battery meter

hi anyone know where i can buy a battery meter reading thingy for 72v? can only find them up to 60v cheers

Or even 100+V. I think the Cycle Analyst does. I know I can hook up a cheapo multimeter, but I'd like something a little more elegant too. Anyone know how to mod a Turnigy Watt Meter for 100+V?

A cycle analyst will work. If your looking for something cheaper then I have no idea.

just put a 12V zener diode in series with the "60V thingy" you mentioned. 1N4742, 1N5349 etc. will do just fine for the voltmeter

rick

crimsonsnake0 said:

hi anyone know where i can buy a battery meter reading thingy for 72v? can only find them up to 60v cheers

Click to expand...

As luck would have it, I just recently finished tackling this exact same problem when I upgraded my 48V Ego Scoota to 72V. I ended up helping to develop a kit for e-bikes that could deal with 60V, 72V, or for that matter any multiple of 12V. It's a 12 state, 6 LED readout that can replace many existing meters, and is cheap as chips (£10 + change).

The full story - including links - is here:
http://zenid10.wordpress.com/category/4-how-to-guide/4-3-battery-meter-upgrade/

If by battery meter you mean just a simple (and almost totally useless on anything other than SLA batteries) voltmeter, then that's pretty easy - there are literally dozens of sellers on eBay who offer digital voltmeters that cover this range.

If by battery meter you mean something really useful, like the various RC wattmeters (like this one: http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=10080&Product_Name=Turnigy_Watt_Meter_and_power_Analyzer by Turnigy etc), then you're really stuck with having to use a Cycle Analyst from ebikes.ca, as I don't know of any others that will allow you to go over 60V.

The zener diode mod suggested will totally screw up the meter readings, as it will subtract 12V from the voltage indicated, which will also bugger up the power indicated, as the unit calculates power by multiplying voltage and current.

My recommendation would be to buy a Cycle Analyst, as it's very well proven, has the capability you need and offers lots of extra useful functions for an electric vehicle. The only other alternative would be to look at modifying a cheap Turnigy (or similar) meter. I think it is probably possible to modify the power supply part of one of these units (which is almost certainly where the voltage limit comes from) and retain an accurate voltage reading (assuming there is a bit of headroom on the analogue to digital converter input on the chip). It'd be more work that it's worth though, when the CA is available off the shelf.

Zenid,

Although your LED meter is probably vaguely OK for monitoring the very approximate state of charge of an SLA battery pack, it would be pretty useless for trying to determine anything useful about the state of charge of either of the two types of lithium battery that the vast majority of ES forum members are using. Both LiPo and LiFePO4 cells have a virtually flat voltage vs charge state characteristic, so a voltmeter would read pretty much the same right up until the point where the cell voltage collapses very quickly at full discharge. It's for this reason that the majority of us are using proper power meters, that measure true power taken from the battery, both in watt hours and amp hours. These give a pretty foolproof way of monitoring consumption, provided you know your battery pack usable capacity and that it pack was fully charged at the start of the monitoring period. For voltages less than 60V these things are cheap - just $24 for the Turnigy one from HK.

Jeremy

If you are electronically inclined, you can also build one similar to the premade LED meter for any voltage you like using some PNP transistors, diodes, and a stack of LEDs. If you want fine readings of voltage, use a lot of LEDs; if you have only a tiny space use just a few.

TL431s can also be used, as in the voltage indicators in throttles. I haven't drawn a schematic of them yet, but in my DayGlo Avenger thread there is a series of pics of the parts and PCBs sufficient to recreate one from.
http://www.endless-sphere.com/forums/viewtopic.php?p=314579#p314579

But as Jeremy explains, the best option is the CA.

It is also possible to modify the Turnigy meter and the WattsUp to run on >60V, but they will not measure the voltage correction above that point, and thus all the readings other than Amps will be off.

Here's a stupid question...can I split the voltage somehow? I.E. run two turnigy meters, each with half the voltage running to them, then join back up the voltage afterward?

itchynackers said:

Here's a stupid question...can I split the voltage somehow? I.E. run two turnigy meters, each with half the voltage running to them, then join back up the voltage afterward?

Click to expand...

