LightningRods mid drive kit

[teklektik]

One last post in this burst of CA interloping in this thread....

One thing that I thought might be useful to get the throttle under control for more powerful mid-drives would be to use gwhy!'s throttle box rigged to the derailleur cable and feeding the V3 AuxPot input in lieu of a three speed switch (direct connect - not extra electronics required). This is a sleazy means to make the CA 'gear aware'. The cable motion would change the AuxPot control voltage which could be configured to limit current. Adjusting the Aux->MinAuxIn and Aux->MaxAuxIn settings would allow the CA so 'see' as much of the cable pull as desired (as many gears as desired) and scale it to whatever percentage of max current seemed appropriate regardless of the actual amount of cable motion (ie need not be the full motion obtained with the original throttle setup).

The idea would be to limit the current in the lower gears and increase it in upper gears. This would make balancing a 'good' AGain setting across all gears easier for Current Throttle mode. The throttle would be scaled across the full range of motion 0-100% of the AuxPot -limited current, so the full throttle would be useable in any gear - just more power in the upper gears. This should automagically make the low gears drivable as perhaps a 500W bike for bike path use while increasing to full power with upshifting.

This idea was floated in the V3 Beta thread but AFAIK hasn't been tried. I don't have a mid-drive to give it a whirl, but it might be helpful for certain riding situations. As always, a switch could be afforded to disable or override the feature.

Anyhow - good idea or bad - just a thought....

 

[eTrailster]

One last post in this burst of CA interloping in this thread....

Anyhow - good idea or bad - just a thought....

Interloping? Hell no! We all appreciate this discussion immensely.

I'm in the process of ordering 2 boxes from gwhy!. How would you propose rigging the box in the line and still switch gears? Split the cable somehow? Once running reliably, it seems like a fun weekend that could potentially yield very satisfying results.

I need to order a CA-3 for my soon expected small block. The one I need is the CA3-DPS for mid-drives right? The CA3-DP is for DD hubs.

 

[teklektik]

How would you propose rigging the box in the line and still switch gears? Split the cable somehow?

Ideally this would use something like a Hidden wire Brake Sensor with a ratiometric instead of digital hall - but I though the thottle box would be easy as a proof of concept (wish I was a machinist... ). So - some sort of ad hoc cable doubler seems in order. Unfortunately the ones I found were a little pricey or big so I might try to whip up a "poor man's cable doubler" that picked up the pull on the exposed derailleur cable run on the top tube.

The idea would be to add to the rear cable stop a kind of piggyback cable stop that slides over the derailleur sheath and butts up against the existing cable stop. This would provide a new rear stop for the throttle box cable. Then run the throttle box wire as far forward as possible to minimize the angle difference between the throttle and derailleur wires and fasten the new wire to the derailleur wire with a common throttle stop. Something like this:


This is probably a sub-par mechanical design, but the throttle box shouldn't require much pull and this level of fabrication is within my skill/tool set. I'm sure the MEs and machinists on this thread could do better. This uses about $10 of parts (Paragon Machine Works / eBay). The eBay link also turns up other useful dual cable clamps, etc that might work out more easily...

Anyhow - anything similar (even zip ties) is good enough to evaluate the AuxPot concept - no point in going crazy until the idea is proven to have merit.

I need to order a CA-3 for my soon expected small block. The one I need is the CA3-DPS for mid-drives right?

Exactly right.

 

[gwhy!1]

2 stroke auto scooter throttle cables are normally split cables ( carb and oil pump ) e.g http://www.ebay.co.uk/itm/THROTTLE-OIL-PUMP-CABLE-50cc-2-stroke-Scooter-/191574905050?pt=LH_DefaultDomain_3&hash=item2c9ac110da

you can also just by the junction box for the splitters http://www.ebay.co.uk/itm/Throttle-cable-splitter-junction-box-1-cable-in-2-out-free-UK-P-P-/201335976150?pt=LH_DefaultDomain_3&hash=item2ee08f30d6

but seem quite expensive

 

[eTrailster]

teklektik, the mechanical aspects seem fairly easy and I don't mind risking money on this kind of experimentation. It's educational and entertaining. Has Justin or another CA expert provided input on your idea? The expected CA response is way beyond my understanding. Now, gwhy! seems like the ideal candidate to test. What do you think gwhy!?

I finally got my bike together. She rides, but I have to fix a few screw ups. Although the tires have terrible rolling resistance, the I9 hub engages instantly. Hopefully this will ease shock on the system. The other thing that stands out to me is the chain wrap I got with the Saint. It seems better than most I have seen around a 12t cog. Also, the Saint spring seems strong and the pull/return sides run parallel, which is cool and maximizes chain ring engagement.

