LOW VOLTAGE ALARM

[Burgerman]

http://www.wheelchairdriver.com/lithium ... -alarm.htm

This is a page dedicated to the low voltage alarm/display that I intend to use on my BM3. I ordered a handful as they are so cheap. I can use for testing hobby batteries as on this page, as well as fitting two (they each do 8 series cells and I need 13) on (under the seat) on my BM3 chair.

This is the other half of the GET RID OF THAT CRAPPY BMS solution! You need a proper balance charger (hyperion) for charging, High C rate cells so you cant exceed the limit, and a low voltage warning so you cannot drive too far.

Then everything is safe and no battery murdering unreliable BMS required.

Take your pick... High quality, programmable, reliable loud and very bright! 8 cells or rather up to 8.
http://www.ebay.co.uk/sch/i.html?_trksi ... &_from=R40

And if you fit one to an anderson connector, tell it its a LIFE 4 cell pack, and connect the pins together with 1k resistors it can stay connected to your unused powerchairs as a monitor on each seperate battery. OR on a short cable plugged into the normal charge port (neutrix) as a 24v monitor (8 cell) on any chair. Residual drain is extremely light. I will make a few of these and add them to the page.

[flagman1776]

When you order these, do they come with connectors & pigtails or do you have to find these on your own?

[ex-Gooserider]

Don't know, you'd need to read the product description to find out... However it shouldn't be a big issue, as the items pictured look like they use a standard .1" spacing header, which should be an easy item to find at any reasonably good electronics shop.

Definitely a standard item that is used in lots of applications - that .1" header spacing is pretty much the default for any kind of board stacking, and plug connections that don't have other special requirements to make them use something different.

ex-Gooserider

[Burgerman]

You can order a connector for an 8 cell charger balance cable. Hobby shop. Or just solder and heat shrink since they are cheap...

[shirley_hkg]

I'm using this / clone Celog8 to alert me if a cell is 2.8V low. BMS is for charging only.

However, it runs on the power of the first 2 cells; 0V -- 6.6V ONLY ; and it eats up an extra 1.5 Ah of that particular 2 cells in 10 days. That means you can't turn it on 7/24.

I turn it on when it gets 25.5V or lower , or when necessary. Put a switch in the 0V path SO YOU CAN SWITCH IT ON / OFF.

Endless-sphere has thread on how to even the power drain on all cells connected.

[g7pvs]

Hi,
I purchased a second monitor and am going to unsolder the 7 segment LED displays and replace with sockets so I can plug/unplug the displays.
I'll then set it up for the voltage I want the alarm to go off at, then remove the 7 segment LED displays.
As most of the current is likely to be used by the LEDs, removing them should massively reduce the power used and the alarm should still work OK.
Best Regards
Trev

[woodygb]

As most of the current is likely to be used by the LEDs, removing them should massively reduce the power

... a bit OTT ...the leds will draw VERY LITTLE... mA'S

http://www.kpsec.freeuk.com/components/led.htm

[g7pvs]

Hi,
1.5Ah over 10 days = 0.15Ah per day = 6.25mA per hour or 6.25mA current drawn.
So although only mA over time it does make a difference.
The only other component drawing current on the PCB is the microcontroller which should be uA.
Best Regards
Trev

[Burgerman]

http://www.ebay.co.uk/itm/8S-Balance-Ch ... 43b5669781

Connectors like these are what you need, any good hobby shop.

I now have one connected to 2 andersons. So I can plug it into any unused powerchair (via the twin connectors) and see battey voltage. Thats total volts. And on pin 1 and 9. And the OTHER pins are interconnected via 1k resistors.

So POS-24V R R R MIDDLE-PIN 12V+ R R R NEG-0V

So I can plug one in and read 12v, 2 in and read 24v, and see a 3v reading in the "spaces" due to the resistors. The checker will be set to 14.4v charged or full LiFe and used on lead.

OR

8 cell LIFE connected to ACTUAL cells...

Or 2 of these connected to 13 series cells.

Any cell goes below 2.7v and it warns me clearly. I can then stop as I should, or risk it and get home or charge it...

[flagman1776]

OK, so if I wire 1 flat plug to each LiFePO4 pack... I can use the alarm to protect the cells. Can the same plug be used with a balance charger for that use or does it require a separate charging plug?

[Burgerman]

You could just use your balance connector. But I will connect two (13s pack) with 8 and 7 cells monitored under the seat. Hard wired. So if it ever beeps, I know a cell or more is low.

