Ca3 dps in legacy mode limit problems

Djhandz

100 mW
Joined
Sep 14, 2017
Messages
49
Hi guys could do with a little help.
I am trying to use limits on the cycle analyst v3 dps
But it don't seem to be working.
I am seeing the limit flags flash on the ca when limiting and I am also seeing the tho voltage come down in voltage to override my throttle but it does not seem to have an affect.
It's is like the controller is ignoring the throttle out from the cycle analyst

Setup
Cav3
Ca3 3 pos switch (presets)
Xie chang/infinion 18fet (still confused what these are called my first ebike)
Mxus 3k 4t
100v 24s battery

Wiring scheme is option c from the ca manual
My throttle is wired straight to 18fet controller
Ca direct plug plugged to 18fet controller

Settings
Ca3 dps (modified tho moved to thd)
Throttle out control mode off wot
Throttle out voltage 1.2v to 3.8v

Xpd infineon 4 settings
Block time/overcurrent detection tried 0 and 0.1

Things I noticed when briefly testing
Limiting not working
Not getting even half the power I had before trying this setup or my 3 speed switch(infinion) is no longer working and it's stuck in pos 2 (low power)

I did not have my speed sensor on the fork while testing not sure if this would make a difference

The infinion 3 speed switch is just there from old setup i won't be keeping it if I get this sorted

Possible theory. I messed up soldering the green wire tho to thd(I think I can test this with a voltmeter at the ca direct plug)

Any help would be much appreciated thanks guys
 
Many details. Good!
Speed sensor makes no difference in this case, so that part is okay.

(1)
I'm not sure I understand what you were saying about the 3-speed controller switch, but you need to unplug the CA and verify the controller is 100% okay before moving forward - otherwise too many variables. It sounds like you are saying you only have half power, the CA limiting looks like it's working by the CA display, but there is no perceptible varying limiting effect on the controller. Is that correct?

This sounds like the middle wire of the three speed switch may have broken. These controllers default to Spd#2 if the switch is disconnected. Since you can program it, you can just set all the speeds to 100%...

(2)
Are you sure you have a 'Large Screen Compatible' controller"?
That is, was it labelled so or did you do the test on page 15 (4.4.2.1.1 Direct Connect: Controllers with CA Interface Connectors")?
If you have an old legacy controller and also made the ThO to ThD diode change, the CA would need to reduce ThrOUT about 1.5V below you controller max ThrIN before the limiting had any effect (so maybe down to 2.5V or so).

(3)
Your ThrO->Min/Max sound okay, but it's pretty harmless to set them the limits for legacy mode (0-4.99V respectively) at least to get started. This will pretty much ensure there is no unintended limiting effect. By 'unintended, we mean that the CA doesn't think it's limiting but the max voltage is incorrect and it's causing limiting even though the CA isn't trying (so - limiting w/o Limit Flags).
 
teklektik said:
Many details. Good!
Speed sensor makes no difference in this case, so that part is okay.

(1)
I'm not sure I understand what you were saying about the 3-speed controller switch, but you need to unplug the CA and verify the controller is 100% okay before moving forward - otherwise too many variables. It sounds like you are saying you only have half power, the CA limiting looks like it's working by the CA display, but there is no perceptible varying limiting effect on the controller. Is that correct?

Yes that is correct

This sounds like the middle wire of the three speed switch may have broken. These controllers default to Spd#2 if the switch is disconnected. Since you can program it, you can just set all the speeds to 100%...

Yes the 3 speed controller switch no longer does anything so it's defaulted to speed 2 which is set for low power. Yes I will program all speeds for 100% so this may not be a problem but what I will say it was working fine before I made these changes

(2)
Are you sure you have a 'Large Screen Compatible' controller"?
That is, was it labelled so or did you do the test on page 15 (4.4.2.1.1 Direct Connect: Controllers with CA Interface Connectors")?
If you have an old legacy controller and also made the ThO to ThD diode change, the CA would need to reduce ThrOUT about 1.5V below you controller max ThrIN before the limiting had any effect (so maybe down to 2.5V or so).

