Cycle Analyst V3 preview and first beta release

[justin_le]

Can anyone tell me which way to adjust [Rbatt] to make the fuel gauge (battery icon) go down quicker, i am going to keep adjusting it manually rather than use the actual IR so that it works over an 80% DOD instead.

Simon, it doesn't work that way. The RBatt term when properly adjusted simply makes it so that the battery icon doesn't move up or down when load currents or regen currents change the apparent pack voltage. If you are running LiFePO4 just ignore the fuel gauge until the Ah accumulation is integrated in the beta software here.

[Tench]

Thank you Justin, i am running lipo, 18s 16ah, yesterday i used 10.5ah and the gauge was still showing 3/4 full so i wanted to adjust it so that using that amount of my battery would display the battery as 2/3 used.

[teklektik]

Unlike v3B12 where I could not get Power Limiting to work even with WGain=1, v3B13 works pretty well although I cannot use values above WGain=15 without getting dramatic surging from a dead stop.

Interesting, I found on our test ebike (9C motor with 48V pack and 25A controller) that a value of about 50 did a good job. However, a lot of the infineon controllers have a very annoying discretization of the throttle input signal which is especially pronounced at low speeds. So they will jump from like 3% PWM to 10% PWM but nothing in between, and this makes anything trying to run with feedback at low speeds / dead stop really problematic.

I am running Xlyte Analog 12 FET 72v controllers at 66v (20s2p Headway). At this voltage these controllers seem to have a pretty steep step in the throttle response curve in the low to mid range which is a big reason for my enthusiasm about non-linear v3 throttle mapping. Although there are no discontinuities due to discrete quantization as you have described, the net effect is likely similar.

I think this harkens back to our email exchange where I mused about attributing the difficulties in stable feedback due to the pronounced low end torque of the dual gear motors - my view was too narrow in focusing exclusively on the motors and ignoring non-linearities of the controllers as part of the loop.

The oscillation/surging problem seems to be most pronounced when hitting the throttle pretty hard from a dead stop on an incline. When the weather clears, I can try some additional tests on different inclines with one/two motors, and different effective max power limitations to see if I can better characterize the behavior...

[justin_le]

Thank you Justin, i am running lipo, 18s 16ah, yesterday i used 10.5ah and the gauge was still showing 3/4 full so i wanted to adjust it so that using that amount of my battery would display the battery as 2/3 used.

Hmm, you don't happen to have a discharge profile for your pack do you? I based the lookup table on the LiPo packs from AEnergy as used in an older batch eZee batteries so I'd be curious to see how the voltage versus state of charge values compare. I've been surprised at how different of a shape curve you can get from the same nominal chemistry but from different manufacturers, certainly with LiFePO4 there is quite a contrast in the sharpness of the knee and final slope.

In a pinch for now you could just set the CA's #Cells to 19S instead of 18S, and then that would probably scale things down closer to what you want.

-Justin

[Tench]

I dont have a graph for my own pack but this one i found does seem to be close to how mine discharges in respect of similar voltages at half used etc, it is now sat at 3.75v p/c after using 10.5 of 16ah, which is near enough to the 2/3 used voltage of this graph.
My packs are Zippy 30c 8000mah.

http://www.google.co.uk/imgres?um=1&hl= ... s:15,i:105

[ColinB]

Hi,
This is a cool new development, and hopefully will draw in newbies like myself. I like how the kits offered at ebikes.ca are "open source" vs. Bionx. However, I like how the Bionx rides. Sounds like soon I will be able to have both. My brief tests weren't even though - comparing a heavy EZ bike with internal gears vs. a mid level hybrid with a 350w Bionx isn't exactly fair. But it was enough to tell which my preference was.

One question: I know the torque gauge adjusts the power based on your input, but is the amount variable while you are riding? For example, on a Bionx bike, when you are tired, you can press the "+" button to increase the assistance level. After you get to the flat you may turn it back down to conserve your battery. Will the CA be able to do this?
Colin

[hillzofvalp]

I just got some of the 615-1016-ND thermistors.. for whenever I receive a V3.
http://search.digikey.com/scripts/DkSea ... 1997269967

[k.ewin]

Hi,

the main issue with the new CA V3 design is the huge amount of wiring necessary to use all features. It would be much better to add an ANT+ receiver to the hardware. This would allow using wireless sensors and would open up a huge amount of possibilities like:

• Wireless power/cadence sensors. Even in pedals like this very expensive new one: Garmin Vector
• Heartrate monitors
• Wireless speed sensors
• Temperature sensors

See http://www.thisisant.com/ for additional information.

