Cycle Analyst V3 preview and first beta release

[justin_le]

Holy crap now i did it
When i was checking for the 5v i on the lcd header i slipped and shorted something out.. Now its completely dead

Ha whoops, probably best at this point to just send an email to info@ebikes.ca and request an RMA to send it back to be repaired at the shop.

[justin_le]

Thanks again to the rigorous methods of Teklektik we've got one more firmware ready for the users here. This should should should be the last version that carries that "prelim" tag on it and will be our official 3.00. The full firmware bundle is available from our downloads site:
http://www.ebikes.ca/downloads/CA3-0p11.zip

And I've also attached the no-eeprom hex file and release notes on this post too. The release notes go over in some detail all the changes so have a look at that. Majority of things are very small refinement tweaks that most users would never have noticed. However there are two updates to the setup menu which should be well appreciated by those who play around with their settings frequently.

First off is that there is now a LONG LEFT PRESS function throughout the menu structure. If you are editing digits in a field, you can press and hold the left button to save and go back to the previous digit rather than always moving to the next digit. And if you are navigating menu items, you can press and hold the left button to exit the setup menu altogether, in most cases that will be a faster than scrolling all the way left or right in order to reach the end of the menu structure:

Hold Left to Exit.jpg (52.45 KiB) Viewed 3285 times

Secondly, the left/right arrow indicators which illustrate where you are at in the menu list have been updated too, and now have an indicator so that you can tell if the parameter is global or is custom settable for each mode preset. If you see dashed lines like this, then you know that limit or option is being changed just on this particular preset.

Per Preset Dots.jpg (63.19 KiB) Viewed 3285 times

So if you haven't updated your CA3 firmware in a while (lots and lots of pple running the P6 code), now for sure would be a good time!

[r3volved]

Oh man, that left button is something I've had dreams about! Thank you

[NeilP]

Oh man, that left button is something I've had dreams about! Thank you

What he said !

[mchlpeel]

Can anyone help me.

I am looking to use the DC power jack from the Cycle analyst v3 for 3 things at once.

Can I use a couple of these splitters http://www.ebikes.ca/shop/ebike-parts/w ... lit90.html

to power these devices

Cycle Lumenator MAX DRAW 10W
http://www.ebikes.ca/shop/ebike-parts/l ... en-mb.html

Rear Light MAX DRAW 0.6W
http://www.ebikes.ca/shop/ebike-parts/l ... b-jk.html#

USB for Mobius Camera MAX DRAW 2.5W
http://www.ebikes.ca/shop/ebike-parts/d ... -dual.html

Total Draw 13.1W


Is this possible or will this exceed the limits of the DC output from the cycle analyst V3?

Is there a better way?

[ferret]

IIRC, the limit of the DC output depends on the pack voltage, there is a table in the manual specifying the limits.

You can use splitters on the pack main connector instead of going through the CA.

Avner.

[justin_le]

IIRC, the limit of the DC output depends on the pack voltage, there is a table in the manual specifying the limits.

Indeed. You can pull 1 amp no problem, but at 2A you'll be at risk of tripping the resettable polyfuse. Mchlpeel if your total load is just 13 watts then you'll have no issues at all. It's when you get up to the ~50 watt territory that you start approaching this current limit with standard voltage ebike packs.

[justin_le]

Oh man, that left button is something I've had dreams about! Thank you

What he said !

Alright, well my dreams for that go back over 3 years now too, so thanks Tek for finally making it happen!

It turns out though that there were a few silly bugs that crept into the P11 release (like the Ah data output column from the serial stream was missing) and these have hopefully all been fixed in P12, which is available from here, along with all the release notes:
http://www.ebikes.ca/downloads/CA3-0p12.zip

No new user functions over P11, but we'd be really happy if a number of people could take the time to update their firmware to this so that we can have it thoroughly vetted as the 3.00 code. When you unzip the package, there will be one firmware option labelled CA3-0p12_firmware_NoEeprom.hex. You can flash that to your CA using the CA firmware update tool and NONE of your settings or statistics will change or be overwritten. Everything should still be exactly as before but you'll have these new features available, like this left button hold to quickly escape the setup menu or save and move left during digit edits.

[Scotty T]

CRAP! I blew it. I got my conbag, and my crimper, I had everything hooked up nicely, and working, and did the PAS last. This is where it blew.

I have a basic PAS, it has red, black and blue wires. I routed them as:

red > white (pin 1 10v)
black > black (pin 2 gnd)
blue > blue (pin 3 dir)

I'm guessing I should have done blue > yellow (RPM)

Now the CA won't power up

Fuckity frock!