Things are looking promising. Now to order my CA3, tune the bike and wait for Mike. I guess I could crank the shock/fork pressure and peddle .

 

[teklektik]

you can also just by the junction box for the splitters http://www.ebay.co.uk/itm/Throttle-cable-splitter-junction-box-1-cable-in-2-out-free-UK-P-P-/201335976150?pt=LH_DefaultDomain_3&hash=item2ee08f30d6

Thanks! Looks perfect.

Has Justin or another CA expert provided input on your idea?

No - I haven't run it by Justin, but I wrote the V3 Guide and can say with certainty that It will definitely work electrically and it should give a broader gear range to the AGain adjustment for Current Throttle.

But - the real question is one of drivability and that's a 'try it' kind of thing. Even if it works, the matters of adjustment, particular application, and personal preference will certainly come into play. Even working as planned, the perceived effect may not be useful or pleasing - and some folks just don't like the V3 - Current Throttle response is too laggy for aggressive/technical riding. But - different strokes...

I just wanted to fly the curve shaper and gear-scaler ideas as non-obvious V3 tricks that some might find useful. No promises.

 

[gwhy!1]

Now, gwhy! seems like the ideal candidate to test. What do you think gwhy!?

I don't own a CA or a e-bike with gears so not in a good position to test

I used 2 of my boxes on my trials bike and a homebrew CA I used 1 box as a clutch and works 100% just like a real clutch on a gas bike. The clutch lever would set the current /phase limit on the fly and the throttle was a speed based control. So if you link the gear selection with current limit on a ca then this should also work just as well..

 

[givitago]

some folks just don't like the V3 - Current Throttle response is too laggy for aggressive/technical riding

[strike]I use the current throttle for some pretty technical riding and I have to confess, it isn't ideal. What is really lacking is the slow speed/startup response, in particular, when you are climbing and have come to a near stop for a tight switchback, the delay is enough for you to loose all your momentum and balance. I have tried to compensate by using a fast ramp rate, but with a low gain setting to try to smooth out the power delivery. It sort of works, but isn't ideal - you cannot power wheely with a setup like that but it is much more manageable on the trail.[/strike] Edit - using 1BFC's config below makes current control much better. Thanks 1BFC!

One gotcha with this setup is that if you use an ebrake and leave the throttle slightly open when at a stop, when you let off the brake, you end up with full throttle. Apparently the ramping algorithm resets when you put the e-brake on, but the gain algorithm does not. Someone else was injured because of this - and it is an irritant for me for sure.

I'm still waiting for a good FOC/Sine wave solution to try out. Adaptto looks fairly interesting but they really haven't targeted the mid-drive market. I understand that you have to use your ebrake connector to use an external speed sensor, and even that sensor gets disabled once the motor starts moving. That may work with a gear motor, but due to the variable ratios of a mid-drive, it messes up the settings. Not only that but you have to short your throttle to get your e-brake back. I'm hoping Grin's BAC 800 might be a good option. It would be nice to quieten the motor a bit too.

 

[r3volved]

What would be nice (imo) is an adjustable ramping curve rather than linear ramp rate.

 

[LightningRods]

What's wrong with the 3 position power switch for dealing with low speed throttle control?

 

[gwhy!1]

What's wrong with the 3 position power switch for dealing with low speed throttle control?

you got to remember to use the switch... and never switch up with the throttle open in a low gear :lol:

 

[LightningRods]

Hi gwhy! Post a link to information about your throttle please. I'm very interested and I'm sure a lot of other people on this thread are as well.

 

[1BFC]

So I modified the cav3 settings some more and I can now lift the front with throttle on demand from low speeds. Lag time between twisting the throttle to wheel lifting is instant. This is achieved without killing the ridability of the bike either. You can still creep and roll on smoothly.

Settings I have found that impact the same desired result that may not be immediately apparent are:

1. The aGain setting directly impacts throttle voltage ramp rate.

2. Clamping aGain to <150 (on my setup) prevents the throttle voltage output signal from rising very quickly, even if you set voltage ramp to 99/sec. This makes twisting the throttle and power being applied feel extremely laggy.

3. Set aGain high and reduce voltage ramp to 0.8 to 1.2v/sec.

4. Fastramp stays at 5

5. Be sure your wGain isn't fighting the aGain. If you are using both amp limits and power limit's they can result in mixed results as they fight each other.

Maybe I should have known, but I didn't think limiting current ramp would impact the voltage output ramp rate. Learning this fact allowed me to dial in the throttle nicely.

Try it.

Also, going to need a brace.....anything over 1500w and the kit starts moving around and creaking. To be fair, I have had some "oops" moments while dialing in throttle where the kit got hammered by full instant unlimited power that was pretty brutal.