Yes I did the test on page 15 to determine if large screen compatible. (Throttle in controller not ca throttle control off wot then thro to 2v exit setup and wheel spun)

When I was testing I seen the thro go all the way down to 1.3v but had no effect

The controller looks like identical to these
https://youtu.be/t2g7voSMGEM
It has no label on it

(3)
Your ThrO->Min/Max sound okay, but it's pretty harmless to set them the limits for legacy mode (0-4.99V respectively) at least to get started. This will pretty much ensure there is no unintended limiting effect. By 'unintended, we mean that the CA doesn't think it's limiting but the max voltage is incorrect and it's causing limiting even though the CA isn't trying (so - limiting w/o Limit Flags).
I will chance to 0 to 4.99v

Sorry I am not sure how to change colour of writing :D
 
Quick update

Ca unplugged controller works as expected
Changed voltage to 0 to 4.99 still no change

Would testing the voltage at the thro ca direct plug confirm I have swapped the wires in the ca correctly?

Many thanks for your help much appreciated
 
Another update.
I have now tested the voltage at the ca direct plug across the neg pin and green thr pin. I am seeing 0.8v closed throttle and voltage rises as I open throttle.
I expected to results the other round as in 4.9v closed thr then drop in voltage as i open thr and the ca attemps to limit. That would match what I am seeing on the ca diagnosis screen

Does this mean I have done something wrong when swapping the green wire in the ca?

As far as I can remember the green wire I moved was next to an empty pad which I am sure was marked thrd and thats the pad i moved it to.I maybe wrong
This was on the side of the unit
 
Djhandz said:
Another update.
I have now tested the voltage at the ca direct plug across the neg pin and green thr pin. I am seeing 0.8v closed throttle and voltage rises as I open throttle.
Okay - that is correct.
However, since your power is being limited, we are concerned about the voltage at full throttle as well.

It would be good to unplug the CA and measure the controller CA-DP pin 6 (ThO) voltages. Then plug in the CA and repeat. Since you are being limited somehow, we would expect the wide open throttle reading to be lower in the second test - specifically, the throttle voltage should rise smoothly exactly as it did in the first no-CA test and then stop rising when a limit is hit.

Djhandz said:
I expected to results the other round as in 4.9v closed thr then drop in voltage as i open thr and the ca attempts to limit. That would match what I am seeing on the ca diagnosis screen
Not at all. In legacy mode, the CA3 works like a CA2. It does exactly nothing to the user throttle signal (which you see on the CA-DP connector coming out of the controller) until a limit is reached. The purpose of the diode is to make the CA and the operator throttle signals play a game of "The lowest guy wins!". This is why we set the CA ThrO->MaxOut to 4.99V - the operator throttle will always be lower and will always 'win' (i.e. the operator throttle will work normally, no matter what voltage it puts out). When limiting kicks in, the CA lowers its signal until it is lower than the operator throttle and the CA 'wins' and prevents the controller from seeing any operator throttle voltage above that of the CA. (As a detail, there's a fixed voltage drop across the diode, but that's not important for the explanation - or in practice.)

Your expectation of the CA_DP pin tracking the OUT voltage on the Diagnostic screen is what happens with the CA in normal mode (i.e. without moving the wire to the diode connection). In that case, the CA drives the controller instead of the operator throttle so you would see the voltage go roughly between 0.8-4.0V on that pin as you opened the throttle (the throttle being connected to the CA).

Djhandz said:
Does this mean I have done something wrong when swapping the green wire in the ca?
From the one ZERO throttle voltage you measured and provided, it looks okay. We need to look at the voltages when the operator throttle is higher as mentioned above try to figure out (1) why you only reach partial power and (2) why the programmed limiting isn't working.

Obviously, opening the CA and checking the soldering and connection is a quick way to get some questions off the table, but one thing you might test is to change ThrI->CtrlMode to Off(ZERO). This should send Throttle OUT to 0V which should make your operator throttle completely inactive. This is essentially 'super limiting'. If your operator throttle still works, there is clearly something electrically wrong and you will need to open the CA, check the wiring, and perhaps measure the voltage on the ThD pad.

  • As an aside - it's unclear why you are trying to run in legacy mode which disables so much functionality. At the very least, I would always recommend running in normal mode with a PassThru throttle. The throttle will essentially work the same as in legacy mode, but you can get throttle ramping and can use the CA to match throttle and controller voltages to eliminate throttle dead zones. Those points alone make the CA3 a winner over the CA2. I guess you have considered this to go thru the trouble of messing with the factory connection, etc, but it's an unusual strategy. So - not trying too hard to dissuade you, but...
 
