Need help troubleshooting 9C / Lyen 12 FET

[GMUseless]

I'm sure this has been covered a thousand times...but I need help troubleshooting a 9C / Lyen build.

I've got a 9C 2810 connected up to a Lyen 12 FET, EB212 4110 controller. I bought the controlled used, and all of the standard connectors had been cut off...so I had to re-wire all of the headers. And, the phase wires were so garbled I had to pull them completely and re-solder new ones to the PCB. I ohmed everything to make sure I didn't leave any bridges or cold joints. Anyway, everything seems to check out ok, but I can't seem to get a good forward motion.

When I try the standard 'swap blue and green' wiring on the hall and phase at 48volts, I get a very slow forward motion. It's smooth, but like 60 rpms or so. The other combo's give reverse, stuttering, or lock ups. The one that does run reverse...seems to run very fast compared to the slow froward one.

I was working on the impression that every hall combo should have one could phase combo...so I've only tried the six phase combinations. Do I need to try other hall combos? Do some of these 6 valid hall / phase combo's in fact run slow?

The throttle out of the controller is 5 volts...and the throttle is returrning a good signal from what looked like 0.75 to 4.4 volts (it was hard to tell since I tested it all with two hands...didn't have any clips handy.)

Being used, I wasnt sure what program was in it, so I reprogramed it using XPD. I set the Hall angle to AUTO. Could this be a problem?

Any ideas on troubleshooting this one?

 

[amberwolf]

He might have a more updated version elsewhere, but this is John in CR's method that seems to work for many m/c combos:
http://www.endless-sphere.com/w/index.php/Determining_the_Wiring_for_a_Brushless_Motor

Since it runs faster in the reverse combo than the forward one, you might also want to ensure that the X3 pad isn't jumpered to ground (it is typically used to engage reverse, and reverse is usually a lot slower than forward).

What is the no-load current for the various combos? You should get one that's an amp or two *at most* with full throttle smooth-spinning wheel. More than that and it's probably a false positive.

 

[GMUseless]

Thanks man.

This I'll try switching the halls around and retrying the six phase combos for each. If that doesn't work, I'll open the controller to confirm that X3 pad didn't get bridged. I did solder in the 3spd switch myself, so I might have bridged it without noticing.

That said, the three speed switch, does seem to change the speed of the wheel...it just stays super slow...so I'll get something like 30 rpm 60rpm and 70 rpm.

 

[pwbset]

I just went through this the other day with my used 9fet lyen controller. I finally found the right combo just by process of elimination. Maybe my color scheme will help you out as well.

I'm running 18s at 75v hot and I'm seeing around 25mph max out of my rear 2810. I thought I'd get more speed than that even knowing the 2810 is a slow wind, but it'll be enough. Nice and torquey though which is more what I need anyway for hill climbing. Let me know what speeds you're seeing when you get it wired up right. Thanks.


http://endless-sphere.com/forums/viewtopic.php?f=6&t=39348#p586477

 

[GMUseless]

Thanks. Reviewing your pics, that's the combo I started with, which is basically Method's rule of 'swapping blue and green'.

So tonight I swapped the halls and the phase and tried all 36 combos: 3 of which ran forward like they should...all slow. I pulled the controller and checked for any solder bridges, and everything looked clean. Also, DX3 is not jumpered out.

I put it all back together and got some jumber pins from the breadboard to shove into the backs of the little JST connectors at the controller connections so I could measure some things. I then noticed some problems with the throttle.

Power out to the throttle is 4.41 volts. But the signal returning at full throttle is only 1.6 volts. I had tested it before installing...but the lowest source I had on the bench at the time was 9 volts...and I didn't take the time to put together a divider...so I just tested it against the 9v source. I'm wondering if that might have damaged it? Time to pull it apart....

 

[pwbset]

Power out to the throttle is 4.41 volts. But the signal returning at full throttle is only 1.6 volts.

Hmm... this makes me wonder if I'm getting full power out of my half twist throttle also. I got 4.4v to the throttle, but I never measured output coming back. I feel like ~24-25mph is tad slow for 75v on a 2810, but maybe I'm wrong. Wish the ebike.ca simulator had a 2810 in it.

