Putting together a GPR-S

[nedfunnell]

In this earlier thread, I introduced my new-to-me GPR-S.

Yesterday, I heard the news of the EVTV Fluence ZE batteries becoming available, and it got me excited about the project. So, this afternoon, I went outside and started popping panels off of the bike for the first time. I wanted to assess the problems with the bike that I was told about: power stuttering during low-speed acceleration, and the chain jumping off of the sprocket during cornering, especially when accelerating.

The seller's suspicion on the stuttering was a dragging brake rotor or a motor problem. I was able to eliminate the brake rotor idea because the rear wheel spins slowly and evenly to stop when off the ground. If it were dragging, it would stop more quickly and slow down more at whatever point it was dragging. I can't eliminate a motor problem at this time, but my next discovery makes me wonder:

The chain seems loose to me. My Katana's spec for chain tension is 1.0 - 1.4" in the middle of the span. The GPR-S chain seemed to be about 1.5 -2.0" inches of slack. The chain also seemed dry, so it is possible that the both of those could be causing the sprocket-jumping issue. I could also so it causing power delivery to stutter, where the slack in the chain makes the bike act like an impact wrench when starting off- the motor takes up the slack very quickly, then BAM the impact of full tension pulls the wheel along, even over chain speed, creating slack, creating a vicious cycle until consistent chain tension is established at speed. I will tension it before the next ride and see how it does.

Also, Test rides of the bike told me that there was something funny about the throttle- it seems that throttle position commanded speed, not power. It is strange to imagine the controller somehow controlling speed on a PMDC motor without an encoder- it has no feedback! I'm not entirely sure what is going on. Here's what I'm observing: if I apply throttle and get up to speed, then coast by returning throttle to zero, reapplying throttle results in no power until I reach a similar point of throttle I was at previously. I expected to receive some amount of power as soon as I got out of the deadband. I think that my GPR-S has the non-upgraded throttle that was originally available for purchase from EMS. Power delivery from the controller doesn't start until a decent amount of throttle travel has occurred, so maybe there is something funny about the low end of the scale on that throttle. That could just be deadband, too.

The controller is an Alltrax AXE7234. Does anyone know about the programming, if perhaps it has a setting that would cause this?

[chipper6]

If you are seriously interested in one of those packs, you should get it now. I have a suspicion they will be sold out by Friday.

Sent from my HTC One using Tapatalk

[Richard230]

I seem to recall that my GPR-S had a pretty tight chain slack. I think it was about one-half inch. To verify proper chain slack, all you need to do is to set the chain slack for a slight bit of play when the chain tension is at its tightest point as you move the rear wheel through its suspension-limited arc. This is easy to do with the bike's rear wheel (place a jack under the frame) off the ground and the suspension units unbolted.

My GPR-S came stock with an older Sevcon controller that seemed to be designed more for a fork lift than a motorcycle.

[nedfunnell]

I went out riding today. I wanted to test a theory about some of the bike's being caused by a loose chain. I tensioned it and the worrying thak-thak noise went away, although I still hear something funny. I loosened it a bit from there and I think it's okay now. I also realized that I'd been riding on under inflated tires for my test rides so I inflated them to near sidewall pressure. (Does anyone know where to get a manual for this chassis?)

After this, I did a ride to experiment with consumption. The Wh/mi on the CycleAnalyst read 140 when I got the bike, which was worrying. I've heard 80-120 Wh/mi was average and had been calculating with 120 as a worst-case scenario. I reset it and went for a 2.5 mile ride, which was a little bit hilly, a mix of 35/45mph areas and included a 55mph section with a hill. When I got back, I had used 130Wh/mi. I wonder if I need to update my expectations for range based on that higher amount of consumption. Does anyone else think that 130 seems too high, indicating a problem? I need to do some more data-gathering with real routes I will ride.

[Richard230]

I used to have a Tiger 250 owner's manual, but I gave it away when I sold my bike. Unfortunately that manual wouldn't be much help to you as it dealt mostly with the IC engine and told the owner just about nothing regarding maintaining the bike. I guess in Thailand no one maintains their own motorcycles. They either bring it to the local wrench or just keep running it until it falls apart.

When my 50 Ah batteries were new I could go 35 miles on level ground at an average speed of 35 mph or about 10 miles at 55 mph. I once ran the pack down completely and used exactly 40 Ah. I can't recall what watts per mile I was getting, but it must have been somewhere around 100. Does your bike roll easily? My chassis rolled like a bicycle, with very little brake or wheel bearing drag.

[podolefsky]

Did tightening the chain help with the stutter? I like your theory that the slack / tight motion could make it stutter.

I'm pretty sure your GPR-S has a Magura throttle. Those are known to get flaky with time, could cause a low speed stutter. You can test it by measuring the output with an ohm meter, turn it slowly and make sure it goes smoothly from 0-5k ohm.

Could also be bad brushes or a messed up commutator, but that's unlikely.

As far as speed control, the Alltrax is voltage control. It's not technically speed control, since as you say it doesn't have feedback, and it's not torque or power control either. Here's the deal - when it's spinning, your motor produces back EMF (basically voltage) proportional to RPM. BEMF counters the voltage from your controller. Let's say at 1500RPM the BEMF is 36V. And let's say your pack is 72V. Then your throttle varies voltage from 0-72V. 50% throttle gives you 36V. Below that, you have less than 36V, so you can't overcome BEMF, thus you get no current flowing to the motor.

Above 50% throttle, you get more than 36V. The current then is proportional to applied voltage - BEMF, so it acts roughly like torque control. But as the motor speeds up, BEMF increases and current goes down until it finds a steady state, and then it acts like speed control. It's kind of wonky, you just have to get used to it. There's no other option with an Alltrax.

[chipper6]

Manuals:

Owners:
http://www.tigermotor.co.th/ProductA.../Boxer_Eng.pdf

Parts:
http://www.tigermotor.co.th/ProductA...r250RSPart.pdf

[nedfunnell]

Noah,
Thanks for your response. I think that the stutter was abated by the chain tensioning. I think it really is an effect produced by the combination of voltage control (which you explained elegantly) and slop in the system. The motor is trying to keep relatively constant speed with the constant voltage, but the slack-tight-slack-tight of the chain really messed it up. I haven't ridden it since posting this (it is SNOWING right now, what the heck) so I can't tell you based on much experience.

I am not a big fan of speed control via voltage! I am bummed to hear that that's what the Alltrax does. Too bad. Replacement with a current-control controller may be added to my someday-list, way down below having a functional bike. The throttle that is on there now is a JiangLe or something like that, not Magura. I'll bust out my multimeter and check its sweep next time I'm working on the bike.

[nedfunnell]

Chipper- Thanks very much!

[podolefsky]

Damn snow!

To be honest, once you get used to voltage control it's fine. I rode with it for over a year before getting my AC system, and I'll have it on my next bike using an Alltrax.

I'd recommend upgrading to the new Domino throttle. It's a step up from the Magura - all sealed components. I don't know about the throttle you have now, but my guess is it would make a big improvement in throttle response.