izeman
1 GW
Not? So what are you calling those? That comment ain't make any sense to meJohn in CR said:
Are you saying that they are too weak for Ebikes?
Cycle Analyst V3 preview and first beta release
- Thread starter justin_le
- Start date
Are you saying that they are too weak for Ebikes?
izeman
1 GW
As a last consequence that would mean that doing wheelies with any bike would lead to bad consequences for those bikes. I think it should be up to the users responsibility to use any features appropriately.2handy said:
I don't think that there will be any impact from a few hundred Ebikes using a CA and from those few hundred there will be some percent that may wheelie.
izeman
1 GW
I guess it could even be used to learn how to wheelie a regular bike (once you know how it feels to ride in this weird position). A boy's dream would come true - can't remember how many trousers I ruined to try it, and how many bleeding knees.Kepler said:The wheelie control would be so cool. I could pretend I was a skilled rider then.
One significant change that Justin mentioned only indirectly in his review of new features is a simplification of the PAS Modes. He has organized functionality into just three modes and the operation seems (to me) much easier to grasp.
Here's a snippet from a draft of the revised Guide:
Here's a snippet from a draft of the revised Guide:
izeman said:
We can agree to disagree.
Whatever the number of the CA3 that are sold, that is only the tip of the iceberg. As ebikes become more popular & the versions of the CA continue to be developed & bought in greater numbers;- if the CA allows those purchasers/riders to do things that draw attention to them & their ebikes, I think that this will very likely end up having an effect. Though trying to predict the future is always a dodgy proposition.
As far as your "any bike" statement is concerned, those kinds of bikes are probably legally road registered & not ebikes that are so often sneaking around the sides of the laws of the land that they are being ridden in.
Eventually we will be faced with inescapable technological scrutiny from the police force, there is no two ways to that.
My post was all about trying to put that time off for as long as possible. Relying on the self control of ebike riders is an uncontrollable variable, if we can do something, why won't we do it.
rp3
100 W
The fun police are here! 
:lol:
How bout we get back to talking about this awesome little device. :wink:
:lol: How bout we get back to talking about this awesome little device. :wink:
pendragon8000
100 kW
This is why I love tha ca and its aux switch power mode. i set it for sensable pwer for comuting and then high power mode for wheelies in the park etc.2handy said:. Relying on the self control of ebike riders is an uncontrollable variable, if we can do something, why won't we do it.
also, consider most ppl can speed in their cars and all those kids with illegal hotted up cars...
Kepler
10 MW
I can see it now, ebikes all over town wheelieing down the road bringing the hobby into disrepute. :lol:
In fact lets set the the current limit on the CA3 so it cant be raised above 250W just to make sure we dont draw attention to ourselves
Seriously though, the way I see it, the CA3 has been designed for the ebike enthusiast in mind. Its feature rich and capable of far more tweaking then most people will ever need. I seriously doubt adding wheelie stability would ever have an impact on the mainstream ebiking community however for the enthusiast, it would be a blast.
BTW, the last time I tried doing a wheelie, I managed go over the handlebars (dont ask
) and crack two ribs. I seriously do need help :lol:
In fact lets set the the current limit on the CA3 so it cant be raised above 250W just to make sure we dont draw attention to ourselves
Seriously though, the way I see it, the CA3 has been designed for the ebike enthusiast in mind. Its feature rich and capable of far more tweaking then most people will ever need. I seriously doubt adding wheelie stability would ever have an impact on the mainstream ebiking community however for the enthusiast, it would be a blast.
BTW, the last time I tried doing a wheelie, I managed go over the handlebars (dont ask
) and crack two ribs. I seriously do need help :lol: 
rp3 said:The fun police are here!
How bout we get back to talking about this awesome little device. :wink:
On that subject, attached is the latest Version3 Prelim5 software. This has not been rigorously tested by Teklektik yet (he had the Prelim4) but I think I've been reasonably methodical in making sure not to break anything since the Prelim4 I sent him.
