Adrian's 2nd Build - Giant AC Full Suspension

[adrian_sm]

Great ride indeed. Jells Park is beautiful spot. Thanks again for suggessting it, and shouting me a coffee.
Was great to see the souped up bomber in action. It is such a well construction machine, and perfect for playing around on bush tracks.
Was also great to have a go of one of your eboo.st drives. It was very easy to use, and so light. Can't wait to see how this developes. Keep up the fantastic work.

Pity my ride home wasn't as fun. My battery BMS tripped, even though I still had 2Ah left, and I was too stupid to work out how to reset it.
(Note: you need to totally disconnect the battery, not just one of the leads. I think it has something to do with the controller caps not allowing the BMS to reset)
I also managed to bend my rear rim, a few k's from home too. I didn't realise my rear tire was a bit low on pressure.

So now need to get new rim, spokes blah blah. :x

In the mean time I am riding my road bike to work. The world seems to move by so much slower on this bike for some reason.....

- Adrian

 

[Kepler]

I sure didnt envy your ride home. 10km of peddling a 408 hub is hardwork to say the least. Just as well you're fit. Who says ebikes dont give you exersize.

Glad you liked the eboo.st. A completely different feel to a hub but still very easy to live with.

You did well on those steep hill especially clipped into your pedals. No room for error when you cant put a foot out when you need to. We will doing again soon. Maybe get a few of the other boys along next time. I found some great tracks out near Ringwood on Sunday we will need to have a go at.

 

[adrian_sm]

Finally got around to putting some balance leads on my headway pack. This allows me to monitor at a cell level. I only have a single CellLog8, so I can only do 8cells at a time.

I managed to run my pack down to LVC, so thought it a good time to see how my cells are holding up.

The full charge took in 9.7Ah, here is the data for cells 1-8. Looks like cell 2 (the green one) is the runt of the pack. It started with the lowest voltage, and hit highest voltage first. Assuming the celllog is well calibrated, it also shows the difference in bleed voltages.

 

[adrian_sm]

Looking at the last graph (and the fact that the charger was still putting out 19W even though the gren finished light was on) I realised I didn't give it time to finish balancing. So I put the charger back on for a bit, only got a tiny bit more action. But here it is.

The first graph only plots the first 10 minutes, but cell 2 (green) still got bled off more, down to a recorded voltage of 3.430 V.


Here is what is happens at the end of 2hours or so. The BMS seems to have decided that we are relatively balanced as a lot fo the cells start getting these little hiccups in voltage. Which I asssume is the bleed resistors coming on and off.


Still something weird happening on cell 2.

 

[chroot]

Look like your 2nd cell dying or dysfunctional. You might want replace it before it get bad possible lead fireworks.

 

[adrian_sm]

Okay bit more data from the same cells 1-8 of the headway pack. This if from a 15km 5Ah commute to work. so ~50% DOD from a full pack. Light green is still cell 2, and it is obviously very different internal resistance than the other cells.

From looking at this I think I am running the pack too hard. I get peaks on the CA of ~50 Amps, but they are just spikes, but I do see ~35 amps on take off. I think I will reprogram for ~30 amps as seen by the CA. So I'll drop the battery amps down from the current 30 amps to 25 in my controller, and see how it goes.

It is also interesting to see how high the voltage spikes go above nominal voltage, when I hit the regen. All very safe at <3.5V peaks. Will have to do the same to the LiPo pack, as that is in series so voltage peaks could be quite different.

From looking at all this data, I would only rate these as 3C cells. Nothing better. I must have got the older cells in the pack, as the newer headway cells definetely perform much better.

 

[Mark_A_W]

I don't think it's worth worrying about yet.


My pack is similar, and when a cell starts tripping the BMS before 8Ah (under my normal 1500w average load), then I'll fix it.

 

[adrian_sm]

I'll probably ease back on the battery amps. As I think it was cell 2 dropping too low that tripped the BMS at ~8Ah when I hit the throttle going up a hill, when I was on the ride with Kepler.

Unlike you I am out of amp limit fairly quickly, so I shouldn't notice too much difference.

Here is a plot of 4 of your headway cells for reference. Couple of interesting notes for others:
1) how much better your new replacement headway cell is compared to the others
2) how consistent you under load voltage is. Which means your controller is under battery current limit conditions most of the time
3) Mark uses from memory 8Ah out of his pack for this ride so ~80% DOD.

When you look at mine, the voltage keeps creeping up as I accelerate, since the controller gets out of current limit, and BEMF starts taking over.

 

[adrian_sm]

And here is what my LiPo 6s pack is doing on the ride home.
Notes:
- used about half the pack capacity from a full 4.1V charge.
- All cells behaving well.
- Regen doesn't create any nasty voltage peaks.
- Pack did recoververy slight after a 5min rest while I ducked in to the shops at ~33mins by ~0.01V.
- Obviously a very different discharge curve than LiFePo4.
-- LiPo has the slow voltage decline the whole way
-- LiFePo4 is pretty flat until near the end, but sags a lot under load (at least for my headways)

 

[Mark_A_W]

That's not correct.

