Cycle Analyst V3 preview and first beta release

Get real world experience and user feedback on the electric bicycle products.
User avatar
NeilP   10 GW

10 GW
Posts: 4956
Joined: Nov 27 2010 3:18pm
Location: 49.17303, -2.05354 Jersey, Europe
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by NeilP » Dec 30 2013 3:44am

Either Speed option ( hall or wheel magnet) still has same problem of inaccurate speed reading due tyre deformation.

If you need to worry about accuracy of speed that much, GPS or better still a Doppler radar speed detector as fitted to some modern farm tractors
:) Only Irish coffee provides in a glass all four essential food groups: alcohol, caffeine, sugar, and fat :)

User avatar
teklektik   10 GW

10 GW
Posts: 4020
Joined: Mar 26 2011 1:15pm
Location: CT, USA

Re: Cycle Analyst V3 preview and first beta release

Post by teklektik » Dec 30 2013 5:26am

RC_guy wrote:But when I bought this one, V3 was not available and everyone told me it would be pretty much plug and play to add a CA later :wink:
The V2 is plug and play with that older controller. Your V3 would work fine like a V2 as a plug and play with no wiring changes... it just wouldn't have the added V3 features. This is described in the Guide. The new controllers are plug and play for a V2 or a V3.

The resistor and changes to the V3 itself are because Paul likes to use the LM35 temp sensor instead of one of the types that the V3 actually supports. From his perspective it is a better choice so folks with just a multi-meter can read the temperature directly, but it's a bit of a PITA to make the V3 use it. Frankly, when my MAC was delivered, the motor was open and not yet assembled to the wheel, so I just snipped the leads on the LM35 and stuck in a thermistor so the CA would require no changes.
RC_guy wrote:Wait a second here, I bought the DPS version only because they told me that is what I needed.
Now I am thinking that it would make a simpler installation (less wires) if I used the halls for my speed reading. Would I need to modify much?

But frankly I never liked those magnets that spin in wheels and then you either loose them, a branch catches the wire or there are many other good reasons to try to not have a speedometer cable
You have a MAC - a gear motor. Unlike a direct drive motor, the rotation of the motor is not tied to the rotation of the wheel due to the internal freewheel. You must use a DPS model with wheel pickup because the hall signal is not guaranteed to be related to speed. For instance, if you close the throttle and roll down a hill, there are no hall pulses at all....

If you have real problems with boonie bashing and a front wheel pickup - move it to the rear wheel.

If you don't want to use the yellow CA-DP wire to carry the temp signal, you can route another wire forward and use the existing external temp sensor plug.
RC_guy wrote:Can it be a regular resistor? The manual does not mention a pull down?
Use any resistor that will physically do the job - the familiar thru-hole type is good. The wattage is miniscule so 1/4W, 1/8W, etc - is fine - whatever is available.
A resistor that ties a signal to Gnd is a pull down... don't worry about the term.
Visit Grin Technologies at www.ebikes.ca
Build Thread: 2WD Yuba Mundo V4

User avatar
100volts+   10 kW

10 kW
Posts: 722
Joined: Nov 20 2012 10:05pm
Location: Houston

Re: Cycle Analyst V3 preview and first beta release

Post by 100volts+ » Dec 30 2013 5:50am

subscribed
http://endless-sphere.com/forums/viewto ... =1&t=57440

Unlike puppets we have the possibility of stopping in our movements,
looking up and perceiving the machinery by which we have been moved. In this
act lies the first steps towards freedom. - PETER BERGER Sociologist

RC_guy   10 W

10 W
Posts: 71
Joined: Nov 14 2011 12:31pm
Location: Montreal

Re: Cycle Analyst V3 preview and first beta release

Post by RC_guy » Dec 30 2013 1:40pm

Gees. I feel like such a dork about wanting to pick up the hall wire signals on my motor.
I wanted to open it anyway and if I need to replace anything I already have a spare clutch and gears assembly.
I will change the thermistor inside the motor.
So if I just get a 10K NTC thermistor?

