Changed fron SLA to LiFePO4 lost power

[oldbiker]

Hi, new guy here,
I have an electric tricycle that I have been riding for 4 years or so. I installed a brushless 36V 700W Motor Bicycle Kit and I use 36V 12AH SLA batteries, however, I purchased a LiFePO4 36V 20AH setup, and when I throttle up it I get a gruesome hum from the motor and the movement is extremely slow.
There was a light hum in the original setup but nowhere near as bad as it is now. The hum will subside, slightly, when I am pedaling but power is completely gone.
I fear something has burned out, a Hall sensor probably.
I am using the same controller as with the SLA batteries; as a matter of fact I get the same response when I changed back to SLA batteries. So until something is fixed the trike is unpowered, which is bad news for my weak old legs.

Any help will be appreciated.

 

[amberwolf]

What brand is the new pack? Do you know what it's "hot of the charger" voltage is?

It is possible that it's fully-charged voltage was higher than the controller could take.

 

[deardancer3]

Welcome to Ebikes/etrikes and to Endless sphere. sorry about your recent woes.

Can you give brands/description of controlle rand motor? Dealer purchased from?

Hopefullly someone here can help. What is your location?

BEst

d

 

[oldbiker]

Hi,
Thanks for your prompt replies.
The pack is marked as a V-Power issue. The voltage “hot” off the charger is 43.4V, and it is very possible that there is too much juice for the controller to handle.
It was my understanding that the controller could handle 41.4 volts. It seems a bit ridiculous that 2 volts would cause such spasms. Evidently the manufacturer did not allow for this slight overrun, if this is the cause of my problems.
Every thing is in Chinese so it is difficult for me to relay any information to anyone other than the basic data I have been given on the labels.
The motor kit was purchased from elephant on ebay, and the battery pack was purchased from cammy_cc, also on ebay. Now it looks like I will have to purchase a controller; probably on ebay.
I have little more than a basic understanding of electro-mechanics, and so prefer a DIY situation but I need to learn more regarding E-Bikes.

 

[swbluto]

It was my understanding that the controller could handle 41.4 volts. It seems a bit ridiculous that 2 volts would cause such spasms. Evidently the manufacturer did not allow for this slight overrun, if this is the cause of my problems.

For "absolute voltage ratings", there is no leniency whatsoever. There will be variation of a given part as far as the "absolute voltage rating", but you cannot exceed that voltage or you will blow the part. It's not really a time-duration issue as it is with current, it's just a definite no-no.

However, if the controller was designed for 36volts, it does seem a bit ridiculous that the absolute max would be in the low forties considering charge voltages and all. So, in that case, the manufacturer was pushing it just a little too close.

 

[amberwolf]

Agreed. If even SLA were charged up beyond 13.8V or so each (41.4V for 3), they'd pop the controller. If the controller was actually hooked up during charging to many SLA chargers that do trickle charge or float charge, you'd see 15V per battery, or 45V pack, which would do it in for sure if that 41.4V is really absolute max (probably is).

The surface charge (hot off the charger) voltage of lithium can be pretty high, especially after a balancing phase, although it typically will burn off quickly to bring it back to it's nominal fully charged voltage.

Even my NiMH pack would blow a controller with that low a max voltage, when it's fully charged, and definitely would do it when it's charging if hooked up to controller too.


My guess based on the voltage thing is all you need is a new controller, or to repair the one you have. It could be as simple as replacing an LM317 or similar regulator, or it could be as bad as the MOSFETs and other voltage-sensitive components (like the MCU) fried.

Either regulator or MOSFETs is technically easy to do, if you have the equipment and skills. If you have the tools to open it up, and can take good pics of it, as well as do a few measurements with a multimeter, we might be able to help you fix it on your own for a lot cheaper than a new controller.

If the MCU or small SMT components are damaged, it is likely to be cheaper/easier to get a new controller unless you're familiar with getting and replacing these types of parts. But it is not that likely that these parts are bad, unless the regulator itself failed shorted, allowing pack voltage to feed the low-voltage parts.

No way to know what's damaged till at least basic tests are done and possibly opened up for other tests. Is there anyone in your area you can plug your motor into their controller to test it? Or your controller into their motor?

 

[oldbiker]

Hi,
Thanks for your reply.
If I were to purchase a 48V controller would I be losing any power? This thing is no speed demon as it is/was, and I would hate for it to go any slower.
Or maybe someone has a way to fix what I have, and to upgrade it to 48v.

If anyone has a schematic of a 36V charger I would like to see it. I have been searching for a schematic for years. There are many 12v charger schematics, with float, available but at this time I am unable to find a way to alter one for 36v.

 

[oldbiker]

Hi,
Thanks amberwolf.
I can open it up to see if there is something I can fix but in the short term I’m going to have to get another controller so I can scoot around.
I’ll open it up, nose around, and maybe even attempt a fix if I find anything askew.
As for trying other parts; I do have another 36v controller hidden away in my parts vault somewhere. I will probably receive a new controller before I find the old one. 

 

[dnmun]

maybe before you open it you could test the controller to see if it is bad. see if there is power to the throttle and hall sensors and check to see if the output FETs have shorted out.

do you have a picture of your equipment or a voltmeter?

 

[oldbiker]

Hi,
I already ran a test, sort of; I reconnected my SLA batteries, and got the same response that I had with the LiFePO4s connected.
I do not have a camera that will properly show the details required.

 

[amberwolf]

If I were to purchase a 48V controller would I be losing any power? This thing is no speed demon as it is/was, and I would hate for it to go any slower.