My battery current limit is 35A or so (over 1800w).

But at cruising speed I am drawing 1500w.


It is Back-EMF limited for most of the time.

 

[adrian_sm]

Huh. :?

Sure looks like you are pulling the same current. I would have expected a bigger voltage drop while you are pulling 35A, versus your 25 amps (1500W/60V) while you cruise.

 

[Mark_A_W]

Yes, me too.

It's possible the logger is missing those peaks (well, troughs really). The would only be for a few seconds at a time, compared to ages at 1500w.

What's the interval between samples?

 

[adrian_sm]

It was set at 2 secs when we used it, but can go as low as 0.5 seconds. Feel free to borrow it again.

Still doesn't sound like BEMF is limiting current if you see "1500W for ages".

I'll drop the battery current limit on my bike tonight, and see what happens for me.

I also want to get some data of what the WOT power draw verses speed is for my different "speed settings" on the controller. So I either need to work out data logging from the CA, or hooking up my little video camera to point at the CA screen, while I goose it up some hills.

 

[Mark_A_W]

I meant ages on end - while I cruise. ~27A or 1500w roughly.

During acceleration it goes over 1800w, over 35A.



It's not sitting on the current limit. When I had the current limit at 50A it was still cruising at 1500w (27A). It's BEMF I tells ya.


And yeah, I still need to log 3 banks of 4 cells.

 

[adrian_sm]

All right. All right.

Grab the celllog tomorrow if you want. I'll leave it on my desk, and set it to 0.5seconds for you.

 

[adrian_sm]

I did some poor man's datalogging of the CA power/speed/volts.
I mounted my camera so it could see the CA screen, then went for a ride and played with the speed switch.
The 6fet is programmed for 25Amp Battery current, but I see 32 on the CA, so calibration is a bit out.

Here is the current settings I am using for the controller.
View attachment 2

And here is the power output for 50%,70% & 100% speed settings.
I have also included a graph of the power my bike requires to maintain speed on the flat.

So the switch clearly drop the peak current, as well as the top speed. It is a bit like switching to a lower battery voltage on the fly.
View attachment 1

I also plotted my battery voltage vs current draw, to look at the sort of sag I am seeing. This was a combined voltage of the 16S 10Ah LiFePo4 Headway pack, and the 5S 10Ah LiPo pack.

 

[adrian_sm]

And here is what regen looks like in terms of speed vs power back towards the batteries.

 

[adrian_sm]

Update.

I really need to replace this dodgy cell. I only got ~7.8 Ah out of my "10Ah" pack before the BMS tripped on the way to work today. I should be getting closer to 9.2 Ah asssuming good cells.
As I never usually use more than 5-6 Ah, I don't notice anything wrong, except when I have to make a few detours and go the extra distance.
The bigger issue is that it is probably affecting the other cells during the charge cycle, by allowing them to be overcharged all the time

According to a watt-meter I only put 7.4Ah back in with final cells voltages once the charge had finished, and cells had rest for a bit, still connected to the charger, of:
1 cell @ 3.48
5 cells @ 3.56-3.58
10 cells @ 3.67-3.69

Here is what half the packs voltqages looked like at the start of the charge, and at the end.
Cell 2 (the light green curve) is the dodgy one, which now obviously has a higher internal reistance.

I just hope that it wasn't the BMS that killed it. so when I put a new one in I get the same thing happen.

Time will tell.

Start of charge
View attachment 1
End of charge


- Adrian

 

[adrian_sm]

Finally order a new 52T 104BCD chainring from Rennen Design for $40 + shipping.

Arrived yesterday, got it on the bike today with a new chain. It just clears the rear swing arm by about 1mm.

Took it for a quick test ride, and it's great. I can still add useful power even going flat out down a gentle hill now. It will also allow me to use the bigger cogs on my freewheel more, which should improve the life.

I find a cadence of ~100rpm to be comfortable, 120rpm max.
This now equates to speeds of about 58kph, and 70kph when in top gear 52/11, previously 48 & 56kph for 42/11.
And when this freewheel wears out, I might be able to go to a quality shimano 7 speed 13tooth, and drop back to similar gear ratio I had before.

It also freed up my old 42T chainring for my hardtail MTB, that is now the test mule for my friction drive developement. So finally that bike is on the road.

Here are a few pics.




- Adrian

 

[SpeedEBikes]

Adrian, you got a good buy on that frame.

I've got a Giant AC1 that I've abused the hell out of and the frame has been bullet proof. I found the stock suspension too plush for the higher speeds and weight of my e-bike and put in much stiffer springs and heavier oil to stop the diving action during hard braking and lessen drive wheel hop on rough surfaces. That probably increased the stresses on the frame a bit, but it hasn't been an issue. I've got a heavy Crystalyte 5305 with a Methods 100 volt 200 amp controller programmed to limit at 85 amps with a 15 ah 24 s lipo pack. The bike is about 8 years old now and the frame is still solid. Though for the first 6 years it had a 20 s liion pack and a 35 amp controller, ie only about 1/3 the power of the current setup.

