Radwagon 72v GMAC Phaserunner Build
- Thread starter RSwannabe
- Start date
Bullfrog - Thanks. Its been a fun project and I'm loving riding the bike and how it performs now. I'm actually showing 16.6 WH/mile in my most recent charge cycle and 16.7 in the previous.
On a separate note, I've been seeing some interesting statistics regarding my regen braking. I've read plenty of comments and posts that regen braking is not worthwhile, or if it is, it's benefit is only so you can save wear on your brake pads and discs. I've read that you only get about 5% back at best, but I'm seeing far more than that. Attached are pics of my CA3 for my most recent charge cycle: I used 8.8 amp hours of battery (72v 10AH battery), but gained back 1.3 amp hours through regen, for a net expenditure of 7.5 AH in about 30 miles of riding. That's a regen recapture rate of 17.7% and is way higher than I have seen any other reports of, but I believe it. My riding is very hilly in city with lots of stops and starts, and I use only regen braking probably 98% of the time. I can do this because the GMAC geared motor regen braking is much stronger than what you get with a direct drive hub. The stronger regen also means its putting more power back into the battery. Also, the proportional regen control where I use the throttle lever to increase or decrease the amount of regen, gives me great control for faster or slower stops.
Anyone else seen results like this?
On a separate note, I've been seeing some interesting statistics regarding my regen braking. I've read plenty of comments and posts that regen braking is not worthwhile, or if it is, it's benefit is only so you can save wear on your brake pads and discs. I've read that you only get about 5% back at best, but I'm seeing far more than that. Attached are pics of my CA3 for my most recent charge cycle: I used 8.8 amp hours of battery (72v 10AH battery), but gained back 1.3 amp hours through regen, for a net expenditure of 7.5 AH in about 30 miles of riding. That's a regen recapture rate of 17.7% and is way higher than I have seen any other reports of, but I believe it. My riding is very hilly in city with lots of stops and starts, and I use only regen braking probably 98% of the time. I can do this because the GMAC geared motor regen braking is much stronger than what you get with a direct drive hub. The stronger regen also means its putting more power back into the battery. Also, the proportional regen control where I use the throttle lever to increase or decrease the amount of regen, gives me great control for faster or slower stops.
Anyone else seen results like this?
Attachments
The advanced regen available with the GMAC and their matching controller is what attracts me to the GMAC potential. The regen available with this setup is NOTHING like what most people are talking about that's available with DD hubs - for the reason you mentioned. This thing is gear driven and turning 5 times faster than a DD making for a MUCH more capable regen function. Regen that's available right down to nearly stopped, vs. maybe 12-15mph for some comparatively meager braking from a DD.
It's not hard for me to see you're replacing a significant %/amount of battery power in use. -Al
It's not hard for me to see you're replacing a significant %/amount of battery power in use. -Al
goatman
10 MW
RSwannabe said:
im running a phaserunner with a DD hub 17s 10ah battery. I got 13% regen the otherday. my other controllers would only give me 400 watts regen but by adjusting the phaserunner I saw 780 watts and the braking is awesome I rarely use the brakes. im only operating the regen with a push button are you using a separate thumb throttle for regen?
as noted above, because the gmac is being spun up by gearing, and because you have lots of stops/starts plus hills, you'll see higher regen than typical.
Ahicks - yes, I am of the opinion that not many people have experience with geared hubs without clutches that are therefore capable of regen braking. Almost all comments regarding regen braking are in regards to direct drive hubs. With the GMAC, I can use regen braking to bring me to a complete stop.
