Cycle Analyst V3 No Throttle When Connected Directly to CA

dagher58 said:

Hello ive been testing and digging through endless sphere and the unofficial manual for the Cycle Analyst the past few weeks and still had no luck getting my throttle to work when plugged directly through my CAv3

Click to expand...

I think you mean "indirectly", since if it were direct it would be plugged into the controller, not the CA. Indirectly, because it goes into the CA first, then the CA creates it's own throttle output signal, then you wire that to the controller. (the distinction is important, because you use the word directly for both cases, which is confusing)

So...that said, exactly what have you created for wiring from the CA to the controller? Are you wiring from the CA's CA-DP connector to the controller's throttle connector? I am not aware of Sabvotons with CA-DP connectors, so you would not be able to plug the CA directly into the controller via the CA-DP connector, and pass it the throttle that way.

If yours does have the CA-DP connector, you should verify that the wiring from it goes to the correct places inside the Sabvoton, *and* that it is not the wiring for the old CAv2, which uses a diode in line with this signal and isn't directly able to do what you want without bypassing that diode.

If you look at the Diagnostic screen on the CA, the first one to the "left" of the main screen (one push of the left button when on the main screen), what are all of the things you see there? Since it is all text you can type it in exactly as you see it, including any letters that are capitalized or lowercase exactly as the screen shows. See the end of the post for a quote of Grin's page and a picture of the screen.


You should also verify that the actual voltage at the controller's throttle input wire is indeed changing with the CA output readings, using a voltmeter or multimeter, so that you know it's not a simple connection or wire problem.

Display #12 – Diagnostics

The final display screen is most useful when initially setting up the CA for various feedback and control modes. On the top line you can see both the input throttle voltage from the user, as well as the output voltage that the CA is sending to the motor controller.

CA Screen #12, Diagnostics with Throttle Input and Output Voltages

The bottom line has a row of limit flags showing which throttle limiting feedback loops are currently engaged. So if you are exceeding the speed limit, the ‘s’ becomes a large ‘S’, and similar flags for the current limit (a/A), power limit (w/W), low voltage rollback (v/V), and thermal rollback (t/T). If for instance you have the bike setup for proportional pedalec assist, then the commanded power limit of the CA is being set by your pedal effort, and so the power limit flag ‘W’ will be set showing that the CA is in a power limit control loop.

Click to expand...

dagher58 said:

Hello ive been testing and digging through endless sphere and the unofficial manual for the Cycle Analyst the past few weeks and still had no luck getting my throttle to work when plugged directly through my CAv3

E-Bike Specs:
72v 40AH Battery
150a Sabvoton 72150 Controller
8000w QS 273 Hub Motor

Sabvoton Controller Settings
1.3v min Voltage (when I set to 1v via Bluetooth app the wheel will spin on its own)
4.4v max Voltage

Cav3 Settings
Throttle in
Input Throttle Mode: Pass-Thru
Zero Throttle Threshold: 1v
Full Throttle Threshold: 4.08v
Fault Throttle: 4.46v

Throttle out
Min Throttle Output: 1.2v
Max Throttle Output: 4.4v


CA detects throttle in idling at 0.85v and throttle in max at 4.23v. No matter what settings i try i cant get my motor to spin when I have my throttle plugged directly into my Cav3. It only seems to work when my throttle is plugged directly into my Sabvoton controller. Any help would be much appreciated.

Click to expand...

Hello Amberwolf and thank you for your reply, I definitely meant "indirectly" it goes through the CA first then the throttle output wire is connected to my sabvoton controller throttle connector.

My controller doesn't have a CA-DP connector so I had to set it up to an external shunt. Ill attach an image below of the exact wiring diagram I followed in order to wire up my cycle analyst to my sabvoton controller. In the attached photo you will see a 2K Resistor on the throttle line, I do not have that connected although I have tried to put one in the past thinking it may solve my issue but it didn't work lol.