Thanks

User avatar
hjns   100 kW

100 kW
Posts: 1273
Joined: Aug 20 2011 11:05pm
Location: Basel, Switzerland

Re: Cycle Analyst V3 preview and first beta release

Post by hjns » Dec 30 2013 2:19pm

Why do you want to open your motor? Only to put a thermistor inside even though it has one already?

If you are going to open it, better prepare to change the phase and hall wires as well. At least that will make it worth all the hassle it takes to open the darn thing.
Henk


All men dream: but not equally. Those who dream by night in the dusty recesses of their minds wake in the day to find that it was vanity: but the dreamers of the day are dangerous men, for they may act their dreams with open eyes, to make it possible. T.E. Lawrence

RC_guy   10 W

10 W
Posts: 71
Joined: Nov 14 2011 12:31pm
Location: Montreal

Re: Cycle Analyst V3 preview and first beta release

Post by RC_guy » Dec 30 2013 3:53pm

Just because my ride to work is 26 kms and I am at WOT for 3/4 of the trip. My motor started whining a little more at WOT last summer and the outside casing was a bit hot sometimes. I just want to make sure everything is OK

Animalector   10 kW

10 kW
Posts: 810
Joined: Mar 17 2012 4:50pm
Location: Mareeba - Queensland

Re: Cycle Analyst V3 preview and first beta release

Post by Animalector » Dec 30 2013 4:52pm

I think what hjns meant was that it is likely more work to open the motor and replace everything than it is to do the resistor miss on the ca. And the outcome would be the same as what you have now have a read of the manual again I think it will be clearer what you need to do once you have the ca open and are looking at everything
Andy

RC_guy   10 W

10 W
Posts: 71
Joined: Nov 14 2011 12:31pm
Location: Montreal

Re: Cycle Analyst V3 preview and first beta release

Post by RC_guy » Dec 30 2013 9:40pm

Ya thanks guys I understand but I was going to open the motor for inspection anyway. It's -15 celcius here for the next few months so I have time.

User avatar
teklektik   10 GW

10 GW
Posts: 4020
Joined: Mar 26 2011 1:15pm
Location: CT, USA

Re: Cycle Analyst V3 preview and first beta release

Post by teklektik » Dec 31 2013 11:38am

RC_guy wrote:I wanted to open it anyway and if I need to replace anything I already have a spare clutch and gears assembly.
I will change the thermistor inside the motor.
So if I just get a 10K NTC thermistor?
Yes. Check the Temp Sensor section of the Guide - you need a device with a beta that lies in the acceptable range. There is another brief discussion of the V3 requirements in this post along with the effect of the beta spec.
RC_guy wrote:My motor started whining a little more at WOT last summer and the outside casing was a bit hot sometimes. I just want to make sure everything is OK
RC_guy wrote:...I was going to open the motor for inspection anyway. It's -15 celcius here for the next few months so I have time.
I have run MAC gears in my BMCs and in my experience they tend to whine a bit more than the BMC variety which are a bit thinner and have chamfered edges on the teeth, so what you are hearing is likely normal. It's no big deal to pop one of these motors open (there are some threads on doing it (search ES), but on the other hand, it's annoying enough that I don't do it lightly.

If you have the motor open, then swapping the sensor is easy, but I'm not necessarily recommending it as a policy to avoid the CA mod - let your situation and skill set be your guide... :)
  • It might be worth noting that EM3EV now sells a new stronger clutch assembly. It needs the new thinner gear set but the cost is very modest. If you are pulling the motor apart, you might like to consider swapping one in and stashing the present unit as a backup. The new unit looks like the BMC V4 clutch which I have found to be much tougher than the old style MAC or BMC clutches.
  • BTW - BMC recommends MobileGrease 28 for their motors and I have found it effective - a cartridge will last a heck of a lot of rebuilds...
Visit Grin Technologies at www.ebikes.ca
Build Thread: 2WD Yuba Mundo V4

User avatar
Arlo1   100 GW

100 GW
Posts: 8171
Joined: Apr 26 2009 10:36pm
Location: Nanaimo
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by Arlo1 » Jan 02 2014 1:51am