First off I would just like to say a big thankyou for taking the time to help me and give such detailed answers that I understand

It would be good to unplug the CA and measure the controller CA-DP pin 6 (ThO) voltages. Then plug in the CA and repeat.

I have just done this and I am getting the same results each time .8v closed throttle and 3.5v full throttle without ca plugged in and 3.5v with ca plugged in and limit flags flashing

Obviously, opening the CA and checking the soldering and connection is a quick way to get some questions off the table, but one thing you might test is to change ThrI->CtrlMode to Off(ZERO). This should send Throttle OUT to 0V which should make your operator throttle completely inactive. This is essentially 'super limiting'. If your operator throttle still works, there is clearly something electrically wrong and you will need to open the CA, check the wiring, and perhaps measure the voltage on the ThD pad.

I also just did this and my operator throttle worked. Just to be clear on all test throttle connected to controller not ca

I will take ca apart and upload a pic and add it to this

As an aside - it's unclear why you are trying to run in legacy mode which disables so much functionality. At the very least, I would always recommend running in normal mode with a PassThru throttle. The throttle will essentially work the same as in legacy mode, but you can get throttle ramping and can use the CA to match throttle and controller voltages to eliminate throttle dead zones. Those points alone make the CA3 a winner over the CA2. I guess you have considered this to go thru the trouble of messing with the factory connection, etc, but it's an unusual strategy. So - not trying too hard to dissuade you, but...

I have tried using limits in normal mode and although I did have it working I wasn't happy with the feel so thought I would give this method a try

Edit pic uploaded and tested continuity and it's good

Second edit. Struggling to upload a pic it goes though process but no pic
 
Do I need to add a diode or something or is It built in to the Cav3.

Proper confused with it all. It's well over my head
 
Djhandz said:
I have just done this and I am getting the same results each time .8v closed throttle and 3.5v full throttle without ca plugged in and 3.5v with ca plugged in and limit flags flashing
Okay. This is strange. The limiting is not having the desired effect on the voltage but I assume the controller power is still limited. This is looking electrical...

Djhandz said:
I also just did this and my operator throttle worked. Just to be clear on all test throttle connected to controller not ca
This looks like there is a wiring error. The OUT value on the Diagnostic Screen should be 0V and as you reported in other cases, this is having no effect on the operator throttle. Not Right.

Djhandz said:
I have tried using limits in normal mode and although I did have it working I wasn't happy with the feel so thought I would give this method a try
To be honest, it might have been good to post up about the 'feel' problem. Slow or surgey response isn't unusual out-of-box and can be tuned away.
 
Djhandz said:
Do I need to add a diode or something or is It built in to the Cav3.
The diode is internal to the CA. You put it in play when you moved the green wire from ThO to ThD (D is for 'diode ;))

Open the CA and check that you did not accidentally attach the green wire to the "G" GND pad next to the ThD pad.



CA3_legacyMode.png
 
I think this is correct Screenshot_2018-02-17-16-33-04-1-1.png

My head is done in
What am I doing wrong ???

Thankyou teklektik
 
Yep - that looks correct.
This is puzzling. I need to think about this.

What firmware are you running (see splash screen on power up)?
I'll load it up and verify that legacy mode works properly....
 
okay - 3.03 is a production release and works fine. No problem there.

It really looks to me like the diode is faulty and the CA is essentially disconnected from the controller when the green wire is tied to the ThD pad.

Except - there is this business about not being able to reach full power when the CA is plugged in. Can you describe this a bit?

First, check the continuity between the CA ThD PCB pad and the CA_DP pin 6. If there is none, then this is the problem.

Otherwise, I would recommend backing out pins 5 and 6 on the controller CA_DP connector. This will leave just the power and shunt connections to the CA. Hopefully, this will allow the motor to reach full power. Here's a youtube video on how to remove JST pins: https://www.youtube.com/watch?v=kRvDgiX4Sk4. The technique for the female part is in the second half of the video. So - give this a try and see if the controller/motor work properly/normally.

While you have the pins out, try setting the CA for OFF(WOT) and OFF(ZERO). Each time, measure the voltage on CA-DP pin 6 - probe the back of the CA side of the connector or the ThD pad on the PCB. Without the controller connected we should see the voltage the CA is trying to apply. Here we expect to get the 4.99V and 0V for OFF(WOT) and OFF(ZERO). If these don't track the expected voltages then either the diode is faulty.
 
