Lyen controller dead a 2nd time..or is it a LVC thing

[NeilP]

Wonder it anyone has any ideas where i shoudl start looking.

Here is the problem
I got a 65 amp Lyen 18 FET controller for my bike, and it ran for about a week before a couple of the output fets shorted out and it would barely roll.
I sent it back to Lyen and got Maxwel65 to send me what ever he had in stock. So I got a 40Amp 72 Volt Crystalyte contoller. taht was in April and all has been well ever since.

Lyen repaired my 65 amp unit and it came back last week..
I fitted it today, went out for a gentle test ride...and noticed the 3 speed switch was not making any real difference

I re programmed it to 100% 50% and 30% and went out for another ridall seemed fine so went for a high speed run...all fine...till no power...then back again after a few seconds.

This continued for a while and now then it failed to come back on...CA is still working but nothing else...No lights come on on the Cruise Control.

The battery was fully charged up to 83 volt, and was only down to 80 when the issue occured.

At that point I gave up..as I knew the next repair tool I reached for would have been a large lump hammer and the controller would have been on the receiving end, and I would have been back to the Crystalyte 40 amp unit again



Just had a thought...

i did change the settings for the controller LVC to reflect the R12 regen braking mod that Lyen did...but did I go the correct way...pretty sure i did..as i had Lyens instruction in front of me.

he has set the R12 resistor to be 400 Ohm and he said to use this to calculate the new LVC

Program LVC = (real LVC + 2) x 0.75.
So i want 70V LVC I set the controller program to 54V ...

No i cant have got that wrong...all i can think of is that maybe it reverted to the old settings ...so it sent 70 and not 54...no that would give 91 volt LVC..so it t would not have gone at all

But if it was LVC issue...would that stop the cruise control lights from coming on?

 

[FMB42]

I'd double check every electrical connection followed by re-programming the controller (while paying extra attention to your regen settings).

Otherwise, you may want to look over this ES thread:

http://endless-sphere.com/forums/viewtopic.php?f=2&t=27898

Hang in there, you'll get it sorted out...

 

[gensem]

Thats odd... Lyens controllers use to be reliable. Just set the LVC to something verylow to checkout and keep an eye in the CA. Btw do you happen to have any real range (>5%) gains using regen?

 

[itchynackers]

Not sure about the cruise light issue, but I'd recommend starting your LVC setting lower than you think it should be. Then run the battery down to the lowest you are comfortable with. If the LVC doesn't kick in, then re-program, bumping the programmed LVC up a volt or so. It is an iterative process, but the only real way to determine what the actual LVC will be. I've also found the supposed formula doesn't work exactly as it thought to.

 

[NeilP]

Btw do you happen to have any real range (>5%) gains using regen?

Difficult to tell with less than half a mile riddend before it died...and only a weeks riding the before the first time it died.

The regen did not work at first because the R12 mod had not been done..Lyen sold me the regn brake levers...and knew I was using 84 volt...but forgot to mention that that regen would not work at that voltage.


I did check the connections last night, briefly, then gave up after it still did not work.

I do have a copy of the new XPD programming software...maybe I'll take a copy of that to the workshop computer ( no internet in workshop) and give XPD a try.

lets just hope it is the LVC...

 

[gensem]

Go for the XPD, its much better!!! I have it in all my computers. Drop a line about the watts you got back when the setup works again.

 

[NeilP]

will do.
It is now time 0400hrs here...I could not sleep...my mind kept thinking things over, so i got up to check the e-mails as my eyes could not see the ipod screen well enough at this time of the morning.
Going to try and get a couple more hrs sleep before it gets too light

On one run I did see a reverse current of 13 amps at on brief point, and that was on the slip/throttle braking as I have not got a brake lever connected up yet..think it will have to be a switch till I can find a reed switch and magnet to stick to the hydraulic brake levers

 

[NeilP]

re-programming the controller (while paying extra attention to your regen settings).

I'll have to have a look, but something may have slipped through. I even got a calculator out before hand to wok out the figures I need

Now this is 4 in the morning and from memory, but I am sure this is what i set
Lyen told me the effective new value of R12 is 400 ohm...so a 33% change
Here is what I got from Ed

Program LVC = real (LVC + 2) x 0.75.

For example, real LVC for 24S lipo is 67.2 (24x2.8), then program LVC = (67.2+2)x 0.75 = 51.9

I have 20 series and want 70V LVC

So that is 70 + 2 = 72
72 x 0.75 =54 v

so set 54 as LVC


Program regen = (required regen voltage/ 1.33).

so in my case for 83 volt regen,
So I set 62.4 or whatever the program allows for the regen voltage

 

[FMB42]

Hey Neil, you might want to check the CA LVC setting just in case...

I would also consider checking the max amp draw settings on both the controller and CA.

Another thing to consider would be an error in Lyen's regen/LVC calculation formula (I've heard that this formula doesn't always factor out correctly). Further research on this could help to shed some light on this issue.

You could, as a last resort, check the "real world" LVC by lowering the controller and CA LVC setting(s) by a volt or 2 at a time, followed by a short test run from a fully charged battery (this will, of course, require that you carefully monitor the battery voltage during these tests so as to avoid over-discharging the battery).

 

[NeilP]

I found the issue. it was the LVC setting.

His figures are way off as was what he told me the new value of R12 was....well I say that,,,but that is of course that in measuring the new vlue myself I have not made a basic error.

This is the question I have in my mid

Can the R12 resistor ( or combined value with new parallel resistor) be measured in circuit? or do other components interfere with the reading.

If it can be measured in circuit, then Lyen was totally off. he told me 400Ω when the measured value ( in circuit ) is 792Ω...
but without knowing what the original value is I can't work out the ratio of change.

From what I have read some of them were 1200 some 2200Ω.




My board is version EB218 AS -1

I'll just have to wait for someone who actually knows what the original value is..and if it can be measured in circuit or not. Then I will be able to do some meaning calculations.

I have also posted here
http://endless-sphere.com/forums/posting.php?mode=edit&f=2&p=439665 and re uploaded Hyenas calculator spreadsheet, as the old link seems broken

 

[wesnewell]

Measuring a resistor in circuit can never get you a higher reading. It is possible to get a lower reading though if other components are part of that circuit. Can't you read the code on it?
http://www.hobby-hour.com/electronics/smdcalc.php

 

[NeilP]

No, because one is a SMD ( the original) and the other is insulated over with heat shrink.
I could rip it off. but would rather not if I can avoid it