Cycle Analyst V3 preview and first beta release

[justin_le]

Hi Bill and thanks for the continued feedback:

1) The unbidden appearance of the AuxLevel screen when in regen still occasionally occurs, but it is now much less frequent. Maybe 1 out of 20 instances where with Prelim8 it occurred in 1 out of 3 instances. This problem is now a trifling nuisance, but I felt I should mention it as the problem has not entirely disappeared.

OK good to know. I've increased the noise bounds even 50% farther in the current build, and we'll see if this eliminates it entirely for you.

2) During the first regen (while the controller's e-brake switch is closed and current is flowing into the battery) the wh/mi display goes crazy, showing a random selection of positive and negative numbers, then as the ride progresses, the display eventually settles down to a reasonable figure.

Whoopsy, I'm glad you pointed this out since I only ever glanced at my wh/km screen after travelling a certain distance when the numbers were all stable. This turns out to be just a single typo in one line of code which has been fixed.

After riding more with Prelim9 I see that wh/mi remains zero in the absence of any regen until somewhere between 80 and 100 watt-hours consumed. At some point that I have not yet been able to observe--I do need to keep my eyes on the road some of the time--the wh/mi display jumps from 0.0 to a reasonable value.

The actual threshold would be at 83.88 watt-hours (2^23)

3) Thermal rollback is only current- and not current- or power-based.

In the end I'm thinking of having thermal rollback moved entirely to the power limit, rather than doing both amps and power, but this is a low priority switch over.

4) I observe a brief and disconcerting surge of full-throttle (750 watts, in this case) upon coming out of regen mode (controller's e-brake circuit changing state from "closed" to "open") while pedaling while in PAS throttle mode when the PAS Watts is adjusted to a modest value like 200 watts. This might occur as I approach a red light that will soon change to green as I continue to soft-pedal while using regen to reduce speed. The light turns green, I release the regen button (open the controller's e-brake circuit), and motor spins up to PAS Watts or it does following the brief surge. If I stop pedaling while in regen, release the regen button, then resume pedaling there is no surge.

This might have been mentioned before but can you clarify if you have your ebrake circuit going both to the Cycle Analyst and the regen motor controller, or just to the CA3, or just to the controller?

I have experienced the issue where if you are pedaling while slowing down with regen, then you stop pedaling and then release the ebrakes at the same time, there will be a short power burst because the pedal stopping hasn't fully registered in the CA3 yet, so it's assuming that you are still pedaling and doing what it should do (pedaling, and no ebrakes, let's power!). If you increase the value of the StopRPM Threshold, then the CA will be quicker to respond to the pedals no longer turning and this effect will be reduced.

I've got a few ideas for a clean way to eliminate this glitch and once I play with that and confirm it then I'll have a firmware to share that should address your points 1,2, and 4.

 

[justin_le]

I have the cycle analyst v3 beta 21 firmware...the throttle is plugged directly to the controller.

Ha, well that 2nd bit is your main problem. You need the throttle plugged into the CA3, not into the controller. The CA3 is there to accumulate all the input sensor signals and then send a single throttle voltage to the controller via the 6-pin CA3 controller plug.

If you have an old device like this you want to bring up to date, then first flash it with the CA3 Prelim6 firmware (this will restore all the default settings without overwriting your calibration):
http://www.ebikes.ca/product-info/grin-products/cycle-analyst-3.html#firmware-and-programming
Then after that, reflash it with the Prerlim9 code if you want to have the latest updates involving the Aux signal display and some improvements to PAS mode.

-Justin

 

[mrbill]

The actual threshold would be at 83.88 watt-hours (2^23)

Hey, thanks for identifying the exact value. Saves me from trying to pin this down by compulsively staring at the display while I'm riding.

This might have been mentioned before but can you clarify if you have your ebrake circuit going both to the Cycle Analyst and the regen motor controller, or just to the CA3, or just to the controller?

Aside from operating the friction brakes, my e-brakes are wired to the CAV3 only. I have a momentary pushbutton that I use as a "regen on demand" button that is wired only to the controller's e-brake circuit: closed when regen is active, open otherwise.