Im not using this version but it is the big CA and Im having problems keeping the current calibrated at 0 amps. I find it seems I have to turn zero it most times I turn it on. I used my dyno sensor to calibrate the shunt.... But should I Zero the amps then calibrate the shunt after? Im not get the right readings at the moment. Anyone have to deal with this?
My Leaf motor controller build. http://endless-sphere.com/forums/viewto ... 27#p963227
My YSR build http://www.youtube.com/watch?v=BRo8r5g4NBg
RC and most types of Lithium batteries you MUST know your individual cell voltages charging and discharging.
Don't keep them were you cant afford smoke or fire!
Never above 4.2v never below 2.7v EVER!!!
Support me on Patreon. https://www.patreon.com/user/posts?u=6842045
http://www.undergroundelectrics.ca/

User avatar
alsmith   100 kW

100 kW
Posts: 1182
Joined: Dec 03 2010 7:25pm
Location: Northumberland, UK

Re: Cycle Analyst V3 preview and first beta release

Post by alsmith » Jan 02 2014 3:59am

Arlo1 wrote:Im not using this version but it is the big CA and Im having problems keeping the current calibrated at 0 amps. I find it seems I have to turn zero it most times I turn it on. I used my dyno sensor to calibrate the shunt.... But should I Zero the amps then calibrate the shunt after? Im not get the right readings at the moment. Anyone have to deal with this?
This is exactly what my V2.2 CA was doing before it started showing regen at 999.9%. Still haven't sorted it but the thread is here.

User avatar
John Bozi   1 MW

1 MW
Posts: 1861
Joined: Apr 09 2013 1:46am
Location: Brisbane, Australia
Contact:

lvc

Post by John Bozi » Jan 02 2014 6:44pm

I started noticing power cuts to my ride for the first time, which were similar to the ones I have set on temperature.

Can anyone confirm that the cav3 indeed does cut down the power as you near the inputed lvc?

does it ramp down?

thanks, otherwise I have another problem....

thanks

User avatar
teklektik   10 GW

10 GW
Posts: 4020
Joined: Mar 26 2011 1:15pm
Location: CT, USA

Re: lvc

Post by teklektik » Jan 03 2014 10:02am

John Bozi wrote:I started noticing power cuts to my ride for the first time, which were similar to the ones I have set on temperature.

Can anyone confirm that the cav3 indeed does cut down the power as you near the inputed lvc?
does it ramp down?
This is an interesting question. I have had V3 LVC cutouts, but never really looked at the exact means of control - it looked like another PI controller because of the Gain setting, but it just never got snooped in depth... But, after a quick ping to Justin:
  • The LVC works similarly to other limit settings on the V3 (e.g. speed, power): when the limit threshold Batt->VltCutoff is crossed, the CA adjusts the throttle at a rate controlled by the Batt->LoVGain setting of the present battery preset to enable/disable power. This works in both directions so the gain setting controls the speed at which power is both rolled off to zero as well as re-applied. Large and small gain values cause faster and slower adjustment respectively.

    The topic is not presently addressed in the Guide - to be added on the next release. Good question - thanks!
In any case, you will see the "V" on the main display flash when LVC limiting is in effect.
John Bozi wrote:...otherwise I have another problem....
Please watch the Limit Flags on the Diagnostic Screen - the flags associated with the present limiting parameters will be asserted.

Otherwise, if this is happening on acceleration and you have a powerful bike, please review section 5.2 on Speed Limiting
Visit Grin Technologies at www.ebikes.ca
Build Thread: 2WD Yuba Mundo V4

User avatar
John Bozi   1 MW

1 MW
Posts: 1861
Joined: Apr 09 2013 1:46am
Location: Brisbane, Australia
Contact:

LVC

Post by John Bozi » Jan 03 2014 5:11pm

thanks teklektik,

It seems I have another problem... It only started after I repositioned my controller and all the wiring, so I think something is different there. I tested again this morning on full charged battery and I still got a jerky kind throttle, but some how it kind of went away for the remaining ride. Power doesn't cut off just seems to be dropping to very low amounts of power, I will have to study the screen more on my next ride, maybe its the motor?