If you can split your battery pack in half, yes. All you do is connect one meter to each of the two packs, connecting the packs up as normal but through the meter that's on the lower pack, rather than direct. What you'll get is a somewhat difficult to interpret display on both, as each meter will read the whole current, but half the voltage and half the power (but both will show the whole amp hours used). You'll also need to find a way to disconnect the meters, otherwise the lower meter will be on all the time (as it's connected across the lower pack - normally the upper meter would be connected after the main power switch).

Jeremy

Jeremy Harris said:

Zenid,

Although your LED meter is probably vaguely OK for monitoring the very approximate state of charge of an SLA battery pack, it would be pretty useless for trying to determine anything useful about the state of charge of either of the two types of lithium battery that the vast majority of ES forum members are using. Both LiPo and LiFePO4 cells have a virtually flat voltage vs charge state characteristic, so a voltmeter would read pretty much the same right up until the point where the cell voltage collapses very quickly at full discharge. It's for this reason that the majority of us are using proper power meters, that measure true power taken from the battery, both in watt hours and amp hours. These give a pretty foolproof way of monitoring consumption, provided you know your battery pack usable capacity and that it pack was fully charged at the start of the monitoring period. For voltages less than 60V these things are cheap - just $24 for the Turnigy one from HK.

Jeremy

Click to expand...

Jeremy,

I'm well aware that the discharge characteristics of Lithium batteries make simple voltage measurements completely useless as a means of ascertaining their charge state, but I assumed that he was referring to an SLA battery bank.

I agree that the voltage is only an approximate indication of the state of charge of SLA packs, but from my pretty extensive testing and charting of the discharge characteristics of my own bike's 72V battery bank, I've found that it's a pretty reasonable one (for a properly balanced pack) providing you bear in mind the that voltages will naturally fluctuate due to gradients in the electrolyte solution, and that the discharge curve is non-linear at its extremes. I custom-set my voltages for each of the 12-states supported by the LEDs to correct for this non-linearity and 'straighten' it out vis-a-vis the LED readout.

You're completely right that measuring the power output is the only accurate way of knowing precisely how much of the batteries' capacity has been used up, but - as you say - this is not a viable, cost-effective solution for systems of 72V or more. The LED solution I suggested isn't perfect, but it's dirt-cheap and - properly implemented - does a pretty good job. If the day comes when I can get something like the monitor you mention for under £20, then I'll cheerfully bin my LEDs and get one of those instead.

Zenid.

Very few people are still using SLA packs on ebikes AFAIK; they are very rarely discussed on here now that the various lithium cells have become so affordable in terms of lifetime cost.

Wattmeters are a perfectly viable way of measuring power and capacity consumed at any voltage, not just below 60V. The Cycle Analyst happily works up to 100V and I believe that Justin may have a mod for even higher voltages if asked. Sure, the cheap Turnigy etc wattmeters only work up to 60V, but that's only because they're designed for a lower voltage market (RC models) where system voltages don't often get higher than this.

The LED system gives a rough indication for those still using lead acid batteries, as you say, but will be affected by temperature (because cell Ri is very temperature dependent), load (because cell voltage drops with current because of Ri) and state of charge (because Ri varies with state of charge). All told this limits its usefulness as a state of charge meter.

Finally, I made a homebrew wattmeter/battery charge state indicator that would work at any voltage and current and the cost of the parts was probably only about $45 or so. It's pictured on this post: http://endless-sphere.com/forums/viewtopic.php?f=2&t=14498#p219880

Jeremy

Jeremy Harris said:

Very few people are still using SLA packs on ebikes AFAIK; they are very rarely discussed on here now that the various lithium cells have become so affordable in terms of lifetime cost.

Click to expand...

Please excuse my unfamiliarity with the layout of the forum, - I only joined a few hours ago. I assumed the term "e-bikes" also referred to scooters (though I see there is a dedicated scooter section further down), where Lithium ion batteries are definitely NOT the norm, as far as I am aware. 12V 28AH Li batteries of the type that could be used by my scooter/'motorbike' and others like it are prohibitively expensive, and I and other owners are frustrated with the lack of affordable Lithium ion batteries that could replace the lead bricks we have to use.

If I am wrong, here, then I am absolutely dying to be corrected on this point. Nothing would make me happier than hearing about a Lithium pack I could use on my scooter that doesn't cost over twice the value of the whole bike.