[Burgerman]

With this to monitor the low end and beep, the hyperion to correctly charge and balance the high end, and cells of more than adequate C rate to cope in use, then you have covered all the stuff a BMS does but without the problems... No BMS needed.

[flagman1776]

But can the buzzer and charger utilize the same connector in turn (wired as it is to each cell)? The load would be wired through big Andersons.

[Burgerman]

Should you wish yes. 8 cells needs 9 wires. These are the same connection points as the balance wires. With a hobby pack you USE the balance connector. Or a Y lead version.

[Sully]

I really like this idea, When we/you first started to talk about the LiFePO4 batteries, the conversation gravitated towards how junky the BMS were, but there was no alternative mentioned. Since running this type of battery low can destroy them or they can reach some unstable point, not having a suggested alternative way to keep from wrecking them seemed to be folly.

This alternative seems to be the answer. Thoses 787 planes are having their troubles, I would not be at all suprised if a solution like this, might be an answer for them as well. I also think the spaces containing these battery banks should be better ventilated.

When I read all the specs for LiFePO4 batteries the larger the AH designation the higher the temps they state they attain and a lesser C rate. Just the opposite of what seems to be desirable. I am probably verbalizing this wrong, but spend some time and see for yourselves.

The 10 and 12 AH individual cells are the best for all uses.

[Burgerman]

Well they are better for high loads, for several other reasons.

1. better cooling with metal cans and much more surface area. Any serious problems dont cause heat runaway issues. And longer high C rate continuous discharge or faster charge is possible.
2. tightly rolled up plate material, and contained by metal, means they dont try to seperate and puff up under high loads as they are clamped together. So have long term lower resistance and higher c rate.
3. Pysically stronger.
4. Easy to build bigger packs with equal self discharge, resistance, capacity and resting voltage in strings by shuffling them about.
5. Easy to change 1 or 2 cells if a problem develops.
6. more flexible sizes and shapes that fill the existing battery area allowing greater Ah to fit in a given space.

Disadvantages

Many connections. And every one must be tight or balance problems...
Gathers dust...
Ends exposed to short out on stuff.

[Burgerman]

The lack of BMS on hobby stuff, RC transmitters, helis etc never much of a problem. As long as:

a. You dont run them too low. In a powerchair that would be right down to 10.8 volts or less. You wouldnt do this anyway as it would be noticibly slow, and all your lights will be out... The meter will stay on full all day, then drop off a cliff very fast and chair will slow way before you hit this point. So low end isnt really a worry - at least if you have any clue at all! It will be obvious. But the alarms are available too.

b. You dont take too many amps. In other words dont rely on a BMS to cut off the power. Use a battery/cell with adequate Ah and C rate that exceeds powerchairs controllers max output by about 2x. A 100 amp controller can pull 200 amps. 100 per motor. On a slope or steep hill for 10 or 15 seconds at least. A BMS also protects the cells by stopping/cutting power dead at some unknown value! Not what you want.

c. On charge, do NOT exceed 3.6v (3.65v upper limit) on any cell Any good all in one balance charger like the hyperion does this very accurately and extremely well. A BMS just doesent! At least if it cant control charger output, and isnt logic controlled. But they claim to do so. All of them allow cells to go much higher or never properly balance. Its a cheap crappy way of selling lithiums to dumb end users. It kind of works for a while at times...

They are the three important things.

Dont run too low. Easy to arrange.
Dont take too much power at once. Use adequate C rate cells.
Dont charge beyond 3.6 volts. Use proper balance charger.

NO BMS REQUIRED.

[Sully]

Now give Boeing a call, and show their engineers where they have taken a wrong turn. From news paper accounts they haven't a clue!

[Burgerman]

I probably could. But most likely the reporters have got it wrong. They are always useless!

[Burgerman]

I really like this idea, When we/you first started to talk about the LiFePO4 batteries, the conversation gravitated towards how junky the BMS were, but there was no alternative mentioned. Since running this type of battery low can destroy them or they can reach some unstable point, not having a suggested alternative way to keep from wrecking them seemed to be folly.

Charger (if a true balance charger) sorts out the high end.
The low end isnt much of a problem, and the SYSTEM you are using can be a better way to cure this. The roboteq can be configured to do something at X battery volts. At low or high end. So I can tell it to shut down at 2.5v per cell (2.5v x 13 in this case) or bring on a light or limit speed, whatever. Theres a lot of ways to do this.