The main addition to the features/tweaks discussed here: http://endless-sphere.com/forums/viewtopic.php?p=728436#p728436 is the way that the torque PAS mode is handled. The combination of averaging over two pedal rotations rather than 1, and also to scaling on human watts rather than pedal torque, resulted in a much longer lag time than the earlier code for the torque PAS mode to kick in while pedaling from a start.
So in the prelim5 firmware, it initially starts averaging over the last full rotation until your pedal cadence gets up to speed, and then it switches over to averaging 2 turns for smoother behavior. As well, if your pedal cadence is less than 55 RPM, then it does the scaling based on your torque rather than your watts. When you are starting off in a high gear for instance, even though you can be cranking hard on the pedals your human watts is remains low because the pedal RPM is low, and yet presumably you would want a lot of assistance in this situation. So when the cranks are moving slow the boost factor will be more than you'd get from a scaling of human wattage. Once the pedal cadence is up to 55 RPM, then the CA will scale your actual human watts to determine the electrical watts to the motor.
I hope that gets the best of all worlds, enabling good off-the-line behavior with the torque PAS mode, while still behaving smoothly at speed, and also having a more intuitive setup of dealing with a scale factor rather than command Watts/Nm.
Accelerometer/Inclinometer, I'm afraid, is a CA V4 project...
-Justin
Kepler
10 MW
Looking forward to testing, especially the new torque PAS settings. Thanks Justin.
Printable Setup Summary for CA v3.0 Prelim5
The setting summary for the newer v3.0 Prelim6 release is available here.
The setting summary for the previous CA v3.0 Prelim1 (aka v3B23P1) release is available here.
Unofficial User Guide is available here.
Please see the Grin Tech Site for a detailed explanation of Setup Parameters.
Distinguishes global from preset-specific parameters.
Printable versions for saving settings - tabular format for multiple presets - formats: 1UP.PDF, 2UP.PDF, XLSView attachment CA_V3-0Prelim5_ConfigSettings.zip
The setting summary for the newer v3.0 Prelim6 release is available here.
The setting summary for the previous CA v3.0 Prelim1 (aka v3B23P1) release is available here.
Unofficial User Guide is available here.
Please see the Grin Tech Site for a detailed explanation of Setup Parameters.
Distinguishes global from preset-specific parameters.
Printable versions for saving settings - tabular format for multiple presets - formats: 1UP.PDF, 2UP.PDF, XLSView attachment CA_V3-0Prelim5_ConfigSettings.zip
It will be just a bit before the updated Guide is released, but for the PAS folks, here's an excerpt with some details to help get the Start/Stop RPM settings squared away. This should be the last noteworthy bit of information about the v3.0P5 release missing from the existing Guide and not already discussed in this thread.
John in CR
100 TW
izeman said:Not? So what are you calling those? That comment ain't make any sense to meJohn in CR said:
Just poor wording on my part. I was trying to say those in your signature look easy enough to lift the front to me, though now that I see the full batt pack in front on the latest, it looks easy enough to keep down too.
MattyCiii said:
I forgot to mention, the CA3 prelim5 firmware I uploaded a bit earlier increase this slow up rate limit to 9.00 sec/volt, and on the fast end it is now clamped at 0.03 sec/volt, rather than 0.05 sec/volt. So hopefully that extra bit on either end makes the difference in your and other setups needing either extreme.
Internally, this setting is stored inside the CA as Volts/Sec, rather than Sec/Volt, and it would be easier and less glitchy to display it as a rate (V/sec) instead of a time period (sec/volt). As an end user, do you find one more obvious or less confusing than the other?
-Justin
pendragon8000
100 kW
Personally from a programmer perspective I would have done the same. However taking into account the type of end user, mostly not being technologicaly uneducated. I think making it more stable would be preferable and with the comprehensive setup notes you guys have created there shouldn't be an issue. IMHOjustin_le said:
MattyCiii
100 kW
justin_le said:
THANKS Justin!
I noticed the expanded range, but have been sick and unable to test it
I would suggest not converting units just for the end user - use volts/sec both inside and out. As Pendragon stated, a few words in the setup guide will make things clear to all users. For those who don't understand, the simple note of "lower value == slower ramping" will make things clear enough to get people safely on the road.