Thankyou mate I will perform these tests and report back. One of the tests I have already done is checking continuity From the thd pcb pad to the ca dp pin 6 and it checks out ok. Update to follow thanks again
 
Ok got some more info
First off I have to put my hand up for giving incorrect information I am so sorry
The ca is not limiting the controller in Any way I had my controller 3 speed switch plugged in the wrong way round. Once changed Everything is working as it should with the ca plugged in. Sorry about that.

OK limiting in legacy mode is still not working. I have done some more testing

I removed the green/throttle wire from the controller Ca dp
I then tested the voltages of both controller throttle and ca throttle(both at the ca dp)

Controller throttle was reading .8v closed and 3.5v full throttle voltage. Readings was smooth on the voltage readout

Next was pin 6 on the ca dp. Having moved the green wire in the ca this would be testing thd
Closed throttle I was seeing 4v on the tested and as I opened the throttle and the ca started to limit this reading dropped instantly to 0V and below I was seeing minus 2v but the reading was all over the place. I lowered my limit gains all the way down to see if it would make tracking the voltage easier to read but it didn't. The reading goes from 4 then when limiting down to zero but jumps about in the minus numbers(not getting a steady reading) I don't know how diodes work maybe this is normal I don't know hopefully master tek will help shed some light on things
 
Djhandz said:
Ok got some more info
First off I have to put my hand up for giving incorrect information I am so sorry
The ca is not limiting the controller in Any way I had my controller 3 speed switch plugged in the wrong way round. Once changed Everything is working as it should with the ca plugged in. Sorry about that.
That's good news. That was really confusing the problem...

Djhandz said:
I removed the green/throttle wire from the controller Ca dp
I then tested the voltages of both controller throttle and ca throttle(both at the ca dp)

Controller throttle was reading .8v closed and 3.5v full throttle voltage. Readings was smooth on the voltage readout
okay - this is good and sort of what we expected from previous reading - so the controller side looks like it's working okay.


Djhandz said:
Next was pin 6 on the ca dp. Having moved the green wire in the ca this would be testing thd
Closed throttle I was seeing 4v on the tested and as I opened the throttle and the ca started to limit this reading dropped instantly to 0V and below I was seeing minus 2v but the reading was all over the place. I lowered my limit gains all the way down to see if it would make tracking the voltage easier to read but it didn't. The reading goes from 4 then when limiting down to zero but jumps about in the minus numbers(not getting a steady reading)
To be honest, we really need a pullup resistor to +5V on there to get an accurate reading, but these wacky numbers don't sound unreasonable without it and it looks like the CA is trying to limit.

Darn, but this looks like it should be working - it looks like the CA is just not managing to pull down the operator throttle on the controller side. Justin has seen all manner of weird failure modes and normally I'd send you off to email him - but I'm going to be meeting with him tomorrow so I'll ask if he has some thoughts to pursue that I can pass along second hand.
 
ok I borrowed a friends voltage tester just to vary my test. The voltage at thd is 4v closed and 0 when throttle open when ca is limiting the readings are steady.
I am now confident that the ca and all wiring is correct but as teklektik says the lowest guy is just not winning

teklektik post.. until a limit is reached. The purpose of the diode is to make the CA and the operator throttle signals play a game of "The lowest guy wins!". This is why we set the CA ThrO->MaxOut to 4.99V - the operator throttle will always be lower and will always 'win' (i.e. the operator throttle will work normally, no matter what voltage it puts out

Is there any xpd settings that I may have missed
 
Right I got some important info.
I just been out on the bike(this is going to be hard to explain)
What I can feel when very throttle is LIMITING (horay) but it's like a pulsing effect. As i open throttle more I no longer feel the limiting/pulsing.

I don't know if it stops when I open throttle or I just can't feel it because of more power

Hopefully you thus info will help shed light

Could these be symptoms of getting large screen display and v2 compatible mixed up and I need to choose the other wiring scheme ??

Thanks for hanging in there
 
I had a chat with Justin about you problem and he allowed that he had seen similar issues when the CA_DP connector on the controller was not wired properly. I actually hadn't seen a wiring issue like that so never really considered it, but it would explain what you are seeing.

