Li-poly batterys 5V?

[alband]

Hello again, this is another one in reference to a tank I'm making.

It is basically a model tank, that has been converted to remote control using aircraft RC gear.

I recently stripped it out though to start again, as I had added on so much stuff that it would be better to retry with all new improvements in mind from the start.

The whole system would run on 5V quite well:
Receiver = 4.8v - 6v
PICAXE = 3v - 5.5v
Motors = 0v - 6v

My previous design used a load of 1.2v NiMH batteries but they only gave 1400mAh. I have now found a li-poly pack that gives 6Ah - much more like it.


The problem is that, being li-poly, it only gives 3.7V.
What would be the best way to bring this to 5V?
Can li-polys be put in series (I suspect not and wouldn't try it unless sure)?
Would this step-up module be able to supply two hungry motors?

 

[BeanieBots]

It is basically a model tank, that has been converted to remote control using aircraft RC gear.

Before anyone has a panic attck and contacts the local authority, Alband has assured us in an earlier thread that he IS NOT using the aircraft band for land based use.

Alband, those units will probably be OK with 3.7v input. I'd like to see a full and proper spec before buying one myself though.

Is that enough current for your needs?
It's not enough to make a single servo even twitch.
What do you need the 5v for? Servos need about 1A each.

LiPo voltage ranges from around 3.6v flat to 4.3v full.
They are a bit like Pb technology in so far as they hate deep discharge or being left flat for long periods of time.


Yes, you can put them is series. Make sure the capacities are identical and use a balanced charger or there will be tears.

Safety warning:
DO NOT discharge below 3.6v/cell. (use a cut-out circuit)
DO NOT overcharge the weakest cell (use a balanced charger)
Plenty of stuff to help you with these issues at your local hobby store.

EDIT:

Would this step-up module be able to supply two hungry motors?

Absolutely NO CHANCE at all.

 

[eclectic]

"My previous design used a load of 1.2v NiMH batteries but they only gave 1400mAh. "

Well a quick and easy solution is to buy more modern >2500 mAh NiMh

e

 

[alband]

@ eclectic, it sounds like it would be, but there was one big problem with using multiple "cylinder" - type cells hich was I had to take them out of the plastic wrapping (the pack wrapping not each cell's wrapping) and then make my own connectors so that they would fit. They still made the sides bulge however, and they frequently disconnected themselves. This makes that 6Ah pack look idea, because it is a single unit that will fit snugly.
Also, it sounds barmey, but I doubt 2400mah or even 3300mAh would be enough. The current draw on the 1400 was enough to drop the voltage a great deal and it only lasted about 15min before the charge started to drop.

Here is a full list of everything that will require power and there VDDs:

Hull:
PICAXE 20X2 3-5.5v
Spectrum 5-channel Receiver (AR500) 3.5-9.6V
2x Drive motors (standard motor 2) "nominally runs at 6VDC"
2x ESC (electronic speed controllers) for the motors (using BEC) 4.8-8.4V
4x iPod backlight LED's. Each ~ 3v (probably going to hook them up straight to the li-poly in parallel)

Turret:
"Ultrafly ST-5 Micro Servo" whatever voltage servos usual take (4.8-6v I think?)
PICAXE20X2 SMT 3-5.5V
Kionix KXPS5 accelerometer 2.5-5.25v (tested it and it does work @ 5v)
HM55B Compass Module 4.8-5.2v
Some kind of motor, perhaps PSP-laser's stepper motor, to drive turret rotation, may have another ESC, same type as before.

Despite whether the rest works at 3.7v the ESC's don't and they can't be swapped (partly because it took me years to find them and lots of money to import them).

Is that enough current for your needs?
It's not enough to make a single servo even twitch.

The previous batteries had 1400mAh and this li-poly has 6Ah so surely it will be enough.

My understanding of step-up and step-down IC's are that they convert ampage into voltage and visa-versa, is this wrong?
If it is right, would it prove more practical to put the li-poly's in series then use a 7805 reg?

The space I have for a battery is 20x57x57mm so this battery pack seems great but is it going to prove impossible to power all the units?

Also, when li-poly's discharge, don't they drop quick quickly from 4.3v to 3.7v then slowly drop to 3.6v?

 

[BeanieBots]

Blimey, so many questions..
and so much missunderstanding... where to start!

4x iPod backlight LED's. Each ~ 3v (probably going to hook them up straight to the li-poly in parallel)

Please video this, I've heard about LEDs exploding, but never actually seen it.

Your voltage ratings are little concern to battery life.
It's current draw that matters.

Some kind of motor, perhaps PSP-laser's stepper motor, to drive turret rotation, may have another ESC, same type as before.

Stepper motor... RC type ESC.... Don't think so!

My understanding of step-up and step-down IC's are that they convert ampage into voltage and visa-versa, is this wrong?
If it is right, would it prove more practical to put the li-poly's in series then use a 7805 reg?

Your understanding is almost correct. (and only true for switching regulators)
They convert power. So, let's say you step 3.7v up to 5v.
Let's assume a typical 80% efficiency.
Let's assume a 1A load.
That 1A at 5v will load the 3.7v battery at 1.6A (80% eff).

