The Cycle Satiator, universal charger for the enthusiasts

[markz]

I am amazed you can do 103V and 5A on that unit.
I am thinking about buying either the 72V because it can do

36V to 84V Nominal Batteries, 103V 5A max.

But the 48V model cranks out a little more amps

24V-52V Nominal Batteries, 63V 8A Max Output.

I am not a speed demon by any means, I like cruising at 30-40kph so 48V is what I like. When I do go 72V its not speed I am after but torque via smaller wheel diameter. I got my 9C clone kit I do 36V or 48V on, my BMC V1 geared I will do the same. The MXUS 3kW is the only one I do 72V on.

I hope Grin puts a bit of a deal on this at the bike show in Van.

[Marc S.]

I am amazed you can do 103V and 5A on that unit.
I am thinking about buying either the 72V because it can do

36V to 84V Nominal Batteries, 103V 5A max.

You can't. 103V x 5A = 515W
Since the Satiator is a 360W charger, at 103V it will max out at 3.5A.
Hence, charging at 5A is only possible up to 72V.

But the 48V model cranks out a little more amps

24V-52V Nominal Batteries, 63V 8A Max Output.

63V x 8A = 504W
Again, they are 360W chargers. At 8A output, it will 'only' be able to charge up to 45V.
At 63V it will charge with up to 5.7A.

[pucksterpete]

Sounds like we might have some defective batches coming off the assembly line in China

[markz]

Ah yessss Thanks Marc S for clearing that up.
Thats what that little astrics is for.

[mistercrash]

Sounds like we might have some defective batches coming off the assembly line in China

I have the same error message. Sent an email to Grin. I hope this can be fixed easily without sending back the unit.

[justin_le]

Hey everyone, nothing too pressing but for those who want the most recent firmware on the Satiator devices we now have a V1.023 release that you can upgrade to. Most of the changes are small behavioral fixes to allow the Satiator to better cope with scenarios of BMS circuits tripping and releasing and some risks of damaging the soft start circuitry if you disconnect and reconnect a satiator from the battery at just the right (wrong) moment. You should be able to just click "Get latest firmware" from your existing satiator software suite or you can download it again from here:
http://www.ebikes.ca/product-info/cycle ... #downloads

In any case, the main reason I'm writing is to mention that we have indeed experienced a much higher than expected issue rate on a recent (late 2016) production run of Satiator devices that we've been able to trace back to what seems to be a compromised batch of film capacitors. This was a real puzzle since we are inside the safe operating zone for the cap over the full operating voltage, currents, and temperature limits of the device, and had no capacitor issues in the first two years of widespread release. There are 3 tank caps on the device and if one of them fails then there will be frequent screen glitches and occasional "Pwr Fault" error messages, but when 2 or 3 of them go then the charger doesn't really work at all. When we disassemble the failed capacitors they all show signs of corrosion/oxidation of the metal film, while those from the earlier runs with 1000's of hours of use are all perfectly shiny all the way through.

FilmCapacitors.jpg (65.02 KiB) Viewed 1458 times

FailedCapacitorFilm.jpg (193.11 KiB) Viewed 1458 times

It looks like a case where there was a manufacturing issue on the cap itself and we are pursuing an investigation to try and trace this.

Anyways, if somebody has issues with their Satiator along these lines regardless of where you purchased the unit from, please contact Grin directly by email and we'll sort things out under warranty right away. The capacitors are easy enough to replace but there are some risks in the disassembly and reassembly of the unit and so we'll generally want to have the device send back to us for repair/replacement, and we apologize for the inconvenience to anyone who is affected by this.

It's not at all what we wanted to have happen and sadly there wasn't an easy way to catch this earlier during the QC and burn-in steps.

[justin_le]

Oh, and I forgot to mention we also released a prototype build of the firmware that allows you to use the serial port for selecting profiles, starting and aborting charge etc. This is just a basic proof of concept implementation, and doesn't support any realtime logging of the charge data, only the charge state (ie whether charging is complete, battery is disconnected etc.) for evaluation by companies that want to use the Satiator in the context of public ebike charging stations and what not.

Both the command instructions and firmware are attached if anyone here has a use for this.

[amberwolf]

I might've asked this before (can't remember) but is there any chance of making an alternate firmware that uses the signal line in an analog way to detect which pack it is hooked to?

Then in the profile for a particular setup the voltage (resistance?) expected on that line would be it's own field. (automatically excluding the temperature stuff for NiXX/etc packs).

This way anyone that has basic electronics skills can install the appropriate resistor in teh pack's charging connector for the third pin of their connector (XLR or whatever), then setup a profile that includes that so the Satiator can activate it when it's detected.

Depending on how the firmware is made, it coudl have a setup option to disable or enable this feature, so that those using temperature sensors for NiMH or NiCd (etc) could still use it normally, and it woudlnt' auto-change profiles, but those that did want to enable it could do so.


It'd be a feature that would have to be explicitly enabled, and profiles explicitly created to utilize it, as a safety feature to prevent accidentally using such profiles on packs that just happened to have a sense line wired in a way that would cause activation of a wrong profile on it. (as unlikely as that would be).