[Mark]

Ordered a pair of these from each of two different suppliers a few weeks ago and the first pair arrived today. Unfortunately only one works as the other just shows all segments including dots lit when power applied. However, at this price, it is cheaper to put it in the bin and buy another than post it back to China! No problem.

Good clear display on the other one - well on both actually, but 8.8.8 is a bit boring all the time!

I've been looking for a pair of cheap voltmeters to monitor battery voltages separately on my powerchairs for some time. I think these might just do the trick. Interesting that they don't light up when 24 v applied across pins 1 and 9, but they seem to be powered by the pins at the lower end. Have you measured the current drain? Will go back to trying them when the second pair arrive and I have some better connectors for them, but certainly looking good and the price is right. Thanks for spotting these. Soon half the bits and pieces around here will be as a result of first reading about them on this forum ! - Many thanks !

[Burgerman]

Stick 1k resistors between each pin 1 to 9. Then connect a battery to the ends. Then you should see battery total voltage and "cell" voltages. Although these will be imaginary lithium cells... But then you can set the alarm for low votage etc. It will see the resistors as cells. Not yet tested... But sure to work. If not, use 470 ohm or 220.

They work great on my 13s pack - two of them.

[Mark]

Yes, I'll play around and report back when the next pair arrive and I get proper connectors fitted. Was rather hoping to avoid the display cycling round all "cells" and just use them to monitor the 12V of each battery. Don't know what will happen if it sees 12 across pins 1 and 2. Might order some more in case I ruin some trying various options to avoid having to wait through the cycle to get the mk monoblock voltage. Lead time from China around 3 weeks so will probably start another order now. Dirt cheap aren't they. Haven't checked accuracy against my "portable standard voltmeter" yet. It's calibration certificate is nearly a hundred years old (May 1918) but it is still one of the most accurate instruments I have.

[Burgerman]

You need one for both batteries. 8 cells - 24v.

You could use one and a centre tap, So 0v - 12v - 24v but that 12v will 4 cells added, and batt 2 will be the next 4...

If you want to monitor 1 battery use pins 1 to 5 with resistors between the pins.

But one logger will do both batteries.

[shirley_hkg]

I have this small LCD voltmetre to read the battery volt all the time and turn on the 8-cells voltage monitor at 25.5V or when necessary !

LCD runs on the battery which it measures thought it's best to feed it with 5V. It sells @ $2.00 US

[shirley_hkg]

HAVE IT ON MY BACKUP CHAIR.

[shirley_hkg]

ACTURAL SIZE .

[Mark]

Well the second pair arrived from China today. Trying to make sense of the very limited instructions on the rear it looks as if "1S test mode voltage range 3.7-30V" means that you can read voltages up to 30 V by using pins 1 and 2.

Tried this and sample one (the working one of the pair that arrived earlier) did read the voltage of 3 AA Batteries in series. The pair that arrived today did similarly, although they quoted the voltage as 3 least-significant digits (lsd) higher. Then tried on a higher voltage. Sample one now read the voltage as 26.9 V and each of today's pairs read the voltage as 27.2 volts. Again 3 lsd difference. Not too bad I thought. Then tried on 29 volts and each of the three samples made a faint noise and then died. So I now have four broken ones.

The second pair were better than the first, as they just cycled between "ALL" and "1", whereas the first sample went through ficticious values for the other 7 open-circuit cells as well on the higher voltage. However, since I ordered each pair from different suppliers on the same day, and since neither packet has a supplier's name that co-incides with the ebay listing, it isn't going to be easy to tell which is which.

Accuracy would probably be ok for what I want (monitoring each 12V lead acid monoblock to check that they are remaining balanced) but I'll need to order some more. I thought it would be easy to identify each supplier, but hardly any info on the small jiffy bags and no info inside. Interesting that all three failed with about 29 volts across pins 1 and 2, whereas the spec / instructions suggested they would work up to at least 30 V. Should be ok, though with each monitoring a single 12V block.

Reminds me of what I used to say at the Research Department "Experience is proportional to the amount of apparatus ruined".

[Burgerman]

I wouldnt have dared put all the volts through 1 amd 2...

Use 1 and 9 for 24v. With the centre pin connected to the joint between the 2 batteries. The connect a high value resistor between each pin?

[Mark]

Yes, there was a risk - now proved - but the instructions did suggest it was possible and I was trying to avoid the display cycling through all the 8 values. Total experience so far is £6.46 ! Just off to spend more than that on my beer tonight.