I doubt that being able to run at a higher voltage would make it any slower; usually it is just the limits of what you can put into it that are affected, as well as the LVC for the battery (which is sometimes programmable one way or another).

Or maybe someone has a way to fix what I have, and to upgrade it to 48v.

Possibly it's the same kind as one that someone else has already modded for such here on ES, or close enough to copy the mods.

If anyone has a schematic of a 36V charger I would like to see it. I have been searching for a schematic for years. There are many 12v charger schematics, with float, available but at this time I am unable to find a way to alter one for 36v.

You'd first need to have enough voltage input to the regulation section to get 36V (45V, actually) out of it. Once you have that, it's likely that one of the existing designs can be adapted. If you have a preferred version of one of those 12V ones I can look at to see if it's modifiable, post it as an attachment or link to it on the web and I'll check.

Alternately, you could just build 3 isolated 12V chargers and hook them all up at the same time (kind of like single-cell lipo chargers are used to balance a pack). I've done that with two 24V powerchair chargers for use with 48V of SLA before.

 

[amberwolf]

I’ll open it up, nose around, and maybe even attempt a fix if I find anything askew.
As for trying other parts; I do have another 36v controller hidden away in my parts vault somewhere. I will probably receive a new controller before I find the old one. 

As dnmun said, check the stuff outside it first, but I do think it likely something in the controller fried from overvoltage; the timing of failure is just too convenient otherwise.

 

[Gow864]

Sounds like the halls got messed up, try checking the phase and hall connections. I've seen this behavior when the connection isn't quite right.

Gow.

 

[oldbiker]

Hi Guys,
Thanks for all of your replies but I'll need more time to take it all in. DO NOT STOP! its just beddy-by for me now.

 

[USAEBIKECO]

you definately blew the controller. I see your kit is a chinese one sold on ebay. From my own trial and error the controllers are designed for only their kits. The controllers cant take lifepo.

The good news is thats all you have to replace. But im not sure if a new controller from the same company is the solution. You might wanna invest in a better system.

I do like your trike!

 

[oldbiker]

Hi guys,
I opened up the controller, and as electronics go, there is no hint as to which device failed. I will start my process of elimination shortly; fun, fun, fun.

I plan on purchasing the Goldenmotor Cruise Speed Controller, item # BAC-0281P: http://www.goldenmotor.com/. I am waiting to hear from them on the particulars. From the specs it looks like it will handle my needs. The price is higher than I care for but what isn’t. Anybody have experience with this device? View attachment GoldenMotor Cruise Controller User Guide.pdf

Here is the schematic of a charger I would like to mod to 36V. It states that the output can be modified for 16V but there it stops.
It has minimal parts, which makes it ideal for my wallet since I have most of the parts required. I guess I could experiment to find the correct configuration but that would surely take its toll on my inventory.View attachment Automatic 12V Lead Acid Battery Charger.pdf

 

[amberwolf]

The major modification you'd need to that circuit will be that T1 has to have a secondary side output a bit above what your pack will be, instead of the 12V it is in that diagram.

Then you will need to figure out what resistances you need R1/2/3 to be at to give you the right voltage range for your pack's charged vs discharged states. Their actual values may not need to change, but their proportions might.

R6 may need to change value to ensure current thru the LED isn't too high.

The SCR, TRIAC, and diodes are all probably fine for existing values. You may need a different zener to compensate for the different voltage limits.

I have not tried to work out how the circuit works exactly; these are all just educated intuition. If you like, you could download one of the PSPICE simulator programs and build the circuit in there, make sure it "works" as-is, then start experimenting with changing the input voltage to see what happens elsewhere in the circuit.

If the person that originally posted that circuit on that site is still around, you might ask them if they know what would need to be changed--it would be easier.

 

[oldbiker]

Hi Guys,
You youngun’s can put an old man to work ‘quick’. I do appreciate all of your input but one of you made a mistake; if I were to get a standard 48V controller I wouldn’t get very far due to the cutoff voltage. Ya see, research does help but its so much work, that’s why I am asking those that know...

I do like being in a forum where users concentrate on the issue at hand, and do not drift too far away on some other subject.
Regards,

 

[amberwolf]

You youngun’s can put an old man to work ‘quick’.

We're not all quite as young as all that; I will equate to The Answer (to LtU&E) come the end of this month.

I do appreciate all of your input but one of you made a mistake; if I were to get a standard 48V controller I wouldn’t get very far due to the cutoff voltage.

That depends; at least with a brushed motor controller like my 2QD or my Curtis, the cutoff is simply down at a point where the MOSFETs wouldn't fire correctly, as they both work for a fairly wide pack voltage range.

Also, some multi-pack-voltage controllers are programmable for their LVC. Probably the ones Lyen here on ES sells are.

I do like being in a forum where users concentrate on the issue at hand, and do not drift too far away on some other subject.

Ah, then we'll have to see how you like it here, being so different from whereever that place is you're talking about. :lol:

 

[oldbiker]

The place is home, and being able to speak to people that some of my interests.

 

[oldbiker]

Hi Guys,
Well… been ‘round the block wit GoldenMotor; it looked like a nice unit but they are asking an unfair price for their product shipping of the perriferal necessities.
How can I get a decent controller to handle my LFP battery; at a decent price?

 

[JRitt]

Have you looked at the controllers that Lyen has to offer? http://www24.brinkster.com/lyen/beta/ He can set the low voltage cuttoff to what you want and maybe recomend one that is right for your usage. send him a PM