Assist cuts out occasionally

[thelarkbox]

Just trace all the sensors etc back to where they plug in, disconnect and label everything except the display battery and pas sensor. Then find some way to replicate the problem, a bike stand would be great I have a loop of cord i can string up over a beam in the garage and suspend the bike rear off the ground..

If the problem persists you have eliminated all but 2x elements the controller/display or the pas sensor.

A pas sensor is the cheapest/easiest to exchange and verify simply.

If the problem can not be replicated with just the pas sensor input try adding back brake sensors and he throttle 1 by 1 until it can be replicated to identify what is the problem.

 

[Nealh]

Very highly unlikely imv to be the PAS as it would occur intimittently more often,
the mosfet scenario in the hotter climate will likely be the cause and easier to replicate by riding then playing about on a bike stand.

 

[saneagle]

Just trace all the sensors etc back to where they plug in, disconnect and label everything except the display battery and pas sensor. Then find some way to replicate the problem, a bike stand would be great I have a loop of cord i can string up over a beam in the garage and suspend the bike rear off the ground..

If the problem persists you have eliminated all but 2x elements the controller/display or the pas sensor.

A pas sensor is the cheapest/easiest to exchange and verify simply.

If the problem can not be replicated with just the pas sensor input try adding back brake sensors and he throttle 1 by 1 until it can be replicated to identify what is the problem.

You need to eliminate the posibility that the controller is over-heating before anything, whichever way you are able to check it. It sits in the compartment at the front of the battery if you have a rack battery and used the compartment provided. Who did the conversion? They should know where the controller is.

 

[fudoka]

You need to eliminate the posibility that the controller is over-heating before anything, whichever way you are able to check it. It sits in the compartment at the front of the battery if you have a rack battery and used the compartment provided. Who did the conversion? They should know where the controller is.

The conversion was a DIY job with a kit from Woosh!. According to the pics on their web site the controller is inside the battery cradle.

 

[fudoka]

Just trace all the sensors etc back to where they plug in, disconnect and label everything except the display battery and pas sensor. Then find some way to replicate the problem, a bike stand would be great I have a loop of cord i can string up over a beam in the garage and suspend the bike rear off the ground..

If the problem persists you have eliminated all but 2x elements the controller/display or the pas sensor.

A pas sensor is the cheapest/easiest to exchange and verify simply.

If the problem can not be replicated with just the pas sensor input try adding back brake sensors and he throttle 1 by 1 until it can be replicated to identify what is the problem.

Replicating the problem is the problem Since it only seems to happen after a hill climb it's almost impossible to replicate on a bike stand. Added to which I don't have a bike stand here, not even a handy beam to hang it from.
It looks like I'm going to have to wait until it happens again and check for overheating and, probably, bite the bullet and admit I need to source a replacement battery.

 

[thelarkbox]

I see, and i did 'tldr' all details so yeah, though without the battery bms cutting out completely or an apparent voltage sag or drop or low voltage warning on the display its not screaming battery problems to my novice ebike fault finding sensors.. and with @saneagle with a un-challenged 'guru' status suggests mosfet overheating as a likely suspect.. I would take that seriously, when given all the info he generally hits ebike issues 'squarely on the head'.

tape a cheap temperature sensor

under the metal base of the slide for a real time monitor? - a lot cheaper than taking a punt on a new battery prematurely

 

[fudoka]

I see, and i did 'tldr' all details so yeah, though without the battery bms cutting out completely or an apparent voltage sag or drop or low voltage warning on the display its not screaming battery problems to my novice ebike fault finding sensors.. and with @saneagle with a un-challenged 'guru' status suggests mosfet overheating as a likely suspect.. I would take that seriously, when given all the info he generally hits ebike issues 'squarely on the head'.

tape a cheap temperature sensor

under the metal base of the slide for a real time monitor? - a lot cheaper than taking a punt on a new battery prematurely

Hmm! I wonder if my meat thermometer would work

 

[Benjahmin]

Your controller is certainly inside the front part of the battery housing. If you slide the battery out it's in the bit that's left behind. It will be tight in there, with all the connectors, but there will be air around the controller making it difficult to measure temperature from the outside.
Really it needs more ventilation or, better, mounting outside of the compartment. My metal boxed controller is mounted to the underside of the front of the rack on the bike in my avatar. All connectors have been wrapped in self amalgamating tape. So far it has survived Welsh weather, though I don't knowingly go out in the rain.