Jeremy Harris said:

Finally, I made a homebrew wattmeter/battery charge state indicator that would work at any voltage and current and the cost of the parts was probably only about $45 or so. It's pictured on this post: http://endless-sphere.com/forums/viewtopic.php?f=2&t=14498#p219880

Click to expand...

Thanks for the info. That's not a bad price at all, and I'll bear that in mind as a prospective upgrade further down the line. Do you have a schematic, and details of where you could find those parts?

Zenid said:

Please excuse my unfamiliarity with the layout of the forum, - I only joined a few hours ago. I assumed the term "e-bikes" also referred to scooters (though I see there is a dedicated scooter section further down), where Lithium ion batteries are definitely NOT the norm, as far as I am aware. 12V 28AH Li batteries of the type that could be used by my scooter/'motorbike' and others like it are prohibitively expensive, and I and other owners are frustrated with the lack of affordable Lithium ion batteries that could replace the lead bricks we have to use.

If I am wrong, here, then I am absolutely dying to be corrected on this point. Nothing would make me happier than hearing about a Lithium pack I could use on my scooter that doesn't cost over twice the value of the whole bike.

Click to expand...

This is the ebike technical section, so tends to be focussed on ebikes (which very much tend to use lithium over all other cell chemistries), rather than bigger vehicles like scooters and motorcycles. Scooters etc are a bit in the minority here, although there are a few posts in the "Electric scooters and motorcycles" section of the forum that may be of interest.

I assume that you're now running a 72V nominal, 28Ah notional capacity, lead acid set up at the moment. This has a theoretical capacity of around 2000 Wh and a usable capacity of maybe 1600 Wh under ideal conditions (probably less, due to the high Peukert factor for LA cells). A direct replacement for your present battery pack might be either a 24S LiFePO4 pack (77V nominal) or maybe a 20S LiPo pack (74V nominal). For the same usable capacity you would want around 22Ah of either lithium cells. A 20S, 5P pack of 5Ah, 15C LiPo (which would give you around 24 to 25 Ah of usable capacity at about 74V) would cost you around $800 plus shipping and taxes, and you would need a to buy a suitable charging system, and, perhaps, some form of low voltage warning.

If you were to opt for LiFePO4, then the best buy at the moment are probably Headway cells. These are around $19 each for 10Ah cells, and, to ensure that you have at least as much power as you have now (in fact around 20 to 25% more) you would really need a 24S, 3P pack (76V, 30Ah). This would cost around $1400 plus shipping, taxes, etc, and again you would need a new charging system and, ideally, a battery management system.

Maybe this is way too expensive, but given that the lithium cells (particularly LiFePO4) may last at least twice, maybe three or four times, as long as sealed lead acid, plus give a big weight reduction and better voltage stability, they are well worth looking at.

Jeremy

Another significant advantage of the Li as opposed to SLA that factors into cost and lifetime is that with SLA, if you are not recharging at the destination end of your trip, for something like a typical workday commute, and you've significantly drained the pack (let's say to something like 2/3 capacity or further), then each time a little sulfation damage will occur to the pack from being left discharged for those several hours a day, most days a week.

That won't happen with Li, AFAIK.


Regarding the usefulness of LED meters vs chemistry: while it's true that SLA is the primary one they're good for, they also work quite well with NiCD and NiMH. While they're not the most common chemistries still in use, they are still sold even by places like ebikes.ca as ebike packs. (and they are also used in larger EVs)

Jeremy and others.

when i suggested using a zener i was assuming that all he wanted to do was adapt an existing voltmeter display for a higher voltage. all that he specifically asked for was a "voltmeter thingy". i assumed like one of those 3-LED ones built into a throttle.

when someone uses highly technical language such as "Thingy" i assume that he is not looking for a Coulomb counter or other more sophisticated display. so what i offered was a suggestion that would allow him to use an existing LED throttle and extend it's range.

but i agree. it is not the most accurate.

rick

Jeremy Harris said:

I assume that you're now running a 72V nominal, 28Ah notional capacity, lead acid set up at the moment. This has a theoretical capacity of around 2000 Wh and a usable capacity of maybe 1600 Wh under ideal conditions (probably less, due to the high Peukert factor for LA cells).

Click to expand...