[justin_le]

I might've asked this before (can't remember) but is there any chance of making an alternate firmware that uses the signal line in an analog way to detect which pack it is hooked to?

Yes, this was discussed in some detail on this thread here:
https://endless-sphere.com/forums/viewt ... 67#p940267

Basically, the idea is there and if there is a decent volume OEM user of the Satiator that would need this feature then we'd be able to implement it. At the moment though, most of the potential users doing automatic pack ID are using a more robust RFID or similar digital handshaking protocol of their own choosing, and can then use this digital command interface to select the appropriate profile on the Satiator and start charging.

The analog solution with resistors in the 3rd pin of the XLR plug would be a lot easier for home and DIY users, but we haven't received much interest in this and I'd be a bit nervous about water ingress in the connector or other things messing with the analog signal (like a broken solder joint) causing the wrong profile to be automatically selected.

[Eerder Metaal]

I've got a problem with a 48 volt satiator which I used succesfully for two weeks, about ten charges. Started with the V1.23 firmware. But now it keeps on going on into pwr fault. Also on the next day and after reinstallation of the firmware. I checked the "Overwrite Profile Data" checkbox and reloaded the profile data from file afterwards. This 8P10S battery I build myself has no bms. I tried a different battery with bms, but no success either. With another Satiator, same model and age, but maybe an older firmware version (have to check), I don't have this issue, so it seems to be the charger who's tripping

What else can I do to get it going again?

cheers Vincent

[justin_le]

What else can I do to get it going again?
cheers Vincent

Hey Vincent and anyone else in a similar boat. Just send an email to info@ebikes.ca explaining the situation and we'll arrange a quick fix under warranty. This almost for sure is a case of one of these bad film capacitors.

[cj0]

I'd wish a Travel Satiator or Satiator Lite.

Primarily with reduced weight, from 1.33 kgs down to 0.4-0.6 kgs.

I can live with reduced features, like:
- (much) lower current / power (2A would be acceptable, 4A is better)
- shorter cables
- no galvanic isolation / no "PE" connection on mains input (a dual conductor philips type cable weights less then the current triple conductor cable)
- no display (instead 3 options to select a charging-profile, like a switch or toggle button and 3 corresponding LED's) [storage 45%, battery-life-optimized full @80%, maximum-range @100%]
- less weight, by:
(a) light weight casing
(b) different charging connector [then XLR]
(c) and/or no Y-splitter cables, only 1 output directly from adapter directly to battery connector
- no-load power consumption according to
(a) EU CoC Tier 2 (<= 150 milliwatts @230V for 50-250W output power)
(b) DoE even @230V (<= 0.500 milliwatts for >250W output power)

I can't live with:
- 36V charging
- reduced efficiency
- no mechanism to program charging profiles

[racingame]

Does this universal charger work with Bosch, Shimano, etc. batteries? Since I know they use some kind of CAN protocol that doesn't allow other charger to work with their batteries. I found a German company who managed to make a really universal charger that could work with any maker battery but forgot its name.

[Alan B]

I might've asked this before (can't remember) but is there any chance of making an alternate firmware that uses the signal line in an analog way to detect which pack it is hooked to?

Yes, this was discussed in some detail on this thread here:
https://endless-sphere.com/forums/viewt ... 67#p940267

Basically, the idea is there and if there is a decent volume OEM user of the Satiator that would need this feature then we'd be able to implement it. At the moment though, most of the potential users doing automatic pack ID are using a more robust RFID or similar digital handshaking protocol of their own choosing, and can then use this digital command interface to select the appropriate profile on the Satiator and start charging.

The analog solution with resistors in the 3rd pin of the XLR plug would be a lot easier for home and DIY users, but we haven't received much interest in this and I'd be a bit nervous about water ingress in the connector or other things messing with the analog signal (like a broken solder joint) causing the wrong profile to be automatically selected.

This could easily be done with a tiny micro board like an Adafruit Trinket or Arduino Nano, or even directly with the Atmel chip itself which could be made to fit in a very small space. Just embed this in the battery pack and respond to the serial stream. Definitely within the DIY range.

Perhaps rather than just select a profile, the serial stream could specify the actual parameters, so it would work even if the Satiator was not programmed, with appropriate checksumming on the values in any case to avoid a communications error selecting the wrong profile or values.

[BVH]

Found my old post and the info. I don't remember it's been so long since I played with Satiator settings. When measuring for cable resistance, do I measure both + and - and add the two for total Ohms?

[cj0]

I do want to share some real world 230VAC no load Cycle Satiator 48V r7 1.025 firmware power consumption results.

Measurements are made with a ZES Zimmer LMG95 precision power meter, at 19ºC 39% relative humidity. The 230VAC is a utility supply, so constant varying. In this case betwee 230VAC and 234 VAC. The power meter samples at 100 kHz, updates the after 100kSamples, and was set to average the last 1000 measurement cycle (=seconds) values.