John in CR
100 TW
Volts/sec is fine, but please don't talk in sec/volt like in your previous post. Consistency and stability are key.
Ignore Pendragon's comment about the end users. Things should be "dumbed down", just not to the preschool level of Apple products.
Ignore Pendragon's comment about the end users. Things should be "dumbed down", just not to the preschool level of Apple products.
Kepler
10 MW
teklektik said:
Did quite a few hours of testing today. I am particularly impressed with the Auto PAS function. Makes it completely viable to install a simple RPM PAS on a high powered setup. I have this set this up on my Fighter. In combination with an adjustable pot or 3 speed switch, the functionality is just spot on. On road I can set the amount of assist I need which is typically only a few hundred Watts but still have my full 3000W available via the throttle. Even makes it viable to use the PAS function off road. I would go as far to say that is my favorite feature on the CA3.
A suggestion to anyone who wants to install an RPM PAS with the CA3. Get the one that Justin sells. I have tried a couple of the cheap $5.00 ones. They work but Justin's unit works significantly better with the CA3.
However, I still am yet to get the torque PAS setup on my Super Commuter to my liking. Ironically I had it working perfectly on the previous version of firmware. The problem I have is that I cant eliminate the power dropouts while pedalling. Just going back over all the notes before declaring a bug with this post covering the problem I am trying to solve.
I dont quite understand the note and would appreciate a litte further explanation. The power dropout is particularly frustrating as it takes away your momentum especially when climbing a hill.
pendragon8000
100 kW
OK so that's a definate v/s and s/vJohn in CR said:
Edit:
Err vote (shrug)
Basically, the V3 is looking at PAS interrupts (which come fairly infrequently) and must decide fairly quickly (less than a revolution) whether you have started to pedal or are stopping. The RPM specification is used to determine the time delays between magnet pulses to detect these start/stop events and the table above yields (start,stop) pulse delay thresholds of about (0.75,0.50) seconds respectively. If pulses arrive consistently between these times, the V3 will alternately turn the assist on and off (dropouts). So - the idea is to make this range as small as possible so you transition through it fairly quickly when you start and stop pedaling.Kepler said:
One thing Justin has mentioned that you might try is to start with the start/stop RPM set the same and adjust if necessary from there. Setting (start,stop) with the reverse relationship should also prevent dropouts but this may not be a desirable setting.
Kepler said:
Actually we don't have any left to sell at this point (more ~ 2-3 weeks away). However, there as been a trend towards higher and higher pole count PAS sensors from most suppliers, and it should be that any PAS sensor with like 10-12 magnets will have a snappy and nice response, while those with 5 or 6 magnets will seem a bit laggy. So if you are shopping for a suitable PAS device, more poles = better. Otherwise they're all pretty much the same so it shouldn't matter where you get it from.
Can you tell me what your start and stop RPM thresholds are and whether you see the human power bar drop off when the power cuts? With the 8 pole sensor on the THUN BB, you can set both the start and stop thresholds to like 12RPM or therabouts. If you disable the quadrature mode, the THUN BB will still work to detect pedaling but it won't be as sensitive to stopping from slight reverse pedal rotation. I've had now a week of riding around with the torq PAS mode and haven't had any periodic cutout issues. However, like you I've also found that the AutoPAS mode with a 3 speed switch is almost as good as proportional torque assist for the way I like to ride.
-Justin
Kepler
10 MW
justin_le said:
The fact that the one you sell wont give an output if turning the the pedals backwards is a nice bonus too even though it doesn't have a direction output. I also noticed that it looks like it has a voltage reg installed in the plug I presume so that you can plug it into the same plug us the thun which jas a 10V output, and still have it work correctly.
justin_le said:
Tried all sorts of settings from the default to equal 10 and 10, to really low figures. Most tests were done with the start number being number being higher the stop number. With this new information, I will continue testing today. I need to have a closer look at the Human power bar drops off again but I think it does.
Similar threads
- Question