Incidentally, I find that I prefer using the minimum EBS Level on the Lyen controller (EBS Level = 0) as doing so allows just as much energy to flow into the battery but causes slightly less braking effect--I can use my friction brake if I need to decelerate harder--wasting less energy when I don't need to slow down as much. Though, to be honest I find it hard to feel the difference between EBS Levels 0, 1, and 2. I'm suspicious that changing this setting may have only placebo effect.

I have experienced the issue where if you are pedaling while slowing down with regen, then you stop pedaling and then release the ebrakes at the same time, there will be a short power burst because the pedal stopping hasn't fully registered in the CA3 yet, so it's assuming that you are still pedaling and doing what it should do (pedaling, and no ebrakes, let's power!).

This may be the same problem, except that I often prefer to continue pedaling through a regen cycle, just to keep my legs moving. The moment I release the regen button (as described above), the controller goes "full power" for a second before settling down to the set PAS Watts.

Perhaps what occurs is the following: When I hit the regen button the controller sees negative current, negative power, and the CAV3 then tries to increase the throttle signal to re-establish the set PAS Watts. Unable to do so, the throttle signal is pegged until I release the regen button, after which the controller sees the pegged throttle signal briefly until the CAV3 pulls the throttle signal down to achieve the set PAS Watts. I will observe the diagnostic screen and the throttle out signal the next time to confirm this.

If you increase the value of the StopRPM Threshold, then the CA will be quicker to respond to the pedals no longer turning and this effect will be reduced.

I went through a lot of trial-and-error to find pleasing RPM thresholds (startRPMThresh = 31; stopRPMThresh=13). Setting StopRPMThresh higher has power cutting out sometimes just as I need it most when powering slowly over a short sharp rise that causes my pedaling cadence to dip briefly. Setting startRPMThresh lower causes an annoying surge of power when I'm doing slow, parking lot maneuvers under leg power alone.

 

[NeilP]

I find that I prefer using the minimum EBS Level on the Lyen controller (EBS Level = 0) as doing so allows just as much energy to flow into the battery but causes slightly less braking effect--I can use my friction brake if I need to decelerate harder--wasting less energy when I don't need to slow down as much. Though, to be honest I find it hard to feel the difference between EBS Levels 0, 1, and 2. I'm suspicious that changing this setting may have only placebo effect.

Are you using Lyen /Keywin software to set regen level or XPD?

XPD allows a higher regen level

Also HVC voltage plays a major part. For 20 series with HoC voltage of 83 , I have my controller HVC set to 90-92 volt.
Regen never gives so much power to actually push pack voltage over HoC,but makes regen more effective
If you lived at the top of a massive hil and the first 1/2 mile of your ride was downwards umder max regen, then the OTT HVC maybe an issue, but otherwise, as soon as you have taken power from the pack, you'll never get regen to push pack voltage above a safe value.

 

[NeilP]

I have the cycle analyst v3 beta 21 firmware...the throttle is plugged directly to the controller.

Ha, well that 2nd bit is your main problem. You need the throttle plugged into the CA3, not into the controller. The CA3 is there to accumulate all the input sensor signals and then send a single throttle voltage to the controller via the 6-pin CA3 controller plug.


If you have an old device like this you want to bring up to date, then first flash it with the CA3 Prelim6 firmware (this will restore all the default settings without overwriting your calibration):
http://www.ebikes.ca/product-info/grin-products/cycle-analyst-3.html#firmware-and-programming
Then after that, reflash it with the Prerlim9 code if you want to have the latest updates involving the Aux signal display

Ummm , but still, and I am treading on thin ice here, contradicting the Guru Justin , running the CA in a legacy fashion , if its settings are correct, should not be giving the power limiting you are experiencing.
That is not the cause of your problem. I am thinking Justin missed your first post, as flashing at the moment, not an option without a cable. So longer term, get one.

If you go to first page of this thread, then either on that page, or one deep in from the first three links, you will find a list of all the Beta Firmwares, and downloadable sheets for writing down your settings..
First think,since you can't flash , is to download that spreadsheet, fill in all the values, then post it back here. I may not be able to help at that point, but it may give others a clue to why your CA is limiting the controller.

