SRM dropouts, troubleshooting thereof

[Dammit]

I have a Rotor 3D+ SRM, 110 bcd, it's 2-3 years old and I'm the second owner- the original owner never experienced the issues I am having with it (I asked him recently).

My history with SRM's is:

1. Wired Cannondale Si with PC-V, which worked flawlessly, all the time. I bought it second hand and later sold it to a friend.
2. Wireless Cannondale SiSl with PC-VII, which had dropout issues which I thought was due to my positioning of the magnet. I bought this new, but sold it to a friend earlier this year.
3. Wireless Rotor 3D+ with PC-VII, which I bought last year from the original owner. It won't connect to any of my Garmins - 800, 910XT, Fenix 2, so all figures are from my PC-VII.

I only spot this happening when I am accelerating up a hill and look down to check what power I am averaging- i.e. can I maintain this, should I push etc.

The power reading changes to zero, then "-", where it stays for a while, then it changes to zero again and finally back to (say) 326.

It happened today, but I didn't think to look to see if the cadence dropped out at the same time, but looking at the data in Training Peaks I can see that both power and cadence dropped to zero at the same time.

Looking at this, notice where speed increases up hill, with zero power and cadence:



This is, not to put too fine a point on it, driving me nuts.

Is it likely that this is a magnet placement issue, a reed switch issue, both or neither?

Also, why won't the SRM speak to any Garmin devices? My Cannondale ANT+ SRM did.

I've also got a P2M and a Powertap, and they a) both work flawlessly with no dropouts and b) speak to any Garmin.

[Dammit]

[rma]

Most likely it's a magnet issue. When you pedal hard, the spider or bottom bracket can deform a little bit and that magnet-SRM distance gets too big to detect the cadence, and then power drops to zero. You should try to keep the magnet closer to the internal plastic of the SRM and, also, closer to the center of this round plastic. Maybe you could put a spacer between the magnet mount and the bottom bracket and use a second magnet attached to the mounts magnet. Just what I think.

I had this kind of trouble on the begining with my Rotor 3D+ and it would not pair with my Garmin. When I did solve the magnet placement problem, everything was OK

[sungod]

as above check the magnet

but if it happens in the same place on the road, the other possibility is a radar trap, the pc7 seems vulnerable to this, there're roads i ride where it'll drop out every time at the same place

while trying to isolate the cause i added a garmin, the pc7 dropped out but the garmin recorded without a break, i had the pc7 checked by srm, but they found no problem with it

[Dammit]

I'll move the magnet closer, however could the spider be deforming enough at ~350 watts to move it out of the magnets range?

I can't pair this SRM with a Garmin - it won't do it.

[Tinea Pedis]

Funnily enough, had precisely the same issue with my 9000 SRM and my Garmins. Great to see I am not alone in this issue. Especially after it went back to SRM and they said there was "no issue" (when they tested with a PC 7, not a Garmin).

It all got messy from there, with regards to refund vs unit replacement. However my solution (which I know does not help) was to refuse a replacement and move to P2M. Which has been flawless.

[Dammit]

I would, right now, happily swap the SRM spider and the PC-7 for a P2M Rotor 3D+ type S spider.

[mentok]

this looks like a magnet problem to me too. the horizontal distance looks ok (3-8mm is fine), but the magnet may need to be spaced away from the bottom bracket (ie, to get to a greater radius on the srm if you get what i mean). my bb30 frames have no real issue with magnet placing, but my bsa (especially the cromoly frame) need spacing and lots of putty to get the magnet into a good spot.

can you reproduce the issue in other locations? it could also be an ant+ interference issue, though that is pretty unlikely.

[Dammit]

Where on the plastic cover should the magnet be positioned? On my other (now sold) Cannondale one there was a white spot - on this the back of the cover has been worn slightly, leaving nothing to line up with.

[sungod]

it looks like it needs to be a few mm further from the axis (fwiw srm's current manual says 4-6mm gap between magnet and powermeter, with magnet base between 21-51 from axis, but not sure if that is with the same magnet)

in this pic (130bcd spider) the two blue things at top right are the sensors, so ideally the magnet radius should bisect them, especially if it is mounted further away from the side of the pm




you could try spacers under the bracket, another way is get a rare earth magnet, 2x5x10mm is about right and put it on the end of the existing magnet - i did this when trying to figure out my own dropout problem

if that works ok, you can either rely on magnetism to keep it in place, or (degreasing first) put a blob on gel superglue on the face of the srm magnet then put the re one on, it'll be secure but with a bit of force you can still remove it if necessary

watch out if the chain gets dropped or is dangling during servicing, the magnet will grab the chain hard!

[Tinea Pedis]

this looks like a magnet problem to me too

In my situation, I tried every magnet combination possible.

Incidentally, I swapped in my 7900 SRM with each new magnet position and it worked without a worry.

That was enough for me - the 9000 was returned.

[mentok]

interesting. the wired srm i have on my track bike is painfully sensitive about magnet/sensor position and the reed switch is plainly visible through the white cover. my mtb wired meter doesn't seem to care where the magnet is distance wise, but the radial adjustment is critical. my wireless mtb doesn't really care at all - you can be 10mm away laterally and wherever you want radially and it just works.

modern SRMs know whether they're moving forwards or backwards with their dual reed switches but i wonder what they do if the reed switches become "sticky". if one of the switches operates inconsistently or slowly then i wonder how it deals with that. could be some interesting error conditions. or it could be nothing

[pritchet74]

I had this problem once before. Solved by shimming the SRM magnet farther away from the frame. I had a strip of rubber (I think it came with a speed sensor so I could wrap it around my chain stay so that I could attach the speed sensor around this larger diamater) and I just put a hole in the middle of it for the screw to go thru. So frame, rubber shim, SRM magnet. Problem solved.

[Dammit]

Ok, so the conclusion is that the magnet needs to be positioned so that it is centred between the two edges of the plastic cover in order that the two reed switches pass directly underneath it?

With a BSA shell that's going to take some doing- better to tape the magnet to the chain stay, or are there BB mounted magnets designed for this?

None of this answers the "why won't the SRM speak to any Garmin device" question, however!

[claus]

Did you update the firmware? There's a recent update that might address the issue:

February 17, 2014

SRM released a firmware update for PowerControl7. The new firmware improves the ability of the PowerControl7 to receive
sensor data.


PS: you might also want to take a look at their FAQ.