RadPowerBike RadCity Commuter

I haven't really investigated it, bit it seems like I read about the variable regen via the throttle. One side of neutral to go, the other to stop?
 
AHicks said:
I haven't really investigated it, bit it seems like I read about the variable regen via the throttle. One side of neutral to go, the other to stop?

That's how the PhaseRunner does it.

The Sabvoton I have in the Borg has a separate analog input for the regen input, and a contact closure to enable it. With the PhaseRunner setup the throttle and regen controls have to be combined into a single analog voltage.
 
Alan B said:
With the PhaseRunner setup the throttle and regen controls have to be combined into a single analog voltage.
That's different from what the product page on Grin's site says:
Proportional regen available through throttle signal or stand alone wire

Maybe that is only for the "v2" units?

Hmm....the manual (for v2) says:
The Phaserunner has an analog ebrake line which is tied to the throttle line, and the regen voltage is mapped so that regenerative braking starts at 0.8V and then reaches maximum intensity at 0.0V.

With the brake and throttle lines tied together this way, the Phaserunner can support variable regen through bidirectional throttles or a V3 Cycle Analyst, with just a single wire for forwards and braking torque.

So that doesn't quite match what the product page says. But I think it is probably wrong, because that's also exactly what the v1 manual says, and I think maybe they didn't update it?

Or else...it just doesnt' make sense anyway. Or there are electronics inside the PR that combine the two inputs to prevent signal conflict?
 
The bigger point from my standpoint is the fact that you don't have to sacrifice the rear brake handle to control regen. Or maybe I misunderstood the thought there.

I understood that whatever they're doing, it's done with the throttle. Beyond that, I'll figure that out when (if) I need to.
 
They tie the ebrake and the throttle wire together inside the connector, and program the braking at low voltages, and the throttle at higher voltages to minimize the number of pins and wires to deal with. If you separate them the voltage ranges don't change (they are configurable in the programming software to some degree), but that would facilitate driving them independently. I haven't done it so I'm not certain how they are disambiguated when they are separate. I had forgotten they could be separated, I'll have to keep that in mind if I mess with them.

I had always planned some kind of variable regen lever on the rear brake lever so the fingers could control regen and brake separately, but have not had to make one.

The PhaseRunner is effectively set up for a throttle that is spring loaded and turns both directions from an "idle" position so one way you get acceleration and the other braking. I don't know that anyone makes such a throttle but it could be a good way to do it.
 
Vectrix uses such a throttle, but they are quite expensive as spare parts. There's a thread around here somewhere discussing possible sources of such throttles.

I'm honestly surprised (and disappointed) that Grin hasn't designed and offered such a throttle in the years this feature has been on their controllers (at least five years I know of).

I'm also surprised (and disappointed) that Grin hasn't designed and offered an ebrake lever that can be used to control this.


On my SB Cruiser trike, I've got a pair of the Grinfineons that use the same setup for proportional regen, throttle controlled, but I do not like this, and need to use an ebrake lever to control it instead.

So I'm still pondering the design of an ebrake lever whose switch will engage a relay that will disconnect the throttle signal from the controller, and connect the analog sensor (cable-controlled pot, probably, pulled by the lever), so the lever then outputs the correct voltage range for the braking portion. Then when the brake is released it reverts signal connections back to the throttle. (the relay will also engage the brake lights, which at present are a separate lever).
 
Here is one example of a device I think could be used for variable regen with a brake lever (must scroll down the page to Cable Pull Throttle):

https://electricscooterparts.com/throttlesstandard.html

A couple more mentioned in other threads, but I didn't save the links.

I considered a double cable brake lever, but was afraid balancing regen and the disc brake would be difficult to balance. Besides, I want max regen before activating the discs. Plan on using a thumb throttle connected to the Phaserunner regen input and using a Littlefuse reed sensor 59022-3-T-02-F to send the eBrake signal to the CA3. The CA3 could on sensing charge current consider that brake application, but it's not a documented feature.
 
The PhaseRunner's ebraking input voltage swing is backwards, so it goes from 0.8V for ebraking to start to 0.0V for full ebraking, so the scheme used will have to account for that. An opamp can be used to invert the signal from a thumb throttle and scale it appropriately. A CMOS analog switch could be used to select which signal to apply to the PR. A Comparator could be used to determine the level and switch between the two inputs at say 0.9V. The problem with analog circuitry is in the details. Getting proper power (possibly bipolar), filtering, shielding, etc. The CMOS switch is susceptible to static discharge. With the various bias resistors, bypass caps and voltage regulation it quickly gets to a number of components. We used to do things this way but not these days.