I am still a bit confused about the lvc.

With the temperature settings, I have a thrash = when it starts limiting @ 90 degrees and a final cutoff @ 125 degrees. That gives as I have noticed pretty clear gradual ramping down, (i've never gotten past 115 degrees because it is so well limited).

With the lvc though I only see one input for low voltage cutoff. That as you are saying is when it starts to work, but how do you set when 100% of power gets cut?

At least you introduced me to the lovgain, but I am not sure what the default 800 actually means or does.

I am running 12s 1p at the moment. Basically I want the power to be limited to zero @ 39v and get there gradually from 43v.

User avatar
teklektik   10 GW

10 GW
Posts: 4020
Joined: Mar 26 2011 1:15pm
Location: CT, USA

Re: LVC

Post by teklektik » Jan 03 2014 6:45pm

John Bozi wrote: It seems I have another problem... It only started after I repositioned my controller and all the wiring, so I think something is different there. I tested again this morning on full charged battery and I still got a jerky kind throttle, but some how it kind of went away for the remaining ride. Power doesn't cut off just seems to be dropping to very low amounts of power, I will have to study the screen more on my next ride, maybe its the motor?

I am still a bit confused about the lvc.

With the temperature settings, I have a thrash = when it starts limiting @ 90 degrees and a final cutoff @ 125 degrees. That gives as I have noticed pretty clear gradual ramping down, (i've never gotten past 115 degrees because it is so well limited).

With the lvc though I only see one input for low voltage cutoff. That as you are saying is when it starts to work, but how do you set when 100% of power gets cut?
At least you introduced me to the lovgain, but I am not sure what the default 800 actually means or does.
I am running 12s 1p at the moment. Basically I want the power to be limited to zero @ 39v and get there gradually from 43v.
Okay - I probably didn't explain the LVC operation too well. Let me try again - maybe this will help.

Like the other limits (speed power, etc), and unlike temp which has special setup to define the roll-off range, the CA LVC is trying to eliminate the LVC limit situation by rolling back the power (throttle). The idea here is that due to resistance in the battery (and maybe a bit of wiring), less throttle will draw less current which will alleviate the voltage sag that arises from the current being applied to those resistances. In short, if you roll back the throttle, the voltage under load will rise, and the LVC condition may be removed.

So - if the CA can limit the throttle by a little bit and this makes the voltage sag recover and LVC alert go away, the CA will be happy. But if your LVC is set really low and the CA must throttle back a bunch to make the alert go away, it will do that. This is the same way the 'speed limit' works - it cranks down the throttle until the speed limit alert goes away.

The Gain value is just a dimensionless number that determines how rapidly the CA makes the correction (it's actually a multiplier for the integrated error, identical to IntSGain for Speed). So - if you pick a small number like Batt->LoVGain = 10, then the gain is low and it will take a while for the CA correction to take effect. If you pick something like Batt->LoVGain = 9999, the correction is almost instantaneous, and the LVC will cause immediate throttle rollback to ZERO.

There really is no way to set up the 39v-43v bracket that you wish (as with temp). You might try setting the limit to 40v and the gain to 50 and see if that more or less gives you the behavior you desire. Mostly this was designed to detect a single absolute limit and then try to hold the throttle to avoid exceeding it.

I'm not sure about the 're-positioning' issue. It may be related. Post back if more info comes to light......
Last edited by teklektik on Jan 04 2014 10:47am, edited 1 time in total.
Visit Grin Technologies at www.ebikes.ca
Build Thread: 2WD Yuba Mundo V4

User avatar
John Bozi   1 MW

1 MW
Posts: 1861
Joined: Apr 09 2013 1:46am
Location: Brisbane, Australia
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by John Bozi » Jan 04 2014 4:44am

if you pick a small number like Batt->VltCutoff = 10, then the gain is low and it will take a while for the CA to 'notice' and the correction to take effect. If you pick something like Batt->VltCutoff = 9999,
Just for clarity for others who might look for this info
I think you mean lovgain = 10 or 9999

not

Bat-> VltCutoff = 10v or 9999v


thanks again.