 

[fudoka]

Ok folks I'm getting a bit overwhelmed now. I've got to admit I didn't expect I'd get quite so much feedback on this and I'm sincerely grateful to one and all (to put it mildly)!
However, I can't check out everything I've been prompted to check right now and, as I say, this is one of those intermittent faults which are always a swine to track down. - following Murphy's Law they never happen when you are actually set up to monitor them.
I'll get back on the bike when I can and definitely check out the heating on the controller and report back if and when.

 

[Nealh]

Ride your bike as normal and if it works ok (without hills /overheating) then one has already narrowed down the issue . In all probability it won't be until one replicates the slow speed up a steeper incline using the same PAS level that one will see the fault again, when it does immediately check the controller heat out put without burning your self. As mentioned once cooled after a few minutes the mosfets will switch again and the bike will work.
If the same does occur then it isn't the battery that is at fault but the controller unit in the battery base unit.

 

[fudoka]

Ok, here's the latest.
I managed to get time to take the bike out on a run today covering the same basic route as the one which prompted my original post. The two major hills which prompted the cut-outs last time were no problem today - Murphy's Law anybody?
The only difference I can think of is that, this time, I wasn't starting with a fully charged battery - I doubt that tightening a bolt on the carrier rack had any effect
I stuck an oven thermometer sensor on the battery casing over where I believe the controller is located with insulation tape to try and get some idea of temperature changes and it got up to 37°C at the top of the two hills where the cut-outs happened last time.
Is the level of charge in the battery going to affect temperature?

 

[saneagle]

Ok, here's the latest.
I managed to get time to take the bike out on a run today covering the same basic route as the one which prompted my original post. The two major hills which prompted the cut-outs last time were no problem today - Murphy's Law anybody?
The only difference I can think of is that, this time, I wasn't starting with a fully charged battery - I doubt that tightening a bolt on the carrier rack had any effect
I stuck an oven thermometer sensor on the battery casing over where I believe the controller is located with insulation tape to try and get some idea of temperature changes and it got up to 37°C at the top of the two hills where the cut-outs happened last time.
Is the level of charge in the battery going to affect temperature?

You can't measure the controller temperature on the outside because there's a massive air gap around it. You need to open the compartment and fix the sensor to the side of the controller that has three screws in it.

 

[Woosh]

I managed to get time to take the bike out on a run today covering the same basic route as the one which prompted my original post. The two major hills which prompted the cut-outs last time were no problem today - Murphy's Law anybody?

it could be temperature related. Controller overheating is a real possibility.
What is the local temperature when you went for the last ride? Also, did you get in touch with Andy?

 

[fudoka]

You can't measure the controller temperature on the outside because there's a massive air gap around it. You need to open the compartment and fix the sensor to the side of the controller that has three screws in it.

Given that I'm using an uncalibrated kitchen meat thermometer as the sensor there's not a lot of point breaking open the compartment to get it closer - this was just as some sort of indicator to give me some idea of the heat involved. When I get some replacement dilithium crystals for my tricorder though ...

 

[fudoka]

it could be temperature related. Controller overheating is a real possibility.
What is the local temperature when you went for the last ride? Also, did you get in touch with Andy?

Actually it may have been warmer on the second ride, I didn't check but summer may have just arrived in the last day or so here in Crete.
I haven't got in touch with Andy - mainly as it's 5 years since I did the conversion so I'm well out of any warranty. I know Woosh are pretty helpful but I didn't want to push it.

The bit of info/reassurance I'm really after is whether the cut-out is likely to indicate a major problem or, assuming it's heat-based, it's just a safety feature to prevent a major problem so I just live with it and take it as a signal to stop and admire the scenery.

 

[Woosh]

if you know that it will probably be OK again after 10 minute rest, then don't do anything just yet. Wait until there are more repetitions so you can narrow down the possibilities. It would be useful if you could post here everytime the problem re-occurs.

it's just a safety feature to prevent a major problem

yes.

 

[Woosh]

The only difference I can think of is that, this time, I wasn't starting with a fully charged battery - I doubt that tightening a bolt on the carrier rack had any effect

that can be. We should not rule anything out at this stage.

 

[Nealh]

Eventually the cut outs will become more frequent with heat issue, the mosfets will eventually fail as they are subjected to more heating.

 

[saneagle]

Given that I'm using an uncalibrated kitchen meat thermometer as the sensor there's not a lot of point breaking open the compartment to get it closer - this was just as some sort of indicator to give me some idea of the heat involved. When I get some replacement dilithium crystals for my tricorder though ...

OK. In that case you shouldn't be surprised if you get stranded due to MOSFET failure.

 

[Woosh]

OK. In that case you shouldn't be surprised if you get stranded due to MOSFET failure.

he rode that bike and kit for 5 years.
The most he had to do is to switch the battery off and on again to reset the controller.