Right on every count. I'm using six 12V 28AH SLAs, total of about 2000W. I didn't realise the usable capacity would be restricted in this way, however, and it's making me wonder if it's worth upgrading the motor to a 2000W one, rather than simply 'overclocking' the existing 1400W motor, as others with this bike are doing.

Jeremy Harris said:

A direct replacement for your present battery pack might be either a 24S LiFePO4 pack (77V nominal) or maybe a 20S LiPo pack (74V nominal). For the same usable capacity you would want around 22Ah of either lithium cells. A 20S, 5P pack of 5Ah, 15C LiPo (which would give you around 24 to 25 Ah of usable capacity at about 74V) would cost you around $800 plus shipping and taxes, and you would need a to buy a suitable charging system, and, perhaps, some form of low voltage warning.

If you were to opt for LiFePO4, then the best buy at the moment are probably Headway cells. These are around $19 each for 10Ah cells, and, to ensure that you have at least as much power as you have now (in fact around 20 to 25% more) you would really need a 24S, 3P pack (76V, 30Ah). This would cost around $1400 plus shipping, taxes, etc, and again you would need a new charging system and, ideally, a battery management system.

Maybe this is way too expensive, but given that the lithium cells (particularly LiFePO4) may last at least twice, maybe three or four times, as long as sealed lead acid, plus give a big weight reduction and better voltage stability, they are well worth looking at.
Jeremy

Click to expand...

Thanks very much for taking the trouble to give such a detailed analysis of my options, here. Though I clearly have quite a bit of homework to do as my understanding of Lithium ion technology is limited to their basic - far superior - characteristics regarding voltage regularity, life-span and weight. As I understand it, I'd be shedding about 65% of my battery weight by switching from SLAs (is that right?), or a whopping 32Kg in my case! Simply losing this weight should add substantially to the power I get in practical terms.

I was pleasantly surprised that your lower end solution is as little as that ($800/£520), as that brings it well into a viable, cost-effective range. Even the $1400(£920) solution isn't out of the question, as since my existing SLA bank costs $360(£240), the 3-4X lifespan alone makes it approach equivalence, pound for pound.

Can you recommend a supplier, preferably UK-based?

There is, though, the matter of battery chargers (are these more expensive than SLA chargers?) and battery management, plus shipping, taxes etc. I have to enumerate the costs involved to get a better picture. One other highly pertinent issue is the matter of regenerative braking. The controller I'm about to get will have this, but presumably it is set up for SLAs, and I have no idea of the implications of this for dealing with a Lithium battery bank. Controllers can be programmed to alter various parameters, but I have no idea if the recharging circuit can be set up for these batteries.

Zenid said:

Simply losing this weight should add substantially to the power I get in practical terms.

Click to expand...

It will also change the handling of the vehicle, probably for the better, but keep it in mind.

One other highly pertinent issue is the matter of regenerative braking. The controller I'm about to get will have this, but presumably it is set up for SLAs, and I have no idea of the implications of this for dealing with a Lithium battery bank.

Click to expand...

Well, SLAs don't generally take a very high charge current, usually something on the order of 1/4C at most. For yours, that would be about 7A, which isn't much in the realm of regen from higher speeds--you could get a lot higher currents than that initially. The SLA might take that ok, if it's infrequent and very short duration.

Li types of cells typically can take much higher charge currents, some higher than 1C, making regen easier and more effective and possibly safer regarding the batteries.

Controllers can be programmed to alter various parameters, but I have no idea if the recharging circuit can be set up for these batteries.

Click to expand...

Theoretically, any BMS you have for them would be what deals with that. But the controller should have some control over the regen voltage and/or current output limits, although it might be via component values (changing resistors/etc.) rather than via programmed parameters.

I'm afraid I've yet to find an affordable UK supplier - all seem to add a whopping markup that makes it cheaper to buy direct from China. I've dealt directly with Headway (Victoria is the English speaking representative) and found them OK, although payment has to be via bank transfer, as they are unable to take credit cards or PayPal. Hobby City are the best supplier for LiPo I believe, they sell directly from Hong Kong, have a pretty good reputation and you can pay easily online.

If you really want to try a UK vendor, then I know of two resellers, LifeBatt who offer 10Ah LiFePO4 cells that are similar, although perhaps slightly better in some respects, to Headway cells, at around £20 - £25 per cell (so perhaps £1400 - £1800 + VAT and shipping). The other reseller is selling Hipower cells from China, these: http://www.chinabatteries.net/english/index.asp but his website is down and I don't have any saved information on price (but recall that they weren't cheap!).