When I am on the main diagnostic screen this is what I see when my throttle is resting

In 0.85 Out 1.2
awsvt 76v 0A 0

and under full load the CA diagnostics screen reads

In 4.23 Out 4.39
awsvt 76V 0A 0


Tomorrow when I get to the shop ill verify that the actual voltage at the controller's throttle input wire is indeed changing with the CA output readings and keep you posted.

dagher58 said:

My controller doesn't have a CA-DP connector so I had to set it up to an external shunt. Ill attach an image below of the exact wiring diagram I followed in order to wire up my cycle analyst to my sabvoton controller. In the attached photo you will see a 2K Resistor on the throttle line, I do not have that connected although I have tried to put one in the past thinking it may solve my issue but it didn't work lol.

Click to expand...

The wiring they show looks correct, as long as the Grey wire they indicate is indeed your throttle signal wire on the controller. (I've attached the linked image to this post in case it goes away at that linked site)

The resistor is not necessary, unless you *also* have a throttle directly connected to the controller at the same connector.

As long as there is a common ground (battery negative, etc) between the CA and the controller, then that plus just the throttle signal wire is enough to get the correct throttle signal to the controller. (as indicated in the diagram, the 5v doesn't get connected between the CA and controller, as each one is independently regulated).

So as long as the wiring is good between the CA and the controller, then it should be responding to the CA's throttle signal normally, exactly as if it were a direct throttle connection.

When I am on the main diagnostic screen this is what I see when my throttle is resting
In 0.85 Out 1.2
awsvt 76v 0A 0
and under full load the CA diagnostics screen reads
In 4.23 Out 4.39
awsvt 76V 0A 0

Click to expand...

Those appear to be normal, except for the controller not drawing any current at full throttle.

The all-lowercase awsvt indicates that no limiting is occuring inside the CA, as any active limits would be capitalized. Similarly, if the CA were limiting, it would be reducing the throttle voltage output, which it isn't doing since the Out value is at your max throttle out value.



Also: It doesn't make any difference to the problem you're having (because you are getting throttle out as expected for the CA itself, at least up to the point the CA reads it's voltage internally), but I noted that your CA .hex file shows it is from the CA 3.1 beta 22 firmware, rather than the latest 3.14 version. Just to make sure, what firmware version does your CA tell you it is running when you first turn it on? (if it isn't 3.1b22, then perhaps the settings uploaded into the CA are corrupt even though it appears to be doing what it should, and that it might not actually be outputting what it says it is--I don't know if it is possible or not, but it is something to verify).

If it matters, there are various things fixed and/or upgraded between the old 3.1 version and the newer 3.14. I don't recall what they are, but the info is in the various release notes files inside the help folder for each different firmware version. If you want to you can use the setup program to upgrade the firmware on the CA to the latest, if you find that it would do something you want to do that the beta doesn't, or that it fixes a problem you are experiencing. I recommend that you write down the Protected settings values in the Calibration section on paper before upgrading, in case they are somehow lost. To see them you have to enable the "show protected settings" in the setup program Preferences menu.


Do you also have a digital aux setup connected to the CA? If not, you may want to disable the digiaux (set to "none") in the digital aux section. It isn't affecting your setup ATM, but various settings in the CA do interact and when a control isn't present that it thinks is connected, it may behave unexpectedly under some conditions.

Similarly, if you don't have a 3-speed switch (wired with resistors inside appropriately) connected to the aux in of the CA, you should also set that to "none".


Same for PAS--set that to "none" if you don't ahve a PAS cadence sensor hooked up.

dagher58 said:

When I am on the main diagnostic screen this is what I see when my throttle is resting

In 0.85 Out 1.2
awsvt 76v 0A 0

and under full load the CA diagnostics screen reads

In 4.23 Out 4.39
awsvt 76V 0A 0


Tomorrow when I get to the shop ill verify that the actual voltage at the controller's throttle input wire is indeed changing with the CA output readings and keep you posted.

Click to expand...

Can I ask,
The CA set up file you posted. Is this the one you are using?
If so, the battery settings are wrong. It will only read 42v on a fully charged battery.
Then for the shunt. Can you say what shunt you have used?

Just for giggles, Santacruz, could you close and reopen the file downloaded, and check the firmware version it is for (in the titlebar) and recheck the battery number of series cells in preset 1?