Potentially a wiring issue too, you'd need to confirm, from the UUG, that internal wiring is correct for how you want to operate, and that controller CA plug is wired in a compatible fashion.

 

[NeilP]

Here, halfway down page 10,
A list of firmware and UUG


http://www.endless-sphere.com/forums/viewtopic.php?p=571345#p571345

 

[Green Machine]

Ok thanks Neil,

I have a friend who has a cable he will lend me and i will try to flash tomorrow.

Unfortunately i found that the throttle is hard wired to the throttle and flashing is easier then re-soldering the throttle cables so i will try that first.

 

[NeilP]

Ummm
Throttle soldered to controller..... The wiring I was talking about is at the board level. You'd need to open up both the controller and CA to check the wiring on the boards.

The CA can pull up or pull down the throttle line on the controller either over riding or driving conttoller input,
Two pads on the CA, that the CA plug green wire can be connected to, and one or two, via a diode or not, at the controller end.
All depends on how old the controller is and how its mating 6 pin CA plug is wired to the controller board.

And regarding the cable, it is the CA programming cable you neec, which is different to a controller programming cable.

You may need to flash two firmware versions, you will have to read up .

Two different operations can be done with the CA USB cable, and two bits of software. One for flashing, one for reading/writing/ saving settings. So two bits of software from Justins website needed.

Once flashed, I'd over write your CA with default settings, since you don't know what the current settings are for/from and start from Page One of the UUG.

 

[Green Machine]

i meant the throttle wires out of the controller are soldered to the wires on the throttle...so i would have to clip them and reconnect them to the wires out of the ca to change them. yes i got that...i got a friend who has a ca programming cable i can use to reflash.

Now it seems like just clipping and connecting the throttle to the CA might be easiest thing to try first.

I am sure now this is a brand new v3 cycle anaylst. Just older firm ware.

 

[NeilP]

Yes, I realised you meant throttle soldered to the controller,

But the wiring of the CA to the controller maybe the issue. It depends on how old the controller is, and how the CA plug is wired.
The green wire on the CA plug may go to the wrong place inside the controller to work with a V 3 CA

 

[mrbill]

I find that I prefer using the minimum EBS Level on the Lyen controller (EBS Level = 0) as doing so allows just as much energy to flow into the battery but causes slightly less braking effect--I can use my friction brake if I need to decelerate harder--wasting less energy when I don't need to slow down as much. Though, to be honest I find it hard to feel the difference between EBS Levels 0, 1, and 2. I'm suspicious that changing this setting may have only placebo effect.

Are you using Lyen /Keywin software to set regen level or XPD?

Hi Neil:

I'm using Lyen's software.

Ideally I'd want a setting that allows for as much regen energy as possible to flow into the battery and not be wasted by heating the motor. I'd rather get my brake rotors hot than the motor hot. Something like the mild braking effect of "natural" regen would be ideal. "Natural regen" occurs at high speed--for me this is at speeds greater than 55 kph--such that the rectified voltage of the motor's back EMF exceeds the battery's voltage, thus allowing charge. When this occurs, I get much less braking effect for a given wattage than the "regen on demand" by closing the controller's e-brake circuit.

XPD allows a higher regen level

Also HVC voltage plays a major part. For 20 series with HoC voltage of 83 , I have my controller HVC set to 90-92 volt.
Regen never gives so much power to actually push pack voltage over HoC,but makes regen more effective
If you lived at the top of a massive hil and the first 1/2 mile of your ride was downwards umder max regen, then the OTT HVC maybe an issue, but otherwise, as soon as you have taken power from the pack, you'll never get regen to push pack voltage above a safe value.

I'm using 14s LiPo batteries, so the 55 volt Regen limit works well. I get no regen at full charge, but after discharging a small amount (about 3% typically), some regen is allowed. At first all energy seems to flow back into the battery as I feel almost no braking force, even up to the limit of 500 watts. But as the battery voltage falls further, I get a more pronounced braking force (and attendant heat build-up in the motor). Regen on demand seems to max out at about 500 watts (~10 Amps) flowing back into the battery, yet given the heat build-up in the motor it's clear that more energy ends up heating the motor.