Adding the brace will fix it. I can clearly see the kit flex and that eyelet mount hole Mike added will nip it in the bud. My only suggestion is that everyone use it from the start.

Nothing is bent or broken, no chain drops either, things just squirm around without a brace and given how much traction I have the rear wheel doesn't slip so all of the torque goes towards trying to rip the jack shaft off the down tube.

I can't fathom someone actually putting 10kw through this from a performance perspective. It's insanely powerful and capable at "only" 2kw. will hold off until I get a brace setup before going past 1500w on a regular basis.

-R

 

[r3volved]

Nice report
I haven't had any flex issues at all. My frame is rectangular and perfectly sized for the kit. Double hose clamp seems to be working well...at least it seems that way right now at 4.7kW peak acceleration.

 

[gwhy!1]

Hi gwhy! Post a link to information about your throttle please. I'm very interested and I'm sure a lot of other people on this thread are as well.

Here is the sell thread for the converters.. it also contain all the info needed to use one and set it up. I not longer sell a compete kit i.e throttle and cable as Most people will want there own throttle choice .. Throttle converter box

edit :

cheap why of splitting a cable 1pull, 2 pulled 8) http://www.ebay.co.uk/itm/BMX-Bike-Gyro-Brake-Cables-Front-Rear-Upper-Lower-Spinner-Rotor-Set-Kit-/221516852997?pt=LH_DefaultDomain_3&var=&hash=item33936eef05

 

[LightningRods]

That's great! Thanks gwhy. I will for certain try one of your throttle boxes to see how it works with my kit. I'll be in touch.

 

[spinningmagnets]

Some riders prefer the "feel" of a common moto throttle (with the hall sensors moved to inside the battery box) which are a very mature design and have a wide selection of makes/models, but another benefit is that the gwhy method makes the throttle very water-proof, for unexpected showers...

 

[1BFC]

Last update.

After adding a small brace and tightening a bolt that was previously just snug, no more flex issues.

Bike hits 42mph on 3kw. Much faster than my needs, happy with the build as it stands today.

I am amazed the stock sram 10spd chain is holding up along with the Novatek rear hub. Bike remains best ridden in the 13t rear.

I will be converting to larger front 219 sprocket as my last change to be able to use 15t rear.

Thanks again to Mike for the support.

 

[LightningRods]

Thanks for the feedback. Yes a torque brace of some sort is recommended for high performance users. For 1500 watt commuter use it shouldn't be necessary but if you want to "hammer" the throttle, please give the kit a chance and use the provided brace mount.

42 mph on a 4.8" tire fat bike. That's as fast as I would want to go on one!

 

[Rotems]

Hello Mike,

Can you please respond to my email regarding the spare parts?

Thank you
Rotem

 

[Remus]

Hello Mike,

Great looking kit, I will get in touch about buying one.

Thanks,
Remus

 

[r3volved]

Here's my mostly-complete (rideable) LR build for all those living vicariously through this thread.

 

[LightningRods]

Very cool! 8)

 

[eTrailster]

Elementary question.

I don't understand how 12ga wires flow the amps we push. 12ga is only rated for 15-20A, yet nearly everyone's batteries push 40A + and there are 12ga wires for battery and motor from my controller. I don't yet have my kit, but I assume the motor leads are 12ga as well. Okay, they are short runs, but still seems undersized. Then, I'm running a backpack battery to the controller, which may be 5+ feet. I'm thinking I should run 10ga for that or maybe double 12ga?

 

[spinningmagnets]

I don't understand how 12ga wires flow the amps we push. 12ga is only rated for 15-20A, yet nearly everyone's batteries push 40A

The high reduction provides an incredible amount of torque multiplication. As a result, the peak amps are only touched occasionally, and even then for only short amounts of time. This is because the motor accelerates to a high RPM fairly quick, because the rider has the option of shifting the gears the motor is using.

If you bog this kit down by starting from a dead stop, in top gear on a steep uphill...it will either get hot, or the rear tire will break loose and start spinning. The 12-ga wire is a tribute to how efficient this kit is at providing exceptional performance from only 50V. Big rear hubs are famous for sucking the battery dry in a short amount of time due to low motor RPMs demanding big amps...

edit: I'm not hatin' on the big hubs. The trend is for off-road to move the motor onto the frame, but...for the street? big hubs (Cromotor, MXUS, Crown, H40, Leaf, etc) are still hanging on, and their users don't seem to mind the weight in the rear wheel when riding on fairly smooth streets. Since a hub frees up the entire frame for holding battery volume, I can see big hubs surviving for a lot of customers...