I wouldn't worry about the motor power rating too much, as it's a very notional figure and doesn't really mean much in practice. Road vehicles tend to run with a pretty high ratio between peak and average power and the motor rating will be based on the average power. For example, a standard family saloon car may have an average power usage of around 12 to 15kW, but a peak power output of maybe 5 or 6 times this figure. The motor rating will be based on its ability to get rid of heat without getting hot enough to cook the windings and Hall sensors (if fitted), so as long as it doesn't get too hot you should be able to up the peak power a fair bit without any problems.

Charging any lithium chemistry battery is far more critical than charging lead acid. The charger itself will be similar to a lead acid charger, constant current up to a limiting voltage, then constant voltage, but each cell in the battery pack need to have some form of individual voltage limiting circuit in order to ensure that the pack stays balanced. Unlike lead acid cells, lithium cells don't self balance - they will be destroyed if allowed to reach too high a cell terminal voltage on charge.

Similarly, lithium cells need protection from being over-discharged, too. They need a monitoring circuit that either warns the user that a cell in the pack is getting critically low, or better still actively prevents further discharge by cutting the throttle.

There is a wealth of information on battery management in some of the threads here, although some of it is deeply buried amongst a lot of other posts!

Jeremy

Jeremy Harris said:

A direct replacement for your present battery pack might be either a 24S LiFePO4 pack (77V nominal) or maybe a 20S LiPo pack (74V nominal)
{...}

Click to expand...

I've just been reading up on LiPo batteries, and saw a couple of pretty horrifying videos of what these batteries can do if something goes wrong. The prospect of a blazing inferno igniting under my ass as I'm riding along at 30mph plus doesn't bear thinking. Are these batteries as volatile as people say they are, or are these claims exaggerated, or has the technology improved such that they're not liable to explode due to a short or minor impact?

The battery fire scaremongering has really done lithium an unfair disservice. Virtually every mobile phone and laptop in the world is running on LiPo, as are a few tens of thousands of power tools and model aircraft, yet you don't hear of these bursting into flames. People on this forum have seriously abused LiPo cells, but again I'm not aware of anyone having a had any cells catch fire (other than one or two who have deliberately caused them to by way of an experiment to see what happens).

The key to using LiPo cells safely is good battery management. The only really critical phase is battery charging, so if this is carefully controlled the cells are as safe as any other battery chemistry. I suspect that more lead acid batteries blow up during charging than LiPo batteries; I well recall seeing safety films years ago with graphic representations of hydrogen explosions caused by charging lead acid cells inappropriately.

Jeremy

deleted

Jeremy Harris said:

The battery fire scaremongering has really done lithium an unfair disservice. Virtually every mobile phone and laptop in the world is running on LiPo, as are a few tens of thousands of power tools and model aircraft, yet you don't hear of these bursting into flames [...]

Click to expand...

Thanks for the reassurance on that point. I just happened upon a link covering a littany of horror stories, including alleged incidents where even a dog bite caused a cell to catch fire.

Does the"20S, 5P pack of 5Ah, 15C LiPo" (74V nominal) offer a capacity and power close to what I have now, and can I draw as much amperage? Clearly the $800 option is much more attractive, but will my performance or range suffer if I don't go for your more expensive option? On the face of it the pack you describe - 20 x 5AH (100AH combined) would look lower capacity than my 6 x 28AH (168 combined AH). What am I missing?

I've been looking for this but am having trouble finding/recognising it listed on the site whose link you gave me:
http://www.chinabatteries.net/english/products.asp?lmid1=115&lmid2=116&lmid3=0&state=show&id=73

Could you point it out for me, or maybe point me to elsewhere where it's listed (preferably with prices)?

Also I'm confused about "24S, 3P pack (76V, 30Ah)": This would be 24 cells right? 30AH per cell would be huge, again am I missing something?

Thanks

Zenid

Zenid said:

Does the"20S, 5P pack of 5Ah, 15C LiPo" (74V nominal) offer a capacity and power close to what I have now, and can I draw as much amperage?

Click to expand...