User avatar
teklektik   10 GW

10 GW
Posts: 4020
Joined: Mar 26 2011 1:15pm
Location: CT, USA

Re: Cycle Analyst V3 preview and first beta release

Post by teklektik » Jan 04 2014 10:48am

Oops! Exactly right. Good catch - thanks!
Fixed.
Visit Grin Technologies at www.ebikes.ca
Build Thread: 2WD Yuba Mundo V4

User avatar
teklektik   10 GW

10 GW
Posts: 4020
Joined: Mar 26 2011 1:15pm
Location: CT, USA

Re: Cycle Analyst V3 preview and first beta release

Post by teklektik » Jan 04 2014 1:05pm

The Errata document for the Unofficial User Guide (available here) has been re-posted with a section to provide an explanation of LVC operation. The new content is similar to the post above.

Once again - thanks for the posts with questions, problems, and resolutions... :D
Visit Grin Technologies at www.ebikes.ca
Build Thread: 2WD Yuba Mundo V4

User avatar
Merlin   100 kW

100 kW
Posts: 1482
Joined: Jan 29 2013 10:00am
Location: europe

Re: Cycle Analyst V3 preview and first beta release

Post by Merlin » Jan 05 2014 9:14pm

thx to you tekle for updating and helping fast as lightning :D

maybe you have also an idea for a little problem i have.

on full throttle and full speed (reached) i have really really short cutouts....
it feels like the throttle has no funktion for 0,1 second
they are so small that someone dont know the bike would never mind.

but i feel it...and i dont like it....

you can feel it only cruising at full speed.

to eliminate settings i tried without ca. just throttle direct to controller and a quick top speed cruise. no cutoffs...

any idea?

User avatar
teklektik   10 GW

10 GW
Posts: 4020
Joined: Mar 26 2011 1:15pm
Location: CT, USA

Re: Cycle Analyst V3 preview and first beta release

Post by teklektik » Jan 05 2014 11:26pm

Merlin wrote:on full throttle and full speed (reached) i have really really short cutouts....
it feels like the throttle has no funktion for 0,1 second
they are so small that someone dont know the bike would never mind.

but i feel it...and i dont like it....

you can feel it only cruising at full speed.

to eliminate settings i tried without ca. just throttle direct to controller and a quick top speed cruise. no cutoffs...
Interesting symptom - that would annoy me as well...
  • Please try watching the Limit Flags on the Diagnostic Screen and see if any go to upper case (limiting) when the cutouts occur. If your cutouts are brief, the flags may not show, but give it a try.
  • What throttle mode are you using? (PassThru? Current?, etc...)
  • Also, when you are at full throttle so the cutouts are ocurring, are you limiting current/power/speed or are the CA settings such that the controller can run unrestrained? That is - are you intentionally limiting one or more of these parameters?
Visit Grin Technologies at www.ebikes.ca
Build Thread: 2WD Yuba Mundo V4

User avatar
Merlin   100 kW

100 kW
Posts: 1482
Joined: Jan 29 2013 10:00am
Location: europe

Re: Cycle Analyst V3 preview and first beta release

Post by Merlin » Jan 06 2014 12:51am

It is my "offroad" mode (3)
no limits at all.

99A-9999watt
speedlimit 500
0.03 ramp time
Throttle is Pass Thru

I will give it another try and maybe film that screens.
As my Signature show......The "Problem" to watch, is the Speed...I have not much time to take a look at it :mrgreen:

User avatar
teklektik   10 GW

10 GW
Posts: 4020
Joined: Mar 26 2011 1:15pm
Location: CT, USA

Re: Cycle Analyst V3 preview and first beta release

Post by teklektik » Jan 06 2014 1:04pm

Merlin wrote:It is my "offroad" mode (3)
no limits at all.
The "Problem" to watch, is the Speed...
This is actually good - the CA should really be doing nothing with this configuraton - much less complicated.