Does the higher regen level available in XPD allow one to push more current into the battery, or does it just increase the braking force (after limiting the charge current)? I'm interested in experimenting with XPD if the former is possible.

 

[NeilP]

I'm using 14s LiPo batteries, so the 55 volt Regen limit works well. I get no regen at full charge, but after discharging a small amount (about 3% typically), some regen is allowed. At first all energy seems to flow back into the battery as I feel almost no braking force, even up to the limit of 500 watts. But as the battery voltage falls further, I get a more pronounced braking force (and attendant heat build-up in the motor). Regen on demand seems to max out at about 500 watts (~10 Amps) flowing back into the battery, yet given the heat build-up in the motor it's clear that more energy ends up heating the motor.

Does the higher regen level available in XPD allow one to push more current into the battery, or does it just increase the braking force (after limiting the charge current)? I'm interested in experimenting with XPD if the former is possible.

I have not really monitored it that closely . but by settign the HVC settign much higer than battery pack voltage, then you do get full regen even Hot of Charge. , but because the regen time is so short you dotn get pack voltage rising above what you woudl want. As i sadi earlier, i guess if you lived at the top of a very long hill, and went down the hill with regen on, with a freshly chared battery, then by the time you reached the bottom you may over volt the pack. but as we all know, the amount of pwer actually put back in , compared to what you take out with even one or two mild throttle openings,m I reckon it woudl take a very logn hill to put in enough power to over charge the pack.

As to eh question of does it push in more current or increase the brakign force,... umm dont know. I always assumed that the greater the current the greater the brakign force. I was under the impression that the Regen levels were merely regen current limiting. SO a low regen level, was only allowing a smaller regen current to flow, where as XPD 'Unlimited' regen, jsut allows max current that the motor tries to push back in.

I have no way of checkign this out in the next few days, as I am jsut to busy to do any trials.

 

[mrbill]

As to eh question of does it push in more current or increase the brakign force,... umm dont know. I always assumed that the greater the current the greater the brakign force. I was under the impression that the Regen levels were merely regen current limiting. SO a low regen level, was only allowing a smaller regen current to flow, where as XPD 'Unlimited' regen, jsut allows max current that the motor tries to push back in.

I have no way of checkign this out in the next few days, as I am jsut to busy to do any trials.

Hi Neil:

I can confirm that "EBS Level" 0, 1, or 2 (the three settings available in Lyen's software) make no difference in the maximum regen current. All three settings max out at just under 500 watts continuous, with occasional higher peaks, under similar conditions. EBS Level 2 has maybe a tad more braking force than Level 0, but that could be merely the power of suggestion. I haven't done any controlled test to assess the difference in braking force, and I'm suspicious that this setting has no effect at all.

 

[NeilP]

When I get a chance I 'll try and see , but it is not going to be tip the end of the month...jsut got too much on..so I may well forget..you'd be quicker getting XPD yourself and setting 'unlimited'

 

[justin_le]

But the wiring of the CA to the controller maybe the issue. It depends on how old the controller is, and how the CA plug is wired.
The green wire on the CA plug may go to the wrong place inside the controller to work with a V 3 CA

Neil is correct about this. If it's an old controller designed just for CAV2 operation, then it might have a diode wired up inline with the green throttle signal on the CA plug. It'll be very easy to figure out if this is the case, even without hooking up the throttle to the CA3 as it should be. Just pick the wheel off the ground, then go into the setup menu and set the throttle input mode to "OFF (WOT)", and once you leave the setup menu the motor should kick up and spin at full throttle. If it does, then great. If not, then almost for sure there is a diode blocking the ability to drive the throttle signal high via the 6-pin plug. It's a trivial thing to fix, either by opening the controller and removing the diode at the end of the green wire, or by leaving the controller closed and swapping with your actual throttle signal wire.