20S would be 20 series cells, typically 3.6-3.7V per cell. 5P is 5 parallel strings of those cells. At 5Ah per cell that means 25Ah. At 15C current draw capability, that means 15 x 25 for the Amps you could draw at any time.

So up to 375A you can draw, at 72V. At 72V and 25Ah, that's 1800Wh.

Your original SLA would probably deliver closer to 14-17Ah out of their rated 28Ah, perhaps less, at a guess. I'm sure it's usable Wh is notably lower than 1800, too. So I'd guess that the Li pack would be significantly superior, even relative to it's cost.

On the face of it the pack you describe - 20 x 5AH (100AH combined) would look lower capacity than my 6 x 28AH (168 combined AH). What am I missing?

Click to expand...

It's not calculated like that. it's not 20 x 5Ah, because 20 is how many are in series, which is multiplied by volts, not Ah. So to find Ah of the pack, you multiply by how many are in parallel, which is 5; that gives you 25Ah.

Your 6 SLA cells are only 1P, meaning none of them are actually paralled, so their Ah does not multiply at all, and they are only 28Ah for your whole pack.

If you wish to calculate a capacity the way you are describing, including the voltage of the pack (number of cells in series), you must use Watt-hours, or Wh, which is Volts x Amp-hours.

That gives your SLA pack, at a nominal 13V per "cell" (battery), 6s (13 x 6) x 28Ah = 2184Wh. But the Peukert effect (much owrese on lead-acid than other chemistries) probably effectively halves that, for around 1100Wh. That depends on the actual current draw and how those SLA are rated by the manufacturer for their Ah. Usually they rate them for a 20 hour draw at some fraction of their capacity, meaning that they could supply that full 28Ah only if drawn slowly over 20 hours. If you draw a lot more Amps, which is typical of any EV, they are going to deliver a lot less. Even if you only drew 28A, they would not last a full hour. If you drew 280A, they certainly wouldn't last six minutes (1/10 of an hour). It's not a linear scale, unfortunately.

Plus if you run the SLA pack down to really zero Ah/Wh, it won't last very long. Maybe a couple hundred cycles or less. Probably a lot less if you did that every time. Usually 50% Depth of Discharge (DoD) is the most you'd want to go to keep an SLA pack working for a reasonably long time. 60-70% at most. So including accounting for Peukert, you really only get 550-770Wh usable from it. Even if Peukert effect isn't more than 25% then you still only get up to 825-1155Wh.

Now for a Li pack, call it 3.6V per cell for 20s is 72V. 72V x 25Ah (5 parallel strings x 5Ah) is 1800Wh. You'll probably be able to use up to 80% DoD on Li without seriously affecting it's lifespan, so you realistically get up to 1440Wh out of it.

Also I'm confused about "24S, 3P pack (76V, 30Ah)": This would be 24 cells right? 30AH per cell would be huge, again am I missing something?

Click to expand...

It's 24 series cells in each of 3 parallel strings. So each of those cells is only 10Ah, but there are 3 in parallel for 30Ah total. Then 24 of those sets in series. (or 3 strings of 24, doesnt' matter which way you think of it).

Zenid said:

Thanks for the reassurance on that point. I just happened upon a link covering a littany of horror stories, including alleged incidents where even a dog bite caused a cell to catch fire.

Click to expand...

I'm sure it's possible for lots of things to happen depending on the cell construction and chemistry. But these days, many cells appear to be made to prevent this.

Liveforphysics here on ES did an extensive destructive test video set of various cell types, and it took some extreme measures to get any of them to burn. You'd have a lot more worries on your hands than the cells burning if you got that far with the ones he tested.

Highlights video:
http://www.youtube.com/watch?v=dree0rTr1HM
The rest should be on his user page.

Zenid said:

30AH per cell would be huge, again am I missing something?

Click to expand...

Depends on the chemistry. There are some flat-type (pouch) cells that Cell_man here on ES was selling (he ran out) that were 20Ah capacity, but not very large.
http://endless-sphere.com/forums/viewtopic.php?f=14&t=15408&start=0

Even the 60Ah Thundersky cells I have here are not all that large, and that includes their outer plastic protective case.


The TS60Ah is the yellow brick, which weighs less than the smaller black 7Ah SLA. The gray SLA is a 31Ah 12V powerchair battery.