I ask because something is very strange. I'd downloaded and extracted that file yesterday, and it showed it was setup for 20s, not 10s as you note (and that it was for v3.1b22 firmware). Just this moment, after your post, I redownloaded it, extracted it, and opened it in the CA setup program, and it now said what you saw--a 10s pack, and it said it was for version 3.19 firmware (which, AFAIK, doesn't exist). Then I opened a new instance of the CA setup program and used that to open the original version of the file from yesterday, and this showed it was the same as it had been yesterday (20s, 3.1b22). BUT: now the version *already open in the other instance* had converted itself to match this one!

I've always been able to open multiple instances of the CA setup program and compare files, without issue, and I still can, with any file I have here (except that one, perhaps).

But...now I open any copy of the file downloaded from this thread, including yet another fresh download of it, and they all come up as 20s, v3.1b22. So...perhaps something in this file is corrupt.

I'm not sure what happen with the Hex file as it sounds like it got corrupted but I setup a new one making sure it was for firmware v3.14 and with all the settings as accurate as I can set them and will attach it below. I reuploaded the new settings to my CA today and ran a few tests but still no luck. Just to note I do have a 3 Speed switch and digital aux installed and to answer your question Santacruz the shunt I have installed says 200a 50mv on it. I remember reading somewhere online that I should set my shunt resistance to 0.500mOhm, I hope that's correct.

Todays Update: So today I grabbed my multimeter and started testing and I think I may have narrowed the issue down... I hope, you guys let me know.

Wired "indirectly" through the CA first then the throttle output wire is connected to my sabvoton controller throttle
CA3 Throttle Connector: + (5v) and - (Neg) read 4.9v
CA3 Throttle Connector: - (Neg) and Sig (Green) wire read 0v

Sabvoton Throttle Connector: + (5v) and - (Neg) read 4.7v
Sabvoton Throttle Connector: - (Neg) and Sig (Green) wire read 0v

Plugged "directly" having the Throttle run straight to the Sabvoton controller throttle
- (Neg) and Sig (Green) wire read 0.8v

BE CAREFUL when testing with multimeter, I tapped in on Sig (Green) and the - (Neg) from the sabvoton throttle connector and the bike went into full speed after giving a bit of throttle, luckily I had the rear suspended in the air and my e-brakes enabled for quick cut off.

Based on the testing I did today it seems like the Sig (Green) Wire from the CA-DP Connector reads 0v even though I can see the throttle voltage changing in the CA diagnostic screen. Maybe that's my issue, my CA is not outputting a throttle signal telling my sabvoton to go ? I cut the CA-DP wire close to the top of the CA to see if maybe I had a nick in the wire but I had good continuity through the cables. I made a new 6 pin connector and put it back together, at least this way I'm certain my wires are good.

Let me know what you think, and thanks again !!

dagher58 said:

shunt I have installed says 200a 50mv on it. I remember reading somewhere online that I should set my shunt resistance to 0.500mOhm, I hope that's correct.

Click to expand...

You can use ohms law to figure out the resistance. If it gives 0.05v at 200a, then that's R = V / A, so 0.05 / 200 = 0.00025 ohms, or 0.250 milliohms.

Wired "indirectly" through the CA first then the throttle output wire is connected to my sabvoton controller throttle
CA3 Throttle Connector: + (5v) and - (Neg) read 4.9v
CA3 Throttle Connector: - (Neg) and Sig (Green) wire read 0v

Click to expand...

First, which connector is this actually being tested at?

If it is actually at the CA3 *throttle* connector, then there won't be any output there, as that is only an input. You'd need to test at the CA3 CADP 6pin connector, the green throttle out wire on pin 6 (opposite side from the red battery positive wire). AFAICT from what you've said in the rest of your post, that is how you have it wired (from the CADP connector), but I want to be explicitly clear to be sure it's not just a wiring issue.

As long as the 5v wire is *not* hooked up between the controller and the CA, but the signal (from CADP connector) and ground wires *are*, then the wiring is correct.

Next, are you testing while moving the throttle position, so that you should be seeing a changing output from the Throttle Out signal wire? This is just to be sure it isn't a setting issue--if it is hardware it would typically stay at 0v, if it's something else it'll still change but maybe not the amount that's expected, as you should be seeing a minimum of 1.2v on that wire.