Teklektik did a great job explaining the options on Page 15 of the unofficial CA3 user manual:

 

[mrbill]

When I get a chance I 'll try and see , but it is not going to be tip the end of the month...just got too much on..so I may well forget..you'd be quicker getting XPD yourself and setting 'unlimited'

I did just that.

With XPD programming EBS Level to "Unlimited", I get no higher maximum charging current and no more or less braking force, as far as I can tell. I am pretty sure now that the regen settings are hard-wired in the controller and are not adjustable in spite of the EBS Level parameter. I have set my controller to EBS Level of "0" or "Disabled" in XPD in the hopes that this minimizes braking force for the given reverse current, but I still get a maximum of about -10 Amps (with some over/under shoot) and the same moderate braking force.

Getting back on the thread topic... With Prelim9 release I still occasionally get an unbidden display of Aux Level and less often persistence of characters from the Aux Level screen continuing to show in blank areas of the underlying display, especially following the use of regen braking.

And, I had the opportunity to observe regen braking in excess of -1000 watts. When a negative four-digit wattage is displayed, the negative sign is not shown, presumably because it would collide with the icons in the lower left corner. This gave me a bit brief episode of cognitive dissonance as I could feel a regen braking force, yet the CAV3 showed over 1200 watts being drawn from the battery. In this case I'd prefer to see the "-" sign, even if it collides with the icons.

 

[NeilP]

They must have changed something within the controller type you have .what model is it?

If I set mine to disabled I am 100 % sure it just does that, disables it.

Again I'll have to double check. but still not been ot the workshop to do bike stuff since the last post.

Between doing jobs for other people (video editing ) and a bloody mac with a totally dodgy and unstable operating system, I am spending every spar moment in front of the computer.

Any one who dares say Mac are better and more reliable than Windows needs taking outside and given a damn good kicking! I have never had as many issues with computers since I moved to Mac. bloody over hyped, over priced and no better than Windows

 

[mrbill]

They must have changed something within the controller type you have .what model is it?

If I set mine to disabled I am 100 % sure it just does that, disables it.

My controller is a ca. 2013 Lyen MKII 12FET controller using IRFB3077 FETs. Although I didn't take a photo of the board the last time I had it open (and I don't want to open it and risk breaking the wires to the power switch _again_ just to check), I believe it is an EB-312-... board.

With this controller I have had trouble with the regen braking being intermittent: sometimes it works when I close the e-brake circuit, sometimes it doesn't. And, if I close the e-brake circuit and get no regen, I can often get regen if I blip the throttle at the same time. Perhaps it really should be disabled, but a reference voltage is resting near a threshold, leading to intermittent functionality. Since the EBS Level setting doesn't affect the actual braking force (something I was hoping it would do), I will set it to the lowest non-disabled level, level 1, I think, and see if this solves the intermittency.

Between doing jobs for other people (video editing ) and a bloody mac with a totally dodgy and unstable operating system, I am spending every spar moment in front of the computer.

Any one who dares say Mac are better and more reliable than Windows needs taking outside and given a damn good kicking! I have never had as many issues with computers since I moved to Mac. bloody over hyped, over priced and no better than Windows

I agree about the over-priced and perhaps the over-hyped. But before you throw it out the window you might consider adding RAM if your machine is short and if your machine allows--Apple deserves a kick in the backside for hard-wiring RAM into their latest Mac mini and offering no user-upgrade possibility. Since RAM is cheaper than a new CPU box, I'd install as much memory as you can.

I recently helped some users set up an old late-2009 Mac mini with stock 4GB RAM after upgrading OS X to "Yosemite" (the latest version) that has a minimum requirement of only 2GB RAM. Everything was slow as molasses and almost unusable even for basic web browsing and email. Users were not happy campers. Instead of throwing in the towel and buying a new computer (600 USD minimum) I took a gamble and recommended they upgrade to 8GB RAM for about 90 USD. It's like a new computer now. Everything runs briskly and reliably.

I have a Mac desktop and a windows laptop. I prefer the Mac for everyday use, but I bought the latter for travel and so that I could run only-available-for-Windows e-bike programming tools at my e-bike.

