/w version is 340. Datecode is 2972013.

My problem is that I don't know what to change/replace but I don't know how to trace the problem to its source. I spoke to a guy at C2 yesterday who told me to clean the flywheel. I did this and it was very clean anyway I also looked at the sensor PCB while I did this and was OK to the eye too, but I don't know how to check the magnets or what I'm looking for wrt them...?
C2 man said that if that wasn't the problem then I would just have to start replacing bits starting with the cheapest first. I'm posting in the hope that someone here can help me go about it a little more empirically, or that someone has experience relating to this kind of problem to help me troubleshoot. I would appreciate any and all attempts to help.
Thanks
Paul

[jamesg]

While pulling mid 20s it can show <10 or >50 and varies rapidly.

Either you stop pulling in the middle of the stroke so the machine thinks you've started the next one, or there's a dodgy connection that stops the impulse stream, or maybe even electrical interference.

The force curve could give you an idea of what's happening.

[PaulDuggan]

Thanks for the response. My technique isn't perfect but I don't think I'm that bad! How do I see the force curve please? Is that something you can do with a PM4?

I'm almost convinced that it must be the sensor/cable at fault, but the weird battery messages I'm getting from the PM are throwing me a bit.

The rower is in a garage, close to two refrigerators and not far from my gas boiler. Could these generate enough interference to throw it?

[chrisl]

Hi Paul, some ideas:

1. Are you seeing these errant readings directly on the PM4 or on ErgData's display too?

2. You mentioned using the Polar receiver. The socket for the receiver is near to the flywheel input socket. I'm wondering if the cable there, or the jack or socket are a bit flaky, maybe got knocked or partially pulled out at some point. Try unplugging the flywheel cable and giving the plug a firm wipe with a clean dry cloth. Then plug it back in and give a little twist back and forward a few times to ensure it's seated all the way in and making good contact, and then rotate the cable 90 degrees out the back of the PM4 so it's out the way of the other sockets.

3. I think your assessment of the cause is good, it looks like either the PM4 is trying to make sense of flaky data (issue with any of jack, socket, cable, sensors, etc) or else is itself faulty. It has the latest firmware. I've noticed that ErgData usually records errant stroke rate readings in the 80s when stopping rowing for a moment. That appears to be an artifact of ErgData not smoothing its output slightly, and is analagous to what you're seeing with the PM4 itself, which you should not be seeing. Hence me being inclined to agree with your assessment of the root cause.

4. If you're friendly with any gyms nearby perhaps you can persaude them to let you borrow a PM4 and flywheel cable overnight one night so you can swap them out one at a time and see if the problem goes away, that could help identify the cause.

5. There's a user on here Carl Watts who is a monitor service technician and maybe he will have some thoughts on other diagnostics or things to try.

To access the Force Curve screen see the link below - if you're pulling correctly then the curve should be nice and smooth, and a flaky connection may result in a messed up line instead.

http://www.concept2.co.uk/indoor-rowers ... orce-curve

Chris

[Carl Watts]

You may have some corrosion on the board, this typically appears in the tach area of the board. The corrosion is on the reverse side of the board that you cannot see after taking out the 4 case screws. You need to be very careful on the PM4 and unplug the RF board that is attached to the rear of the case. Two more screws enable you to remove the board after also unplugging the LCD display. The connectors just slide away by evenly pushing both ends of the plastic.

The plug on the pickup sensor is stuffed and the cable is intermittent.

The tach socket is cracked in half if you look at it closely or is loose on the board.

The battery level fault is one I have not yet nailed down but multiple faults suggest corrosion. Dont ask me where it comes from, I see boards from 2003 that are spotless and others that are from 2012 that are stuffed from it. Possibly its sweat getting in or if its been in a coastal area the salt air or someone sprayed some sort of cleaner on it at some point.

[PaulDuggan]

Chris:
PM4 and ErgData display match each other perfectly.
I've checked and re-checked connections. I've looked into taking it to a gym it's still a 'possible', if I can persuade the suspicious manager that I'm not trying to swap my broken PM for one of his good ones(!). Thing is - since the problem is intermittent - if it exhibited the problems then that would tell me that the PM needed replacing, but if it didn't it would tell me nothing.
Thx for the response and force curve link.

Carl:
Thanks for responding. I will open it up and see if I can see any corrosion or cracks. If I manage to do it without wrecking it I'll let you know what I find!
Can corrosion be fixed by resoldering? Or would replacement be unavoidable?

[jamesg]

Suggest you leave well alone unless you've nothing to lose. The occasional spike in the bar chart seems to have no effect on averages.

I see a few spikes with PM alone, more when I plug in a laptop on battery and a lot more when I connect it to a power supply. Nearby machinery could also have some effect.

[PaulDuggan]

Unfortunately this isn't spikes or glitches. When you turn it on it might work properly for the session or the data might be completely duff for the whole session. Just had another duff one today; the only thing on the display that was correct was the HR. Even the elapsed time was wrong and incrementing about 40s for every elapsed minute, so it looks like the PM has had it.

[Citroen]

You should contact Concept2, there's probably a limited warranty on their refurbished kit.