I do not know whether interference with other e-bike electronics would be an issue.

[macfrank]

It would be much better to add an ANT+ receiver to the hardware. This would allow using wireless sensors and would open up a huge amount of possibilities like:

Great idea to manage the communication wireless with ANT+!

I'm using ANT+ sensors (speed, cadence) with my MTB and it would be amazing to implement the CA too.

Question:
My CA (direct plug with speedo) has been shipped yesterday. It's Version 2.23 but there is an additional cable mounted (looks like a power plug), which is not shown in the user manual for 2.23 model.
Is this because my CA is a new hardware release which will be compatible with V 3.0?
Is there a change to update my CA 2.23 to 3.0 when the new firmware is released or do I have to buy a new device?

Thanks in advance...
Frank

[johnrobholmes]

The 2.xx and the 3.xx devices are not the same in hardware so you can't just update it for v3 features. The extra plug is probably for serial data, and can be used for programming.

[amberwolf]

My CA (direct plug with speedo) has been shipped yesterday. It's Version 2.23 but there is an additional cable mounted (looks like a power plug), which is not shown in the user manual for 2.23 model.

AFAIK that is for passing pack power out of the CA to Cycle Lumenators or other pack-powered handlebar-mounted devices.

[johnrobholmes]

Probably best to crack it open and double check or wait for Justins response then! I didn't realize they had such a plug for lights, thats pretty cool.

[hillzofvalp]

macfrank, fyi that's not my quote.

[amberwolf]

Probably best to crack it open and double check or wait for Justins response then! I didn't realize they had such a plug for lights, thats pretty cool.

There is a thread around here asking about a mystery plug on a recently-bought CA, and it turned out to be for the CL.

To test the theory, simply use a DMM to measure the pin and barrel of the connector, but best to put something over all but the very tip of whichever lead is going to be inserted down into the plug to check, so it doesn't short anything out.

[macfrank]

To test the theory, simply use a DMM to measure the pin and barrel of the connector, but best to put something over all but the very tip of whichever lead is going to be inserted down into the plug to check, so it doesn't short anything out.

I'm going to check with a DMM and post the results....

[Hyena]

Yep, they've had those for a few months now, it's meant for a headlight but it's straight pack voltage passed through so if you're not using the ebikes.ca luminator that's good for higher voltage you'll need a DC-DC converter. Common sense would dictate not to hook up a huge load to it - all the power is coming from your battery over the thin gauge wires to the CA. For most LED headlights they're fine though.
The guys @ ebikes.ca said they're good for up to around 1 amp - but I assume this is 1 amp @ ~36v not at 100v+ like half the maniacs here run - so don't try and pull 100w from it. The luminator draws ~10w so it's good for atleast that

[hjns]

Hi all,

10x 10k thermistors ntc came in, and I connected 5 of them to my CA v3 just to test. Unfortunately, the CA then shows temps between 5 and 8 oC with an actual ambient temp of 25oC, and the CA shows around 28oC when I cover the thermistors between my fingers. Assuming that my fingers are around 35oC, there is a correction factor of around 17oC. I think it would be great if that could be adjusted in the CA. Note that these are the el-cheapo thermistors from Ebay.

[justin_le]

Hi all,
10x 10k thermistors ntc came in, and I connected 5 of them to my CA v3 just to test. Unfortunately, the CA then shows temps between 5 and 8 oC with an actual ambient temp of 25oC, and the CA shows around 28oC when I cover the thermistors between my fingers. Assuming that my fingers are around 35oC, there is a correction factor of around 17oC. I think it would be great if that could be adjusted in the CA. Note that these are the el-cheapo thermistors from Ebay.

Hi Hjns, did you update the firmware with the B13 that was posted last week? In the original B12 the look-up table was assuming we used a 10K pull-up resistor although we made a board change to a 5K pull-up at the last minute so that the device would be more sensitive and accurate at the higher temperature range. If you don't have access to a programmer and want to use the B12 firmware, you can either a) connector two of your thermistors in parallel which will effectively behave like a single 5K thermistor so that the ratio is right, or b) replace the 4.99kOhm pull-up resistor on the PCB (R1) with a 10K pull-up.

-Justin

[justin_le]

Probably best to crack it open and double check or wait for Justins response then! I didn't realize they had such a plug for lights, thats pretty cool.