One way to handle this is to use a small micro that has a DAC output, or a separate DAC chip that the micro drives. Use two analog inputs, one for throttle and one for ebrake. Then write a few lines of software to read both ADCs and map the output DAC appropriately. I would probably use a small Arduino for this, but most anything could do the job. A Raspberry Pi would be total overkill, but whatever is easiest for you. The nice thing here is you could determine the priority so a signal from the brake input overrides the throttle. I built something like this but have never put it on a bike. One more thing to package / mount / power / cable / connectorize / etc.

I've been riding the Borg lately, and it has the variable ebrake setup since there's insufficient space for a rear brake. It is a little un-nerving the way it is working. I have the "drag" or "slip regen" feature enabled and set to a modest level of regen, so releasing the throttle causes mild regen, similar to engine braking. So here's what happens when you come to a stop. First you release the throttle and get mild regen. Then you pull the rear brake lever, and as soon as the switch activates the regen goes to zero, it feels like a mild acceleration, exactly opposite of what you want (rather like pulling the clutch on a motorcycle). Then you pull the brake lever farther (with essentially zero resistance, it just moves) and regen goes from zero to very strong braking, but with a little delay. At very low speed the regen reduces and drops out (like 2 mph). Then the front brake (dual disc) is used to bring the bike to a stop.

When first pulling the ebrake the slight acceleration is unnerving. I wish it would transition smoothly from drag to variable regen. This could be fixed with a circuit that sets the minimum ebrake to be equal to the drag. But controller settings don't seem to cover it.

In general the ebrake doesn't feel quite like a brake. The lever moves a long way with little resistance, there is delay, and the braking varies inversely with speed. It is wonderful, but it is not what your brain expects for a brake. You have to adjust, and with the delay it is easy to think it is not working for a moment.

I tried putting the ebrake on another throttle. I did not like that at all. It was yet another thing to think about, slowing down the process and requiring new motions that are not automatic. The brake lever, even with it's flaws, is more natural and automatic. Years of motorcycle riding has burned the pathways into automatic reflexes.

The RadCity's front spokes are rattling due to being loose. It is like they built the wheel but forgot to tension it. The rear is better. I think I'm going to try slightly tightening an even amount on every spoke. Running it the way it is may be bad for the wheel. I haven't tensioned wheels so I'm going to go slow and then get it tensioned properly on Sunday. With these thick 12 gauge spokes it is easy to go from loose to pulling out the nipples and warping rims, so going in baby steps is required.

We use Topeak MTX compatible trunks, and they have a dovetail mounting system. The rack on the RadCity has a lot of features but not a Topeak track. Topeak doesn't seem to make a track that can be bolted to a rack, strangely enough. I searched and found a design on Thingiverse for a compatible 3D printed track. I also started a design of my own. I may put one or the other on the RadCity. That could be very handy.
 
Re: the electronics, I get your point about going for the brake lever instinctively. That's an outstanding point. Having the brake on the throttle would take some getting used to and still not be near as intuitive for somebody with a long background of doing the same thing another way. A 'youngin would likely adapt much more quickly.

On the spokes, I was concerned about pulling the wheel out of line with my adjustments more than anything else. I went around the rim twice, 1/4 turn at a time, and that stopped the clicking (that was making me nuts). I'm sure there are more accurate or better ways, but that worked for me!
 
The Phaserunner, according to the manual, allows you with the PC configuration program to remap the regenerative braking from the 1-0V backwards sensing default setting to whatever you want. Will probably use 1-4V, in theory pretty much matching a std throttle signal.

By default the regen input is electrically tied to the throttle input. This to coordinate with the CA3 standard scheme using a single throttle. But the wires supposedly can be separated so there are 2 independent inputs. One person has told me if both throttle and eBrake are both active, eBrake wins. I hope that's right, although the reed switch eBrake signal to the CA3 should disable any throttle output to the Phaserunner.

It's been 30 years since I spent much time on 2 wheel machines...but if you reach for the brakes in a instinctive situation, they are still there.
 
1/4 turn twice around sounds good on the spokes. I may try the same.