I made a rear rack which mounts a large pair of panniers. I've had 60 lbs in them, plus ~30 lbs of frame mounted battery and 25+ lb motor & controller plus ~10 lbs of U-lock, tools, pump, etc. and my fully dressed weight ~180 lbs. So figure 240 to 300 lbs of load and never a problem. Not to mention I live & ride year round in Chicago which has a lot of truly horrible pavement and fairly strong weather.

 

[adrian_sm]

Great to hear. I bought it without a lot of research and have been really happy with it.

The rear end is starting to feel a bit loose, when cornering at speed. Not sure if it is my relaced rear wheel needing the spokes tensioned or the wheel skipping out a bit due to poorly adjusted suspension. I haven't modified the springs or oil at all. Did you tweak the front or rear shocks?

Got any decent photos of your bike, the only one I could find was old and blurry.

Have you had to do any maintenance on bearings etc over the years? This is my first dual suspension bike so I haven't had to deal with the upkeep of them before.

Anyway great to hear that your bike has stood up to the abuse.

- Adrian

 

[SpeedEBikes]

I've think I've got some photos somewhere on a back up hard drive but have unfortunately lost track of where they are. Probably easier to just snap some new ones which would also show the latest changes. Although the bike is rather filthy at the moment... One way or another I should be able to post photos within a few days.

The only 'frame' problem I've had was the suspension pivots between the chain stays and seat stays seized up due to winter riding on slushy salted roads. But disassembly, cleaning and relubing resolved the problem for now. I also cracked the rear drop outs when I upgraded to the more powerful controller and turned up the current limit. But the fault was mine for using an inadequate torque arm. I welded the dropouts back together adding in extra material since I couldn't heat treat to get the original strength and then made a pair of much stouter torque arms and have not had any trouble since.

One other problem I had, though unlikely relevant for you, is a piston in the bike's stock front hydraulic disc caliper seized up, again likely due to winter riding on salted roads. I was unable to disassemble or repair it. I swapped the rear caliper in which is working ok for now. I've read that this is a somewhat common problem with the Hayes brakes.

As for my suspension, I reworked both front and rear. The front annoyed me because under braking the suspension would compress a lot. In panic stop situations it was very unnerving. On the rear the main problem was wheel hop over small bumps, especially noticeable while accelerating and/or cornering. Stiffening the rear suspension and damping didn't eliminate the issue 100%, but has made it hugely better. The bike would still be considered poor for motocross type use, but works well as an urban commuter. But this is not a fault of the frame design, but of me bolting on an absurdly heavy rear wheel motor.

If one was attempting to build the ultimate off road machine it would make sense to not use a hub motor. But in my case I think the benefits of simplicity, reliability and quiet operation outweigh the draw backs.

As for your loose rear end, is it loose on smooth pavement? If not, it is likely a suspension adjustment. Otherwise check if you have any play with the bike at rest. If you can noticeably rock the wheel side to side then you have a problem. It shouldn't be too hard to spot what moves and what doesn't and locate which linkage(s) are either loose or damaged. If everything feels tight at rest my first suspicion would be the rear tire. I've had numerous tire issues on my electric bikes that at first made me think I had a suspension or bearing issue. It might just be an unevenly seated bead, or it could be a damaged or unevenly worn tire. If swapping the tire out doesn't fix it, at least you'll have a spare which is a good thing. A wheel badly out of true can also feel loose while cornering although it should be very obvious.

 

[SpeedEBikes]

Ok, here's a new shot of my bike.



A full suspension Giant AC1 with a Crystalyte 5305 dd bldc hub motor using a Methods 100 volt 200 amp controller programmed to an 85 amp limit and nominal 88.8 volt 15 ah lipo battery. Has a lockable battery cage, rear cargo rack, heavy duty kickstand, Schlumpf 2.6x speeddrive, Cycle Analyst, and 3 speed range switch. Rain shield for controller and connections made from 2 liter soda bottles. Stock componetry of the bike includes the front 8" hydraulic disc brake and tubeless front wheel. Tires are Nokian XXX 2.4" tubeless run at 75 psi. They are less prone to a blow out than tubed tires. Since this bike is quite fast the blow out resistance is important.

 

[adrian_sm]

Wow. Thanks for the pics and advice. Your bike looks so different to mine now. A real beast. I like the rear rack you put on, I have to use a trailer for the kids now.

Looks like you have a much better rear shock than the one that came with my frame. I might hVe to start looking and tweaking the suspension because it looks like your tweaks helped with all the things I am now noticing about my bike. Like dive under breaking, hop on bumpy corners etc.

- Adrian

 

[adrian_sm]

Just an update on my 52t chainring. It's awesome. I am no longer pedaling like a loony tunes character when at speed, and it lets me drop down to low gear on the rear, which will probably improve life of the cluster.

I definitely advise gearing you pedal drive train such that you can still contribute at all speeds on your bike.