There is a thread around here asking about a mystery plug on a recently-bought CA, and it turned out to be for the CL.

Yes, we decided to make the DC power port standard on all the CA's a few months ago but haven't yet got around to updating all of the documentation. The idea is to make it a convenient access point for those wanting to power small handlebar accessories (headlights, horns, handgrip warmers, stereos, GPS/ipod chargers etc.) without running an additional cable line back to the battery. Just a note that if you aren't using this, to leave the rubber cap on the DC plug so that it isn't at risk of shorting or corroding since the full pack voltage is present there.

CA_Power_Port_TN.jpg (8.2 KiB) Viewed 451 times

This is going to be standard issue in all the V3 CA's as well and will be included in the pilot batch with all the wiring harnesses. For people wondering what the big yellow square is on the V3 CA boards, that's a resettable polyfuse inline with the power tap in case of a short or excessive current draw. The V2 boards don't have this yet and hence the caution to be careful about shorts.

CA V3 Polyfuse.jpg (70.76 KiB) Viewed 451 times

To test the theory,

No need to test

[justin_le]

Hi,

the main issue with the new CA V3 design is the huge amount of wiring necessary to use all features. It would be much better to add an ANT+ receiver to the hardware

I love the convenience and cleanliness of wireless too, I just wonder (EMI interference problems aside) about the wisdom of having core components of an ebike all needing their own independent battery power. Imaging if you had to have coin cells in your speedo sensor, coin cells in your cadence sensor, coin cells in your throttle, temperature probe, and ebrake cutoffs etc. That's a lot of small batteries to stay on top of.

For something rather peripheral to operation like a heart rate monitor it wouldn't matter much of the sensor battery died in the middle of a trip. But say you have a bike with PAS control and the cadence sensor shuts off because it ran out of battery? Or you run in cruise control mode but the speedo sensor battery dies?

In any case it's something we're keeping an eye on for possibilities if there are cases where it makes sense, but for the most part I'm mostly of the opinion that items that are _part_ of the ebike are best permanently wired up, even if it's a bit more inconvenient at first. -Justin

[amberwolf]

Imaging if you had to have coin cells in your speedo sensor, coin cells in your cadence sensor, coin cells in your throttle, temperature probe, and ebrake cutoffs etc. That's a lot of small batteries to stay on top of.

At least in sunny places like here in Phoenix, maybe use those little solar chargers like some of the keychain LED lights have built in? I don't know that they would ever fully recharge the cells, but they could certainly extend lifespan, if the cells can handle the recharge, and the solar cell can output enough to either recharge it afterward or even fully power the device and transmitter....

[teklektik]

...maybe use those little solar chargers like some of the keychain LED lights have built in....

Seriously?
+1 for copper!

[amberwolf]

I dunno that it would really do much good, but thought I might mention it.

[hjns]

Hi Justin,

No, I still run the B12, as I still need to obtain a TTL-USB cable. Anyway, I put two thermistors in parallel, and it nicely shows ambient temp around 20oC (07.30am here now). I lighted a cigaret lighter and put the thermistors in the warm air above the flame, and the CA showed the temp increase. Within seconds the CA showed 130oC, after which I removed the lighter. Then the CA nicely showed the decrease in temp, and within 1 minute the temp was ambient again.

Very happy with the results! Thanks, Justin for the support and for developping the CA in the first place!

Maybe a suggestion to keep all requirements for thermistors the same between future CA versions and the CAnalogger, or have a simple pot added to the PCB to change the pull-up value.

Hi Hjns, did you update the firmware with the B13 that was posted last week? In the original B12 the look-up table was assuming we used a 10K pull-up resistor although we made a board change to a 5K pull-up at the last minute so that the device would be more sensitive and accurate at the higher temperature range. If you don't have access to a programmer and want to use the B12 firmware, you can either a) connector two of your thermistors in parallel which will effectively behave like a single 5K thermistor so that the ratio is right, or b) replace the 4.99kOhm pull-up resistor on the PCB (R1) with a 10K pull-up.

-Justin

[justin_le]

Maybe a suggestion to keep all requirements for thermistors the same between future CA versions and the CAnalogger, or have a simple pot added to the PCB to change the pull-up value.

Yes, that is the intention, both the Analogger and CAV3 are designed to use the same 10K NTC thermistor which is the most common value we find on BMS boards and inside ebike batteries. The situation of needing to put two in parallel for 5K is just a temporary hardware fix to which can work until you update the firmware.