VD0B1755.jpg (121.51 KiB) Viewed 283 times

The Cycle Satiator DUT was set at brightness 0 (=lowest brightness) and had at least 8 hours of warm up time (actually cool down after a charge cycle). The AC measurements were taken at the socket using the supplied KENIC KE-34 to KENIC KE-23 (Schüko) H05VV-F 0.75mm² power cord (1.5m¹).

The measurement results:

VD0B1753.jpg (37.86 KiB) Viewed 283 times

P: 0.9873 Watts
Q: 35.5115 var
S: 35.5252 VA
PF: 0.02779 capacitive

VD0B1754.jpg (40.46 KiB) Viewed 283 times

P: 0.9876 Watts
Itrms: 152.7978 mA
Utrms: 232.570 Volts
PF: 0.02779 capacitive
---

VD0B1756.jpg (41.85 KiB) Viewed 283 times

P: 1.0017 Watts
Itrms: 153.357 mA
Utrms: 232.695 Volts
PF: 0.02807 capacitive
---

VD0B1764.jpg (45.43 KiB) Viewed 283 times

Itrms: 153.082 mA
Iac: 153.082 mA
Idc: 0.240 mA
Ipp: 698.880 mA

[cj0]

Satiator Mac OS X software suite feature request:

relocate saved_profiles somewhere in the /Users folder

The Satiator suite XML files need to be user editable, for instance by downloading them from the charge simulator directly into the saved_profiles folder.

rel. 1.025

/Applications/SatiatorSoftwareSuite.app/Contents/MacOS/Satiator Software Suite Files/saved_profiles

makes the xml profile files not turn up in Spotlight when a user searches for its file name.

-----------------------------------------------------------------------------------
PS: A downloadable Cycle Satiator icon file like

CycleSatiator-tiny-icon.png (3.24 KiB) Viewed 277 times

is welcome too (to fix the missing app icon)

[cj0]

Cycle Satiator 36V & 48V charge efficiency: η
η = greek small eta = energy efficiency
Friday I picked up the new 48V Cycle Satiator running firmware 1.025.

From the Chinese Shenzen Modiary 13s BC263480050 Li-Ion battery charger I know that the bulk charging power factor is between 0.954 and 0.962, and that the end-to-end (between wall socket and battery terminals) charge efficiency for a 10s/36V pack is 89.46% efficient (EP in [Watthours] versus EP out [Wh]).

Now I am wondering what the results will be for the Cycle Satiator. So I did empty the same 36V 10s6p and a 2 months newer 48V 13s8p Li-Ion battery pack while cycling home. Next day I build a test rig for the measurements and hooked up the LMG95 precision power meters:

VD0B1742_900.jpg (128.81 KiB) Viewed 264 times

The 36V pack was charged at the recommended 2.5 Amps rate till 42.0 volts, and the 48V pack was charged using 5.5 Amps and 54.6 volts profile.

36V pack charge efficiency
η = 89.64% ≍ 418.463 Wh ÷ 466.787 Wh
PF (top display) while charging = between 0.64 and 0.89
Note: this measurement is too optimistic in the sense that charging was ended on the power meters within 5 minutes of ending the charging process, as there is no audible end-of-charge signal on the Cycle Satiator, it is not realistic that a regular user will unplug this short after reaching End-of-charge, thus in real life your numbers will be lower due to standby power consumption of the charger.

48V pack charge efficiency
η = 90.95% ≍ 1.00791 kWh ÷ 1.01812 kWh
PF (top display) while charging = 0.97 (sorry, only photographed once)
Note: this is a more realistic efficiency value. After charging completed my good night of sleep was not over yet. Between 4 and 8 hours of charger standby losses are included.

[cj0]

All efficiency measurements were made using a Cycle Satiator firmware v1.025 that suffers from the

power fault message the moment the satiator starts charging

issue.

Detailed charge progress on display (frame by frame from movie).

1.

Charging to  42.0V

Sati IMG_2137.MOV.png (57.8 KiB) Viewed 137 times

2.

0 Watts   23ºC

Sati IMG_2137.MOV-1.png (56.5 KiB) Viewed 137 times

3.

  Watts     ºC 

note: very short, only 1 frame!

Sati IMG_2137.MOV-2.png (53.06 KiB) Viewed 137 times

4.

Batt V too low 35.7V

note: very short, only 1 frame!

Sati IMG_2137.MOV-3.png (45.65 KiB) Viewed 137 times

5.

Pwr Fault: Restart

Sati IMG_2137.MOV-4.png (45.11 KiB) Viewed 137 times

[cj0]

Downgrading from v1.025 to v1.024 was the solution for me.

I haven’t seen any “Pwr Fault: Restart” before-starting-charge-issue since the downgrade.

[justin_le]

Downgrading from v1.025 to v1.024 was the solution for me.

I haven’t seen any “Pwr Fault: Restart” before-starting-charge-issue since the downgrade.

Yeah this is a very new issue and we've had two people (the poster here and one other user) recently have this situation where the charger works fine with the v1.024 FW but has starting glitches with 1.025 (either Pwr Fault or a premature Charge Complete). If there is anybody else who has witnessed this on the firmware update please get in touch with us. We'll be able to issue an updated debug firmware that outputs a log file which should help us pinpoint just what is going on in the startup sequence.