If so, and there is still only a 0v reading, then this means the CA hardware is not outputting the expected signal. Unplug the CA's throttle signal wire from the controller, and retest just the CA signal wire output. If it now reads correctly, then something in the Sabvoton's input is loading down the signal so much that there isn't any output, or the wiring is not actually correct (such as having ground and signal reversed).

If it still doesn't read correctly, then there's no signal out from the CA. The most likely problem is a wiring fault inside the CA, so at this point I'd open up the back of the CA (just four philips screws) and verify the green throttle out wire from the CADP connector does indeed go to the THO pad on the short end of the CA board inside. If it does, then also measure the voltage on that pad. If it reads correctly there, then the wire isn't correctly soldered to that pad, whcih is easily fixed.

If it still doesnt' read correctly even at the THO pad, then the CA itself may be defective in hardware.

Theoretically it "can't" be a setting or limit in the CA programming because if it was, you wouldn't see the Throttle out voltage tracking the throttle in on the diagnostic screen, and you do see that.



BTW, the new CA setup file does read teh correct 3.14 version, and it also has the right 20s battery, etc.

amberwolf said:

Just for giggles, Santacruz, could you close and reopen the file downloaded, and check the firmware version it is for (in the titlebar) and recheck the battery number of series cells in preset 1?

Click to expand...

Okay, I think I found what caused the reading problem my end.
If I opened the file which booted the program it read 10s for the battery etc.
If I opened the program and used it to load the file, it showed correct.
Maybe something to note in the way the CA program is used.

0.25mohm is correct for the shunt and the CA also needs to be set to High. I don't see this will have any effect on the throttle, but your power readings will be off.
However, I do not know if this has any effect. My shunt is 0.30mohm and the CA is set to high. But my amps feed back gain is set to 900 and my power feed back gain is set to 250. Yours are too low. As you are running in high mode, these setting effectively are increased by 10 times. (Mine are not exact because I am fine tuning), But I think yours should be more like 500 for amps and 200 for watts.

Amberwolf's suggestions are a very good route to follow, I would also make sure the driven wheel is suspended, just in case something powers it up.
Also, did you try it in Bypass on the throttle in, mode setting? That is also another way of confirming wiring without swapping any plugs.

When I mentioned "CA3 Throttle Connector" I was referring to the throttle input in the image you linked, I guess it being an "input" makes sense why I was reading 0v on Sig (Green). The 5v wire is *not* hooked up between the controller and the CA, but the signal (from CADP connector) and ground wires *are*, so I believe my wiring is correct. My CA is set to High, I updated my shunt resistance today to 0.250mOhm and tried having my throttle settings set to "by-pass" as well "pass-thru" but still couldn't get it to work that way.

Today I tested the green throttle out wire on CA3 CA-DP 6pin connector while it was connected to my sabvoton controller, I was moving the throttle position and was seeing a voltage output on my CA Diagnostics screen but was reading 0v on the green throttle out wire.

I unplugged my CA's throttle signal wire from controller and tested again to see if maybe something in the Sabvoton's input is loading down the signal and still read 0v on the green.

At this point I'm assuming there's no signal out from the CA so I opened it up and can confirm the throttle wire and CA-DP 6 pin Green wire are wired to the correct locations. I tested the wires for continuity just to make sure they were good right up to the board and all was well.

I did not get the chance to test the voltage directly from the pad as I couldn't have power to the CA from the table I was working on it from. I ended up putting it back together and to my luck my CA just flashes on and off when powered up. I'm going to open it back up tomorrow to investigate further more and report back.

Do you think its my CA that's defective ? I have 2 more CA's laying around at the workshop I picked up second hand for cheap, both worked fine until I tried updating them from v2.4 to v3.14. Now they just flash on and off just like this one. Cant figure out how to get those working either, any thoughts ? Should I just buy a new CA3 from grin ? I'd really like to be able to have full control over my ebike.

I don't think the v2.x hardware is compatible with the v3.x firmware. You'd have to verify this with Grin Tech, and see if you can roll them back if so.