[amberwolf]

If you're powering a THUN or other BB sensor off that 10V, and your pack is higher than a certain voltage (48v? I forget but it's posted early in the thread somewhere, and I think in the manual), it can blow the main regulator of the CA.

Don't ask me how I know this.

Good news is it's not that hard a fix if you want to do it yourself, Grin should have the parts to send you, or they can fix it.

Or you can power the CA off an external 12V source (I ran mine off my lighting pack (instead of my main pack) for a long time before I blew it up completely via a totally unrelated wiring problem). It'll still be connected via the shunt/etc to the pack to monitor actual voltage and current, etc., it just wouldn't be powered by that pack anymore. (A DC-DC could also be used; I just already had a 12V pack for lights on CB2 so it was faster/easier to fix by using that than any other method at the time).


Might be some other problem, instead, but thought I'd mention that possibility just in case.

[Scotty T]

My system is 48v, the pack was putting about about 53v. I have a DP system and a Grinfineon controller so I can't power it off another source.

[amberwolf]

You can always power it off another source, since the power would come in on that low voltage line, not on the main power line (which would stay wired up just like it already is). But you'd have to add that other source, if you don't have it already available on the bike.

However until you open up and check the CA (or send it in to Grin) you won't know if the problem you have is the same I did, and thus if the fix I used would work for you.

[Scotty T]

Now that you mentioned it and I looked through the manual, I found the section about the PAS and the 10v pin which is exclusive for the THUN sensor and doesn't have a fuse, I can guess what I've done wrong. Someone who knows will confirm it I'm sure.

It looks like I needed to make up a 3 pin JST connector with red to red and black to black connected to the Aux Pot as my PAS sensor is a 5v type. Then only the blue wire should have gone to the yellow RPM pin on the PAS connector. I probably should have just got the PAS sensor which is compatible with the CA3. Damn.

[justin_le]

Now that you mentioned it and I looked through the manual, I found the section about the PAS and the 10v pin which is exclusive for the THUN sensor and doesn't have a fuse, I can guess what I've done wrong. Someone who knows will confirm it I'm sure.

Almost for sure all that happened is that your PAS sensor couldn't handle the 10V supply and blew, and then that drew excessive current form the 10V source which _hopefully_ simply fried the reverse protection diode which can easily either be replaced or bypassed with a short circuit to get the CA operating again. See this post:
https://endless-sphere.com/forums/viewt ... 00#p992104

It looks like I needed to make up a 3 pin JST connector with red to red and black to black connected to the Aux Pot as my PAS sensor is a 5v type. Then only the blue wire should have gone to the yellow RPM pin on the PAS connector. I probably should have just got the PAS sensor which is compatible with the CA3. Damn.

Well any PAS sensor can work if you open up the CA and move the white wire from the 10V pad to the 5V pad, or do as you say and peel the 5V from one of the other plugs, or just put a 10V->5V linear regulator at your connector so that this is stepped down to 5V before it meets the PAS. It's not only the THUN sensors which are designed for ~12V operation, the torque sensors from FAG, NTCE, and Sempu are all meant to natively run from this higher voltage so it makes sense to keep it available there.

[marcn]

Thanks for all your work Justin. Love the left button and readjustment of the PAS/ebrake relationship at slower speeds. Two things we were wanting!

[Scotty T]

I checked the diode and it was dead, I tried the short circuit with a bit of wire and plugged it in, it switched on momentarily then something went pop. I've put it back together, will get another one with the right pas sensor and start again and send this one off to be repaired.

[justin_le]

Thanks for all your work Justin. Love the left button and readjustment of the PAS/ebrake relationship at slower speeds. Two things we were wanting!

Hey appreciate the feedback Marc and nice to see that someone at least has been testing out the latest firmware updates!

I checked the diode and it was dead, I tried the short circuit with a bit of wire and plugged it in, it switched on momentarily then something went pop. I've put it back together, will get another one with the right pas sensor and start again and send this one off to be repaired.

Oops. I do presume that at least you had unplugged the PAS sensor before you powered things up again with the diode shorted? In any case, we'll be able to repair whatever you did for sure so just send us an email and we'll handle the RMA process from there.

[Scotty T]

Yeah only plugged in the ca3 connector from the grinfineon and nothing else. Will get an email off to you shortly.

[John Bozi]

Hope this relevant enough sorry if not...

In my noobness I don't understand what is different in the way CA V3 DP shutsdown.

I have run a lyen and grin controller with either an external switch or on the controller. Normally when you hit off the CA just turns off straight away...