Goatman - It is interesting to see that you are getting 13% recapture with your DD setup. That is a higher number than I have seen others report. With my GMAC, I have it set to allow up to 1500 watts of regen back into the battery. I am often seeing 500-1500 watts regen during braking. I have hall sensor switches on my brake levers, so whenever I pull the brake levers the regen braking engages. To brake, I just lightly pull one of the brake levers which is enough to engage regen, but not have the brake pads engage the rotors. When I have the brake lever lightly pulled and regen engaged, I can then use the thumb throttle to vary how much regen braking I get. Full thumb throttle gives me maximum regen. The base level regen (no throttle) is still stronger than the regen my previous DD hub motor was capable of (the stock Radwagon setup showed up to 750 watts of regen possible). Full regen with the GMAC (with full throttle used) feels like I've grabbed a pretty good handful of brakes. While it may sound overly complicated, it is all very intuitive (for me at least) and easy to use in practice.
Amberwolf - Yup.
Goatman - It is interesting to see that you are getting 13% recapture with your DD setup. That is a higher number than I have seen others report. With my GMAC, I have it set to allow up to 1500 watts of regen back into the battery. I am often seeing 500-1500 watts regen during braking. I have hall sensor switches on my brake levers, so whenever I pull the brake levers the regen braking engages. To brake, I just lightly pull one of the brake levers which is enough to engage regen, but not have the brake pads engage the rotors. When I have the brake lever lightly pulled and regen engaged, I can then use the thumb throttle to vary how much regen braking I get. Full thumb throttle gives me maximum regen. The base level regen (no throttle) is still stronger than the regen my previous DD hub motor was capable of (the stock Radwagon setup showed up to 750 watts of regen possible). Full regen with the GMAC (with full throttle used) feels like I've grabbed a pretty good handful of brakes. While it may sound overly complicated, it is all very intuitive (for me at least) and easy to use in practice.
Amberwolf - Yup.
SamRich
100 W
That's a very clean build! You can barely tell that's it's electric.
I think I might steal that string LED idea. Thanks for sharing all those pics!
If it's ok I thought I'd share a few pics of my similar build.
Had my Radwagon for nearly 2 years now.We bought it used on CL (2016 model I think). The previous owner had made significant changes - new wheels/battery/motor/controller, which I didn't use for this bike. So basically started with a bare Radwagon frame.
First off I built a triangle 52V 30Ah (14S10P PF) battery housed in polycarbonate cover - bike has 30-50 mile range at decent/fun speeds (no speed limiter). The battery (BMS) has bluetooth capability and display to indicate battery temp, speed, remaining capacity and estimated range as well as individual cell voltage for diagnosis.
Second, the 1500 W motor is laced to a rear moped rim and 3" moped tire which pretty much eliminates the risk of rear flats. The controller is a 50A sinewave "dumb" controller for Aliexpress, but has been solid with regen.
Added steel fenders that clear the wider tires and four piston Magura MT5e hydraulic brakes. Finally, added wireless speakers for the kiddo behind the front seat.
Since then I also built a 72V 20s7P 30Q pack that fits in the same PC shell, but the 52V pack works great for now.
The PC shell:
Heated and folded
Honeycomb holder and panosonic cells:
during welding with k-weld:
3D printed case for ANT_BMS display.
I think I might steal that string LED idea. Thanks for sharing all those pics!
If it's ok I thought I'd share a few pics of my similar build.
Had my Radwagon for nearly 2 years now.We bought it used on CL (2016 model I think). The previous owner had made significant changes - new wheels/battery/motor/controller, which I didn't use for this bike. So basically started with a bare Radwagon frame.
First off I built a triangle 52V 30Ah (14S10P PF) battery housed in polycarbonate cover - bike has 30-50 mile range at decent/fun speeds (no speed limiter). The battery (BMS) has bluetooth capability and display to indicate battery temp, speed, remaining capacity and estimated range as well as individual cell voltage for diagnosis.
Second, the 1500 W motor is laced to a rear moped rim and 3" moped tire which pretty much eliminates the risk of rear flats. The controller is a 50A sinewave "dumb" controller for Aliexpress, but has been solid with regen.
Added steel fenders that clear the wider tires and four piston Magura MT5e hydraulic brakes. Finally, added wireless speakers for the kiddo behind the front seat.
Since then I also built a 72V 20s7P 30Q pack that fits in the same PC shell, but the 52V pack works great for now.