Since I haven't tried remapping the PhaseRunner ebrake I don't know how well that works. You'd have to change to a four pin throttle connector. Sounds like they have implemented it with the proper priority. It's been a long time since I reviewed the PhaseRunner thread, I followed it from the start. I have one of the original BAC500 flat prototypes, before they came up with the name, and a beta 2 and a pair of early production models that are on the Bonanza with the built in on-off pushbuttons. Mine are cast fairly clear so you can see the components. Now they are different, mostly different connectors and shells. I did experiment briefly with the regen and it did not seem to work, but that was quite awhile back, and many firmware revisions ago. I do need to work on a Front/Both/Rear control on the Bonanza, perhaps I should split out the regen wire as well. Only the rear DD needs regen the front gearmotor clutch prevents regen.

A PhaseRunner on the RadCity would be interesting. I don't know much about their DD motor characteristics, I wonder what the actual specs are.
 
Just following along at home with no electronics expertise or expectation of ever using this stuff, but I wonder if the switch needs to be so complex? The variable ebrake I guess is to be homebrew, with a 4V neutral up to 0V max pot, so you double it up with a cutoff switch and route the throttle power through that? It's awkward, and maybe the controller design isn't such a good idea as they thought, but doesn't seem like it really needs a relay or more complicated electronics?
 
Not to derail the topic, but electricbikereport has an infomercial (I say that because they charge to review a product) of a Radrover commuter that sells for $1299 delivered. It has a 750w Bafang geared motor and 48V, 14 ah battery. Seems like a steal at that price.
 
Hope the attachment works...the manual shows it like this:

78CA218C-BF8E-4D3D-A953-6C63DB410DA8.jpeg

So I'm hoping I can just add a 3 new conductor JST to Grin stds after separating the throttle input from the regen input as shown? The multipin CA interface (includes throttle) unchanged and the remaining throttle unused. It will require two new wires to the handlebar, analog brake and +5V. Figure I can steal a ground off the CA3 cable. Pretty sure that will work and it's the same as you said with more detail...but this is my first.

The documentation doesn't talk about the new L10 motor connector, but don't think that matters. It seems the doc lags the product, a little. But I have seen threads where regen would appear to be functional now.
 
That (separating the wires and adding a regen throttle connector) should work, based on the documentation. My proposal was for using the original 3 pin connector, but if you take their wiring apart and add a second connector (or use a single four pin connector instead) it will simplify the external wiring and you don't need to multiplex the signals onto one wire.

I think you can also do it with the Cycle Analyst and have it combine the brake and throttle signals. That has the advantage of minimizing the wire runs from handlebar to controller - the throttle and ebrake controls connect to the CA. - update - I reviewed the CAv3 manual and don't see anything about this there.

Due to the flexibility of the PhaseRunner and Justin's foresight there are a number of ways to do it and keep it simple.

Has anyone tried a PhaseRunner and CA on the RadCity with the standard motor and battery?
 
Alan B said:
That's the idea but I'm thinking smaller, more thumb size than palm size.

I didn't find anything on thingiverse, I'll probably just design something.

In the Vespa world we found big o-rings from Catipillar worked great.
https://www.amazon.com/8M4991-Caterpillar-Yellow-Silicone-O-Ring/dp/B00BQ1BL46/ref=sr_1_cc_1?s=aps&ie=UTF8&qid=1492875509&sr=1-1-catcorr&keywords=yellow+caterpillar+o-ring

https://youtu.be/BmwkuzbsiOg
 

Attachments

  • 04207327-BCFC-4CBE-8DF2-680D359CE915.jpeg
    04207327-BCFC-4CBE-8DF2-680D359CE915.jpeg
    98.3 KB · Views: 949
tomjasz said:
In the Vespa world we found big o-rings from Catipillar worked great.

https://youtu.be/BmwkuzbsiOg

That video is titled for a cruise control, not a thumb throttle conversion of a grip throttle?
 
Today my son made his first work commute with his new RadCity. I advised him to wait until we tightened the spokes on the front wheel, but he went ahead. You know how kids are. Even adult ones. :)

So he sends me a text from work. Battery hit bottom of the gauge, and the rear tire is flat. I take that to mean it the battery didn't quite run out, but is close. The flat must have been near or at work or he'd have called for help. So he'll need a pickup after work. So it saved us one trip at least. But now we'll have to remove both wheels, one for the spokes and one for the flat.

He was not having problems with flats on the Peugeot's Marathons and made many trips with them on this route. These Kendas are supposedly shielded against flats. The Marathons are also supposed to be protected, and they are narrower. We'll have to see what type of puncture it is.