If the CA you've been using was v2.x hardware, I'm not sure there is an analog throttle out at all--you'd have to ask Grin Tech about that; IIRC it just has digital output that grounds the controller's throttle input (with actual throttle connected directly to the controller), whenever a limit is reached that causes the CA to begin limiting throttle. This is different from the CAv3 oepration, which uses an analog output to actually create a modified throttle signal based on the actual throttle input signal vs any settings/limits/etc taht are setup in the CA.

But if you don't have any voltage at the CA board inside (and/or the wiring is verified continouos from pad to end of wire and no voltage at wire without connection to controller), but the diag screen does show an output voltage on the diag screen, then something is definitely wrong.

Okay so I just started playing around with the CA3 on my bike and it seems I forgot to plug the Neg wire from 6pin CA-DP Connector to my sabvoton throttle when I was running test last night after putting my CA back together lol. Back to testing we go.

I just opened it back up to test for any voltage inside and I just want to make sure I'm doing this right. I located the 6 pin CA-DP Wires on the CA board and tested voltage between Green and Black wire which read 0v. I also tested voltage for Throttle In on CA board, Green and Black wire which was also 0v, but you said that's normal. Does this mean my CA is reading an output Throttle Signal but isn't actually able to output the signal ?

I've asked grin in the past if v2.x hardware can be updated to v3.x firmware and they told me it wasn't possible. Its strange though as one of the CA was a v2.4 and the other was a CA3 running on V2.4 and both of them just power on and flash after updating them.

dagher58 said:

I just opened it back up to test for any voltage inside and I just want to make sure I'm doing this right. I located the 6 pin CA-DP Wires on the CA board and tested voltage between Green and Black wire which read 0v. I also tested voltage for Throttle In on CA board, Green and Black wire which was also 0v, but you said that's normal. Does this mean my CA is reading an output Throttle Signal but isn't actually able to output the signal ?

Click to expand...

It's not "reading" an output throttle signal, it's *generating* that signal, and the readout you get on the diag screen is what the MCU is sending to the output hardware.

If it is v2.x hardware (regardless of firmware version), it can't output that signal because it doesn't have the stuff to do it, so that would be why you get 0V.

If it is 3.x hardware, it should output that signal if the hardware is working and it is running the correct version of 3.x firmware.


The throttle in will only read a voltage if you have a throttle connected to the CA's throttle input connector, for all three wires (5v, signal, and ground), and that throttle is outputting a voltage. If this is occuring, then the CA will *also* show that throttle in voltage on the diagnostic screen's throttle input voltage reading (to the left of the throttle output voltage reading on the same screen).

If the CA reads a voltage there, but you can't read a voltage with your meter, something is wrong with the meter, the leads, or the connection, or the meter setup wrong or is being used to probe the wrong points (either signal or ground lead is in the wrong place).

I've asked grin in the past if v2.x hardware can be updated to v3.x firmware and they told me it wasn't possible.

Click to expand...

Just curious: if they said it wasn't possible...why do it? :?

Its strange though as one of the CA was a v2.4 and the other was a CA3 running on V2.4 and both of them just power on and flash after updating them.

Click to expand...

It's not strange if the firmware is incompatible, because it means the computer (MCU) inside can't boot and run the program, so they can't do anything "normal" anymore.

AFAIK they are different hardware and can't use the same firmware. If you can't reflash them to the correct firmware with the same method used to flash them with the incorrect firmware, you'd have to send them to Grin Tech for them to do it directly to the MCU chip (if they can).

I retested voltage directly from the pad where the 6 pin CA-DP wire connects to the CA board and read 0v again, I reached out to grin directly and it seems I have finally located the issue. After updating them with where I was at and linking them to this forum for more info they replied telling me

"If you are not getting a throttle out signal from the CA that certainly indicates a CA issue. If the display registers an output signal, but it doesn't reach the output pad, the most likely culprit is a little 270ohm resistor (r18) installed to protect the opamp from high voltage on the signal line."

Turns out it was a resistor causing the issue and ill be sending my device in for an repair. Thank you soo much Amberwolf and Santacruz for reaching out and helping me get to the bottom of this once and for all. I cant wait to get this thing running at its full potential


Also I didn't know it wasn't possible to update v2.x to v3.x until I did it lol, ill reach out to grin to see if maybe I can send it in with my CA3 to have the MCU chip reflashed. Wish me luck

Glad you have found the problem.