 

[NeilP]

Sounds something iffy going on on the controller front..but some day I will confirm if mine disables or not.


As for mac RAM, it is a Mac Pro Tower 2008 3,1 all RAM slots are full, 28GB, Max it will take is 32 from 4GB sticks, but some talk now is saying that it will take 64 GB using 8GB sticks, but not got that sort of money to throw at it.

Just run memtest for 36 hours, three loops , all good.

I think the problem has been taking advice from others..Mac worshippers who believe you can just upgrade and trust it all to Apple. I should have been doing clean installs each time I upgraded the OS, and not trusted to the useless migration assistant that migrates everything , dodgy kexts and plists included. Going to have to do a clean install to sort this. Used EtreCheck the other day and many many failed or non-loading lets and entries in the launch Daemon and Agent Folders. Comes from believing the bullshit that you can just delete an app by dragging the app itself to the Trash Can.

Oh BTW..how did you know I had thrown computers out the top floor window before...last one was a laptop that I folded backwards over my knee, ripped it in half, then smashed its still warm processor off the main board ....great fun, made be feel a lot better

My best running machine is a HP4530s laptop, with OS X Yosemite installed with Clover Boot loader. see tonymacx86 for more details

 

[mrbill]

Sounds something iffy going on on the controller front..but some day I will confirm if mine disables or not.

I would like to know if others still get sporadic regen with their Lyen controllers when EBS Level is set to "Disabled". Meanwhile I've set mine to "Moderate" and will test in the next few days to see if that consistently gives me regen when I close the controller's e-brake circuit.

As for mac RAM, it is a Mac Pro Tower 2008 3,1 all RAM slots are full, 28GB, Max it will take is 32 from 4GB sticks, but some talk now is saying that it will take 64 GB using 8GB sticks, but not got that sort of money to throw at it.

Just run memtest for 36 hours, three loops , all good.

I think the problem has been taking advice from others..Mac worshippers who believe you can just upgrade and trust it all to Apple. I should have been doing clean installs each time I upgraded the OS, and not trusted to the useless migration assistant that migrates everything , dodgy kexts and plists included. Going to have to do a clean install to sort this. Used EtreCheck the other day and many many failed or non-loading lets and entries in the launch Daemon and Agent Folders. Comes from believing the bullshit that you can just delete an app by dragging the app itself to the Trash Can.

Oh BTW..how did you know I had thrown computers out the top floor window before...last one was a laptop that I folded backwards over my knee, ripped it in half, then smashed its still warm processor off the main board ....great fun, made be feel a lot better

My best running machine is a HP4530s laptop, with OS X Yosemite installed with Clover Boot loader. see tonymacx86 for more details

A 2008 Mac Pro is getting long in the tooth, but you've certainly got enough RAM that that shouldn't be an issue unless you're loading it up with (or trying to launch) too many memory-hogging and/or CPU-hogging background processes. Although it's inconvenient, a clean install from scratch gives you a known end state.

 

[NeilP]

A 2008 Mac Pro is getting long in the tooth, but you've certainly got enough RAM that that shouldn't be an issue unless you're loading it up with (or trying to launch) too many memory-hogging and/or CPU-hogging background processes. Although it's inconvenient, a clean install from scratch gives you a known end state.

That is something I avoid at all cost. always try to keep a clean system..then you go and buy stuff like the X-Rite ColorMunki screen calibrator or a Wacom tablet, and their damn software insists on background apps running ..that are almost impossible to get rid of. Even using an un-installer supplied by the ColorMunki manufacturer, still leaves background shit running g. just crap coding .

 

[justin_le]

Getting back on the thread topic... With Prelim9 release I still occasionally get an unbidden display of Aux Level and less often persistence of characters from the Aux Level screen continuing to show in blank areas of the underlying display, especially following the use of regen braking.

Hi Bill, do you recall on what screens specifically you were on when you say characters from the Aux Change linger on the 2nd line? This will be easy to fix.

And, I had the opportunity to observe regen braking in excess of -1000 watts. When a negative four-digit wattage is displayed, the negative sign is not shown, presumably because it would collide with the icons in the lower left corner.