The PC shell:
Heated and folded
Honeycomb holder and panosonic cells:
during welding with k-weld:
3D printed case for ANT_BMS display.
goatman
10 MW
72v 10ah battery and 1500 watts regen, how many amps are you putting into your battery, im at 16 amps, my cells can take 4 amps at fast charge and im 4p but I think my regen amps are at 30 amps.
I notice you connected the phaserunner to a heatsink, have you noticed any heat there from the 1500 watts of regen? isn't that where the heat from the extra regen amps would wind up
I notice you connected the phaserunner to a heatsink, have you noticed any heat there from the 1500 watts of regen? isn't that where the heat from the extra regen amps would wind up
math.goatman said:
1500w / 72v = 20.8a
actual amps (and watts) depends on actual instantaneous voltage at any moment during regen, but if 1500w is an average, then the average amps is all you can determine from it.
goatman
10 MW
amberwolf said:math.goatman said:
1500w / 72v = 20.8a
actual amps (and watts) depends on actual instantaneous voltage at any moment during regen, but if 1500w is an average, then the average amps is all you can determine from it.
the phaserunner is a little confusing to me, if 20.8 amps is going into the battery then I guess that a battery can receive more amps through regen than a battery charger because its only a burst of energy?. I suppose I could pump up the phase regen amps to 60 and battery regen amps to 20 but that may be a little hard on my spokes unless I use a thumb throttle to slowly activate the regen. I was worried about cranking up the phase regen amps beyond what the battery regen amps. the controller has to dump the amps somewhere so I figured it was getting dumped as heat in the phaserunners heatsink.
rswannabe is a 20s3p ga 10 amp/cell battery so I figured the extra amps were dumping into the custom heatsink
on the ca, regen current (on the current readings page, that's "amin", if a negative number) is exactly the current flowing back to the battery, because it is using exactly the same shunt that it uses to measure the current out of the battery, to get the regen measurement, becuase the current is flowing on exactly the same path, just in reverse.
there's nowhere else for that current to go, so it is going back into the battery.
whether the battery can handle that kind of current, or if it's bms can, is a totally different thing, and that depends on those devices themselves.
there's nowhere else for that current to go, so it is going back into the battery.
whether the battery can handle that kind of current, or if it's bms can, is a totally different thing, and that depends on those devices themselves.
goatman
10 MW
okay thanks, ill just max out the phaserunner settings and use the CA
My understanding is many BMS do not regulate regen current. Thus when “charging” a battery via regen through the primary output wires, it bypasses the BMS charging charging circuit. That is the case on my battery. In researching what was safe amount of regen amperage, I was told because regen is usually short bursts you can use a higher amperage than would be safe for prolonged charging. The short durations don’t allow the battery to get too hot or create dendrite structures. Justin at Grin indicated he has not seen issues with higher regen amperage hurting batteries.
In regards to the heat sink on my phaserunner, I’ve never even felt it get warm. I don’t think I run my setup near hard enough to really stress it. More research also indicates I’m far more likely to overheat the GMAC motor before the phaserunner gets too hot. And I’ve not gotten close to overheating the motor yet either.
In regards to the heat sink on my phaserunner, I’ve never even felt it get warm. I don’t think I run my setup near hard enough to really stress it. More research also indicates I’m far more likely to overheat the GMAC motor before the phaserunner gets too hot. And I’ve not gotten close to overheating the motor yet either.
My 20 amps max regen is just under a 4C rate. (CORRECTION- it is actually a 2C rate. See posts further down). My Panasonic Sanyo GA cells have a charge constant of 1.675 amps. Thus a 3p battery configuration gives a 1C rate of just over 5 amps (5.025 to be exact). So 20 amp charge rate is 3.98C for my battery.
To be clear, I am not a battery expert. My research and talking to people like Justin and others with far more experience than I leads me to believe that charge rates of up to 4C for short bursts won’t be significantly detrimental. But I have no solid data to back this up.