I have just connected up a new 2.5kw-3kw greentime for the first time. But here's the deal.

there is no ignition wire or switch on this thing to my amazement. And I am setting it up on the big block so have not connected the phase or halls because the chain would be dangerous to all my wires right now if something went wrong.

When I disconnect my power wires the CA stays on counting down the voltage all the way to zero for quite a while about 30 seconds until it runs out....

Could anyone explain what the difference is? Is the CA draining the power out or is the controller draining or both? If I disconnect the CA first will it drain out or hold the charge longer so that I don't get arcing after 20 second restarting..

All I can say I don't recommend greentime to anyone with the lack of info and follow up. I didn't think it could be possible to have 20 useless wires and not have an essential ignition wire and that's why I didn't bother asking....

[izeman]

When I disconnect my power wires the CA stays on counting down the voltage all the way to zero for quite a while about 30 seconds until it runs out....

Could anyone explain what the difference is? Is the CA draining the power out or is the controller draining or both? If I disconnect the CA first will it drain out or hold the charge longer so that I don't get arcing after 20 second restarting..

as far as i understand it, there is a value for "CA shutdown voltage". If you see your CA count down to 0V before it shuts down, then this value seems to be set to 0V. I have mine set to a bit below LVC of the BMS, so if i turn off the BMS (open the switch) it counts down to 35V within a second and fades out.
in your situation it seems the CA is powered by the big caps in the controller. the slowly drain and feed the CA until there is no voltage left.
if you disconnect the CA it will keep voltage a bit longer, but it will still spark when you re-connect the battery.
you should use a central power switch to dis/connect the mains battery avoiding any sparks. i have soldered the "ignition" wire of all my controllers to battery VCC.

[markz]

3 way switch
3 way-switch
3-way switch
3-way-switch
power levels
power-levels

How do I hook it up so I have different power levels?

OK I found it, took awhile

https://endless-sphere.com/forums/viewt ... 40#p647750

Man oh man this looks tough. I get some of it, but man this brings me back to electronics college. Hopefully I just buy the 3-way switch and plug in 3 wires to a CA connector and go into the CA itself.

[justin_le]

When I disconnect my power wires the CA stays on counting down the voltage all the way to zero for quite a while about 30 seconds until it runs out....

This is just your controller bus capacitors slowly slowly discharging from the controller + CA's quiescent current draw.

Could anyone explain what the difference is? Is the CA draining the power out or is the controller draining or both? If I disconnect the CA first will it drain out or hold the charge longer so that I don't get arcing after 20 second restarting..

It will hold the charge for a bit longer. CA's contribution to the capacitor drain is 10mA. The controller is likely more like 30-40mA.

-Justin

[Scotty T]

Each to their own, I'd rather use a dedicated tool, one size for one job tool specifically for those JST crimps,
If you prefer the more universal 'one tool does many' style crimper, then I won't knock you for it.
At least now he has the choice for himself.

I have to agree that having reviewed the relevant training material on Youtube the ratcheting crimper from Justin was the one for me. It would have been annoying to have to do each part of the crimp separately and difficult to hold the connector without the ratchet. I only messed up one crimp too of about 18 that I did.

I'm all hooked up now and it's just a matter of getting all the settings correct. I definitely have a few values wrong in the setup as everything functions and it seemed ok unloaded but on riding it I have about half the previous speed and it's pulsing.

[John Bozi]

This is just your controller bus capacitors slowly slowly discharging from the controller + CA's quiescent current draw.
It will hold the charge for a bit longer. CA's contribution to the capacitor drain is 10mA. The controller is likely more like 30-40mA.
-Justin

Thank you Justin.

I gave up trying to use that controller for my current bike because I can't find how to connect the off switch.

Would you say it is unusual for this thin red CA wire to be connected in the way of my greentime controller? sorry for the blur... it is the only odd wire not coming from the other side of the board.

[justin_le]

Thank you Justin.
I gave up trying to use that controller for my current bike because I can't find how to connect the off switch.
Would you say it is unusual for this thin red CA wire to be connected in the way of my greentime controller? sorry for the blur... it is the only odd wire not coming from the other side of the board.

Yes, it's normal that the CA's red V+ wire would come from the same part of the PCB where the controllers V+ input is located. On most controllers that have an on/off switch, there will be a set of pads for V+ of the battery, and a nearby pad (sometimes labeled Vk of Vk+) which is the return V+ from the key switch, and this is what supplies the voltage to the linear regulator and all the controller circuitry. If you want the CA's power to turn off with the controller's on/off switch, then you want to make sure that it is soldered to the switched side of the V+ line and not directly to the battery V+.