That's correct. The plan is to eventually fit in a small negative symbol bundled in with the PAS icon since that doesn't use the right most 3 pixels, but there ends up being a fair bit of custom character hacking to pull that off, so yeah now if you have more than -1000 watts it looks misleading. I'm impressed that you are actually getting >1000 watts of regen though, it's pretty rare to see that much getting dumped back into the battery pack. I'm presuming this was at fairly high speed travel?

This gave me a bit brief episode of cognitive dissonance as I could feel a regen braking force, yet the CAV3 showed over 1200 watts being drawn from the battery. In this case I'd prefer to see the "-" sign, even if it collides with the icons.

If I can't super-impose the '-' sign with the PAS symbol, then the easier solution would be to replace the PAS icon altogether with the negative sign instead.

 

[mrbill]

Getting back on the thread topic... With Prelim9 release I still occasionally get an unbidden display of Aux Level and less often persistence of characters from the Aux Level screen continuing to show in blank areas of the underlying display, especially following the use of regen braking.

Hi Bill, do you recall on what screens specifically you were on when you say characters from the Aux Change linger on the 2nd line? This will be easy to fix.

The underlying screens where I observed persistent AuxLevel characters were the Regen% and Fwd Ah/Regen Ah screen and the Watt-Hours and wh/mi screen.

And, I had the opportunity to observe regen braking in excess of -1000 watts. When a negative four-digit wattage is displayed, the negative sign is not shown, presumably because it would collide with the icons in the lower left corner.

That's correct. The plan is to eventually fit in a small negative symbol bundled in with the PAS icon since that doesn't use the right most 3 pixels, but there ends up being a fair bit of custom character hacking to pull that off, so yeah now if you have more than -1000 watts it looks misleading. I'm impressed that you are actually getting >1000 watts of regen though, it's pretty rare to see that much getting dumped back into the battery pack. I'm presuming this was at fairly high speed travel?

I was moving about 65-70 kph down a straight hill with a nice tailwind while intermittently closing the controller's e-brake circuit to get regen braking. The peaks were around -1200 to -1300 watts, although power to the battery declined after a few seconds to around -500 watts. By closing the controller's e-brake circuit intermittently I could effectively get the controller to send more power back into the battery. Although regen power varied, braking force remained constant during regen.

This gave me a bit brief episode of cognitive dissonance as I could feel a regen braking force, yet the CAV3 showed over 1200 watts being drawn from the battery. In this case I'd prefer to see the "-" sign, even if it collides with the icons.

If I can't super-impose the '-' sign with the PAS symbol, then the easier solution would be to replace the PAS icon altogether with the negative sign instead.

That would be fine with me. Since I stopped using a torque-reading bottom bracket the PAS symbol merely serves to let me know when I'm pedaling.

 

[Green Machine]

Let me say one of my pet peeve about the new v3 cycle analyst.


IS it not possible that the v3 cycle analyst could have integrated an on and off switch for the electric bike...Is this possible?

Why was this feature left out? Isnt it easy to do with the stand alone at least? Is it possible with the DP?

Is it possible to add it later with a firmware update?

On and off switch on an ebike is for obvious reasons an important safety feature....not to mention the convenience factor.

I think that DIY ebikes do not have an easy soluiion for an ebike on and off switch really elaborates how backwards we are in the technology game.... i guess just waiting for china to provide one. The problem is most DIY controllers (lyen, cellman) dont have an on and off switch or an easy way to connect one...same with DIY battery solutions. i think this is an area where the cycle analyst should have stepped up and provided an essiential feature which the other DIY companies have neglected.

why did a temperature probe come before an on and off switch?

 

[mfinca]

Grin Technology's controllers come with integrated on off switches. I back that up with a regular car 12 volt battery switch, which is keyed. I never switch it under load and so far it has held up. The better solution is a Circuit Breaker like this http://i.ebayimg.com/00/s/NTAwWDUwMA==/z/UeQAAOSwfcVUDeHa/$_57.JPG

It handles 50 amps and 80 volts at under $20. (Original tip on breaker from teklektik)