To be clear, I am not a battery expert. My research and talking to people like Justin and others with far more experience than I leads me to believe that charge rates of up to 4C for short bursts won’t be significantly detrimental. But I have no solid data to back this up.
Short bursts I guess is the key, as in a few seconds.
Once getting the cells hot, maybe not "damage" but reduced cycle life longevity of the back end.
If already being stressed by a lot of high C-rate discharging, might only be single digit percentages, NBD.
If the pack's otherwise being coddled might be more significant.
Definitely not a good idea as the cell temp is cold, as in winter.
Once getting the cells hot, maybe not "damage" but reduced cycle life longevity of the back end.
If already being stressed by a lot of high C-rate discharging, might only be single digit percentages, NBD.
If the pack's otherwise being coddled might be more significant.
Definitely not a good idea as the cell temp is cold, as in winter.
it doesn't exactly bypass the bms...it's still going thru it's fets. but if the bms uses separate charge and discharge ports, then it cannot do anything about overcurrent or overvoltage coming in on the discharge port (it will turn off the charge port when these happen, but it doesn't "know" it should turn off the discharge port because it shouldn't be coming from there).
if it has a single common charge/discharge port, then it can turn off the port upon overcurrent or overvoltage, so could protect against overexuberant regen currents or voltages.
fwiw, if you have a bms with separate ports, you should be able to use sets of schottky diodes of sufficient power level on the output to block regen from entering the discharge port, along with another set to prevent output from the charge port, and then wire from the controller side of the diodes to the charge port diodes, so that regen current is run only to the input, and discharge current is only run from the output. my brain is shutting down some so i can't draw this out right now.
it shouldnt' be necessary to do this but if you wanted an extra safety feature....
the catch with the diodes is that they do drop voltage, and thus create heat, so the more current you push thru them the hotter they get, and tha'ts all wasted power.
if it has a single common charge/discharge port, then it can turn off the port upon overcurrent or overvoltage, so could protect against overexuberant regen currents or voltages.
fwiw, if you have a bms with separate ports, you should be able to use sets of schottky diodes of sufficient power level on the output to block regen from entering the discharge port, along with another set to prevent output from the charge port, and then wire from the controller side of the diodes to the charge port diodes, so that regen current is run only to the input, and discharge current is only run from the output. my brain is shutting down some so i can't draw this out right now.
it shouldnt' be necessary to do this but if you wanted an extra safety feature....
the catch with the diodes is that they do drop voltage, and thus create heat, so the more current you push thru them the hotter they get, and tha'ts all wasted power.
thepronghorn
1 kW
RSwannabe said:
The capital "C" stands for capacity. Since you are putting 20A into a 10Ah pack, you are charging at 2C during regen.
Tommy L
100 kW
I just found out about the GMAC and I just finished my BBSHD fat build. Omg.... now I need to do another build. I love regen, but my 10 year old DD build is heavy.
Question..... is the GMAC torque sensing during pedaling like a Bionix? I thought the BBSHD would be torque sensing since the cranks are right there.
My Bionix was excellent for a smooth workout while riding long distances.
This GMAC has me intrigued
Question..... is the GMAC torque sensing during pedaling like a Bionix? I thought the BBSHD would be torque sensing since the cranks are right there.
My Bionix was excellent for a smooth workout while riding long distances.
This GMAC has me intrigued
The GMAC does not inherently have torque sensing. The torque sensing will Come from a seperate torque sensing bottom bracket and your controller setup. Mine works well enough, but I’ve not ridden a bionix so I can’t compare. I suspect mine could be tuned to work even better than it does if I were to play with my settings and scales in the controller. I haven’t bothered because it works more than we’ll enough to suite my needs.
Tommy L
100 kW
RSwannabe said:
Thank you for your reply!!!
That’s great to hear about regen.
Do you find the 2000 watts torquie
With it being 5to1 ratio???
Pulls away quickly ???
Take care!
Tommy L sends.....