If this was occurring during acceleration, I would suspect the DSGain setting since you clearly have a lot of power on tap. This is discussed in Section 5.2 'Speed Limiting' of the Guide and there is a correction in the Errata document. I'm thinking you probably already addressed that issue. In any case, since this happening at steady state once you have reached flat out top speed, the DSGain setting is not in play.
  • Are you using a wheel pickup (DPS model) or the controller hall signal (DP model) for your speed input?
  • If the wheel pickup, how many spoke magnets?
  • Please look at the displayed Maximum Speed (MaxS) after a ride when this problem occurs.
If you are getting contact bounce in the wheel pickup or some other noise on the Sp connection, then the CA might erroneously register a very high speed and give a brief power rollback due to exceeding your max speed limit. Contact bounce has been reported before with similar symptoms, although the cutout usually lasts longer than your reported 0.1sec. The MaxS reading will capture and preserve these bad speed readings even if they are too quick to see normally.

This is a puzzler....
Visit Grin Technologies at www.ebikes.ca
Build Thread: 2WD Yuba Mundo V4

User avatar
Merlin   100 kW

100 kW
Posts: 1482
Joined: Jan 29 2013 10:00am
Location: europe

Re: Cycle Analyst V3 preview and first beta release

Post by Merlin » Jan 07 2014 4:19am

hey,
it is a DP Model.
No Cutout at acceleration. It is only on (Top Speed)Cruising.

I have checked max speed after a ride. freakin 107kph. but it was korrekt speed. (I have Ulysee GPS on my Handy running.)

On these useless burning energy heat generating quick sprints the motor temp rises so quick that iam not sure that the overheat protection is the bad guy in this puzzle....
on a normal ride i dont feel incoming heat protection from the ca.
can this be a reason? the temp rises really quick on full throttle sprints. maybe the protection comes not so smooth in while between sensor checks from the ca is in 1 second
90°c and the next second it is 95°c

User avatar
izeman   10 GW

10 GW
Posts: 4742
Joined: Jun 21 2011 8:25am
Location: vienna, austria
Contact:

Re: Cycle Analyst V3 preview and first beta release

Post by izeman » Jan 07 2014 5:55am

you could disable temp cutoff and try again. if it's over 120c for a few seconds it would not harm the motor.

User avatar
teklektik   10 GW

10 GW
Posts: 4020
Joined: Mar 26 2011 1:15pm
Location: CT, USA

Re: Cycle Analyst V3 preview and first beta release

Post by teklektik » Jan 07 2014 10:03am

Merlin wrote:hey,
No Cutout at acceleration. It is only on (Top Speed)Cruising.

On these useless burning energy heat generating quick sprints the motor temp rises so quick that iam not sure that the overheat protection is the bad guy in this puzzle....
on a normal ride i dont feel incoming heat protection from the ca.
can this be a reason? the temp rises really quick on full throttle sprints. maybe the protection comes not so smooth in while between sensor checks from the ca is in 1 second
90°c and the next second it is 95°c
The CA is actually sampling much more rapidly - you are looking at the display update speed which is determined by Pref->Averaging. Lowering this value will speed up the rate of screen update. Even, so, that temp is rising fast.
izeman wrote:you could disable temp cutoff and try again. if it's over 120c for a few seconds it would not harm the motor.
+1 This would be a simple test.

You might also just raise Temp->MaxTemp and lower Temp->ThrshTemp. This would give an expanded range for power rollback. Remember that at MaxTemp, the CA has throttled back to 0% which is pretty extreme. With the default settings of 90/130degC, power is rolled back to 75% at 120degC. The issue is not the actual MaxTemp temperature set, but the temperature at which the motor and CA rollback reach equilibrium. So - settings of 85/140 would raise the 75% power rollback point slightly to 126degC but it would begin sooner and roll off more gently - perhaps reaching equilibrium at a lower temperature (e.g. at 85% throttle, etc)

It seems unlikely that the temp in the stator would drop rapidly enough to alleviate a power rollback in the short period you are seeing. I would expect a lot more latency. That said, the tests are easy and seem worth trying...
  • What temps are you seeing normally when cruising?
Hmmm - running out of ideas....
  • Does the Vmin reading look like your LVC might have kicked in?
Visit Grin Technologies at www.ebikes.ca
Build Thread: 2WD Yuba Mundo V4

Post Reply