So - where did you get the controller? If it's from EM3EV or Lyen then it's certain to be correct. If from another source, they may not have followed the guidelines for the internal hookup.

A common wiring error is pretty to test. With the controller switched off, use your meter to measure the resistance from the operator throttle connector 'Signal ' pin to the CA-DP pin 6. We expect to see at least 1K - maybe a bit more. The suspicion is that it's actually zero which would account for why the CA could not override the throttle.
 
I got the controller from ebikewarehouse.eu here in uk. It looks identical to the one I linked in my second post. I will do the test. Thanks
 
Ok I think you might be right about the wiring

Just to be clear I tested the pins you asked me to with everything connected but not powered up

I am not that great with a multimeter so please bare with me

I tried testing resistance first but was not getting a reading the readout stayed at 1 all the time no matter what range I had the multimeter on

So I then tested continuity(0ohm) and I got a clear buzzing with a reading of .3

As I am unsure please see this vid around the 3 min mark this is what is telling me I have zero resistance
https://youtu.be/bF3OyQ3HwfU

So if this is the case and my controller is not wired correctly. Is there a fix?
And please could you tell me what kind of affect this would have running the ca v3 in normal mode using limits(just wondering if this is why I was getting strange throttle issues in normal mode)

I really appreciate your help in this matter and thanks for talking to Jason about this issue for me.
I look forward to your response
 
Djhandz said:
I am not that great with a multimeter so please bare with me

I tried testing resistance first but was not getting a reading the readout stayed at 1 all the time no matter what range I had the multimeter on
No problem - they are just another kind of hammer you get better at swinging with practice... :D

We'd expect the reading to show something on one of the settings. Can you verify that you have a newish healthy battery and maybe post up a shot or what kind of meter you have. It might be something else entirely, but since since you're not super comfortable with the meter, this is just making sure all is on the up and up.

Djhandz said:
So I then tested continuity(0ohm) and I got a clear buzzing with a reading of .3

As I am unsure please see this vid around the 3 min mark this is what is telling me I have zero resistance
https://youtu.be/bF3OyQ3HwfU
Sounds like you did the test properly, but the continuity reading can be a little misleading - the resistance was really the key. That said, continuity/diodeTest could still tell us if there is diode in there. Please try the continuity/diode test again, and try switching the leads red-for-black. If you still get a beep then there's no diode (which would be good).

Djhandz said:
So if this is the case and my controller is not wired correctly. Is there a fix?
And please could you tell me what kind of affect this would have running the ca v3 in normal mode using limits(just wondering if this is why I was getting strange throttle issues in normal mode)
Sure. We can fix this either way you want without opening the controller.

Here's the thing - if we can't get a good answer on the resistance thing, then we don't really know what was done in there. This makes it pretty sketchy to guess about what normal mode would do or if it is misbehaving, but we could give it a try as a first cut to avoid any hardware mods.

In any case. to make normal mode work with 100% certainty, all you need to do it make the change as shown in case 1 or case 3 below (assuming you moved the green wire inside the CA back to ThO). Case 2 would also work fine if you have a resistor on hand.

To make legacy mode work with 100% certainty, you would make the change shown in case 2 (assuming you left the green wire inside the CA on Thd).

As you can see, all we're doing is bypassing the internal CA_DP throttle stuff with a bit of external modification that we know does the proper stuff. Case 2 is actually what we normally expect to see inside the controller, but they may have done case 3 instead. From the remarks above, you can see that that would allow normal mode to work but not legacy mode. Sort of makes sense when you see the pictures even if the electronics reasoning is unclear...

These mods can be done inside the controller, but it's not necessary. Sort of depends on your skill level.


CA3_interfaceOptions.png
 
Ok I did the continuity test in both directions and got the same results so I went with case 2 with one of these resistors
https://www.maplin.co.uk/p/metal-film-06w-1k-ohm-resistor-m1k
And yes it works the ca is now overriding the operator throttle. Nice
So how does this mode work?
I am getting surging when it limits it's like a on off on off there is no inbetween is this normal ? I had a little mess with the wgain but didn't really notice a big difference

Can I set it so for instance 250watt and receive that at the wheel constantly instead off this on off thingie Can it be smoothed out?
 
Back
Top