It will be interesting to hear how well it climbed the hills with the power it allows.

A friend has offered to tighten and true up the front spokes, I guess I'll work on the rear flat. Another day in the life of ebikes.
 
With the bike flipped over on it's back, tightening those front spokes, and the rear for that matter, are a pretty easy operation.

I had a thorn, no bigger than one on a rose, cause my rear tire to go flat. That was the day I replaced the tube and added Slime. Haven't had an issue since, and the chronic need to add air frequently syndrome gone as well - finally! Have since moved on to Schwalbe Marathons because I had a "lump" in the rear tire starting and didn't trust it enough to ride it any further. It looked like the bead cords were letting go on both sides. I tried letting the air out and reseating the bead area on the whole tire, but no luck. So I replaced them. Marathons are much quieter and WAY less rolling resistance. I can carry a gear higher that I used to with all else being equal.

Point being, don't fall in love with those tires....

BTW, with any luck, the boy got a lesson on the use of PAS if he expects much in the way of running any distance to speak of. -Al
 
I hope he pedals more, but I don't know that has happened yet. His commute is near 16 miles (each way) with moderate grades and to run the battery nearly down that probably indicates very little pedaling. We'll keep working on that.

Better tires are probably going to happen, sooner or later. Maybe much sooner if there is tire damage here.

I have one friend who used incredible amounts of slime and still had lots of flats. Recently he changed to Stans and the flats are gone. I've seen articles on testing of anti-flat liquids, and slime was not at the top of the list. Combine that with the messy stains steers me toward something else. Generally I don't use anything, but if I have to it own't be slime next time. I think Stan's is fairly clear but I've not used it. The tubeless tire folks seem to prefer it.

One of my flats was a little piece of stainless steel wire, likely a short length of a single strand from a tire belt cord. It was amazing I was able to find it at all. It made a slow leak rather than a quick flat, but every day the tire was flat again. I think this was in a Schwalbe Marathon, too. Frustrating.
 
amberwolf said:
tomjasz said:
In the Vespa world we found big o-rings from Catipillar worked great.

https://youtu.be/BmwkuzbsiOg

That video is titled for a cruise control, not a thumb throttle conversion of a grip throttle?

I'm confused by your question... it's an o-ring using friction to hold the throttle position. Works great on MC's and Scooters. I never use twist throttles so I can't speak to the fit on eBike throttles. I had the style pictured. I assumed it was aboutreducing hand stress to maintain a steady speed. Did I misread?
 
The original issue was my son's preference for thumb throttles and this bike came with half-twist. So a ring with a tab was the general idea.

Using an O ring to increase friction is a solution to a related issue. Does "jamming" the throttle by adding friction create a safety issue? Or does the throttle still return to zero just more slowly? On a powerful ebike a stuck throttle could be exciting. :)

In the old days we used to add friction devices to motorcycle throttles as a "cruise control". However it had a lever to engage and disengage this friction. It could still get you into trouble, but when it was off the normal spring return of the throttle worked as usual.
 
Alan B said:
Better tires are probably going to happen, sooner or later. Maybe much sooner if there is tire damage here.

I have one friend who used incredible amounts of slime and still had lots of flats. Recently he changed to Stans and the flats are gone. I've seen articles on testing of anti-flat liquids, and slime was not at the top of the list. Combine that with the messy stains steers me toward something else. Generally I don't use anything, but if I have to it own't be slime next time. I think Stan's is fairly clear but I've not used it. The tubeless tire folks seem to prefer it.

My understanding is there are different types for use with tubes vs. tubeless if you aren't already aware. The bigger point is that SOMETHING in those tires will be better than nothing. That, and the mess you speak of with Slime isn't as big a deal as you often hear about. It's water soluble for starters, so worst case it's an easy clean up.
 
Back home after the drive of shame. Picked him and his flat rear tire bike up at work. :)

He said it went flat just as he arrived at work. We'll diagnose it tomorrow.

I asked him about the power. It took him longer to get to work, the average speed and power are down slightly from the 35A Infineon/9C/Cycle Analyst set to 30A. Not a surprise there. He had to pedal more. Not bad that.

Not much else to say. He didn't take the charger with him, and the battery was flashing on the last bar, so he would not have made it home anyway. Need to get another charger. Wish the pack was 52V. That would be a good thing for RadPowerBikes to do, it would differentiate their product and give them a bit more range and torque in some situations.
 
Back
Top