 

[GMUseless]

I know Dogman has extensive experience with 2810. You might try asking him for some feedback.

 

[pwbset]

I know Dogman has extensive experience with 2810. You might try asking him for some feedback.

Thx GM... I've read most of his threads on the 2810 already, but don't recall MPHs per voltage level so will search around some more. Good idea. I can always wire up more in series if I get a different charger. I'm sure at 24s the 2810 would be really nice.

 

[pwbset]

Tommy L is seeing:

76.8v 24wh/km @ WOT 45kph 1000-1500 watts (battery pack 20lbs)

About 28mph. I'm seeing 24ish at 75v and I weigh about 30lbs more than Tommy L so that sounds about right. Oh well. I'll see how it does climbing 8-10% for 3.5mi and go from there. Can always go 24s & oil bath cooling and/or just get a damn Cromotor. 8) :lol:

 

[GMUseless]

Likely better just to get another motor, wound for speed. My MAC 6t does 35mph on 12s LiPo. I got the 2810 for a low speed cargo bike..I'll be happy with 20mph on it.

Just have to sort out the throttle....

 

[999zip999]

Sent my lyen 12 fet out for repair will run with 2810. My 2810 with a crstalyte analog 40a turned to 25amps with a 18s konion pack runs 27mph at 74.3v off the charger. I peddle down pch 48t-11t wot up to 31mph. Up hills 14-21mph. If peddling over 26mph I use maybe 600watts and as low as 36wh/mi. But it still makes that wrine eeearr. When I get my lyen back I will run it at 24s up to 40amp with A123 20ah to get some more range and speeed for that 2810. Get a Lyen tester. Or I did the lyen test sheet every colum.

 

[pwbset]

My 2810 with a crstalyte analog 40a turned to 25amps with a 18s konion pack runs 27mph at 74.3v off the charger.

How much do you weigh? Any tail wind when you measured those readings? At 75.6v off the charger I'm only seeing about ~24mph on flat ground no wind in a "crouched over the handlebars" position with 2.1 mtn bike tires that are kinda knobby, but not a lot. I weigh around 215lbs though.

I've already decided I'm going to add 6s to go with 24s for a little more speed. Side benefit being I can run better within my controllers 30A limit and get an extra 200wh to boot while only adding 4lbs or so. Downside is I now need to order a couple 48v Meanwells in series for 100v so it's some extra money.

 

[999zip999]

I weigh 220lb my bike is a old comoly D.B. it has always rolled fast I think it's great japanese bearings. I got most all low cells out. Started with 500 cells and fighting for 200. How hot can I run this combo 24s A123 on a 2810 and 12 fet lyen without melting 40amps ? Plus ?

 

[999zip999]

Put the greens togethere and swap the other two

 

[GMUseless]

So I finally got back home tonight, and took apart the throttle. Everything looks to mechanically be in order. I'll try replacing the Hall when I get a chance.

Amberwolf (or anyone), I'm guessing those Honeywell SS41's we got will work?

In the meantime, I swapped the wuxing throttle for a Crystalyte one, and everything fired right up, less than 1 amp no load at 47 volts.

 

[amberwolf]

Amberwolf (or anyone), I'm guessing those Honeywell SS41's we got will work?

No, unfortunatley those are on/off types, and a throttle needs the ratiometric (analog) ones.

If you just hook up the hall in the throttle to 5V and ground, and move a magnet around near it, does it's output signal pin change voltage at all? (I forget if this kind needs a pull-up resistor or not; it'd have one internal to the controller if it does).

 

[GMUseless]

That makes sense as far as these not being ratiometric.

I had forgotten about the pull up resistor. I can almost gaurentee that's the problem. I had 4.4 volts leading to the throttle...and the signal back was between .8 volts and 1.6 volts.

It's a Lyen EB212 controller. I'll have to research what's required.

Do you know if that resistor is internal or something you need to add external?

 

[amberwolf]

If there even is a pullup resistor, it would already exist in the controller. I don't actually think it needs one but I can't remember. I'll have to go measure one of my throttle to see.


But given the voltage range you're getting, it sounds like the sensor is only responding to one magnet (there are usually two), and so only getting the first half of the range.