The previous batteries had 1400mAh and this li-poly has 6Ah so surely it will be enough.

As far as the battery is concerned, better, but not by as much as you think.
However,
The up converter you were looking at can only supply a maximum of 100mA!

Going the other way (higher battery voltage) using a 7805 regulator will also be quite lossy. The 7805 is a linear regulator, so it simply loses the excess voltage and dissipates that voltage times whatever current as heat.

Personally, I think your best bet is to use a higher voltage battery which suites your highest current demand devices, (namely the ESCs) and then use a linear regulator for the low current lower voltage devices (PICAXE & receiver etc).

 

[hippy]

The current draw on the 1400 was enough to drop the voltage a great deal and it only lasted about 15min before the charge started to drop.

A rough 'back of an envelope calculation' suggests that's using 5.6A.

If you need 5V for the ESC I cannot see any alternative other than using two in series; conveniently the ESC will work at 7.2V. If the motors ( through ESC ) can handle that you're sorted.

For other electronics I cannot see how you can avoid a 5V regulator, but others may know better.

 

[BeanieBots]

Don't forget, the ESC BEC gives you 5v.
Refer to the ESC datasheet for current rating but they are designed to run several servos and the Rx unit.

 

[alband]

I will post the video, but don't keep you're hopes up, I've used them many times before, I just can't get any spec.

BTW I know that ESC + stepper = hence the "perhaps" and "may"

Anyway, I'm not surprised quite a lot of my understanding is wrong and its nice to clear it up.

There is surely, then, such a thing as a non-linear, step-down regulator that will turn 11.4v (basically same pack but wired in series) into 5V and give a bit of ampage.
eg.

Take hippy's figure of 5.6A (which sounds right given there will be 4 motors and some extra) current draw.
5V load.
5.6A at 5V = 28
28 / 11.1V = 2.522A
80% eff = 3.15A

The battery in series would have only 2200mAh so should last aprox 40min?

 

[alband]

Don't forget, the ESC BEC gives you 5v.
Refer to the ESC datasheet for current rating but they are designed to run several servos and the Rx unit.

It hasn't got a sheet, only this:

For brushed motors
Micro 8-Bi (8 amps)
4.8V to 8.4V, with BEC
(Up to 12V without BEC)
3kHz switching frequency

So it will be able to supply 8A max but only if there is enough available?

 

[BeanieBots]

You're still missing a fundamental point.
By supplying the ESC with a higher voltage, you don't need high current from the 5v.

Note that Hippy's figure of 5.6A assumes a constant load.
Is that the case? or is it more like move, stop for a bit, move a bit more?
If the latter, the peak current is more likely to be ~8A or even higher.

Yes, you can get non-linear (switching) step down regulators and yes they do work as you have indicated.
They are also big, heavy and very expensive at the current ratings you need.

Something along the lines of this would give you 6v @ 12A peak.
http://www.powerstream.com/dc-buck-boost.htm

 

[hippy]

The battery in series would have only 2200mAh so should last aprox 40min?

I really don't follow that under any context discussed.

 

[alband]

Yes but the ESC's will only take up to 8.4v with BEC which limmits to only 2 cells.

It wouldn't be constant current draw but there would be multiple units going for most of the time.

However, the motors need 6v so if I took 4 of the 2200mAh cells in series (14.8v) and supplied this with them to give 6v, then feed that into the ESC's. That way, the motors get the 6v they need and the rest gets 5v.

I would need to find some more spec on that converter to check if I can fit it and 4 cells into the hull. However, electrically it should be fine and be able to deliver quite a few amps for quite a while?

 

[alband]

I really don't follow that under any context discussed.

I got that figure from what must be a bad missunderstanding of a mAh. Surely if you have a figure of current draw and a capacity then it is possible to calculate how long it should last?

 

[BeanieBots]

I got that figure from what must be a bad missunderstanding of a mAh

I think you missed out a step somewhere in your reasoning. You may well be right but it doesn't follow on from the figures discussed earlier.

I still don't know you want to do this voltage conversion
Compared to your drive motors, everything else falls into insignificance so we largely ignore them.

7.4v is IDEAL to drive 6v motors via an ESC.
Don't forget that the action of the ESC together with the inductance of the motor and the clamp diode forms a switching regulator and hence the associated efficiency.

Also, you have the option of 6X1.2v = 7.2V NiMh or 2X3.6v=7.4v LiPo.
No converison losses.
IMHO it's a no brainer to use 7.2 or 7.4v.

You have two ESCs and hence two BECs.
OK, they're not very efficient, but they won't lose tremendous amounts dropping from 7.2v to 5v. You could spread the servo load and/or keep one as a 'cleaner' supply for the PICAXE.

For some unkown reason you always want to do things the hard way.
Your choice of course, though it would be a good introduction into switchers and how to protect a micro from their noise issues. I've done it myself on a few robots and it does give great versatility when done properly. My latest can take a battery of anywhere between 4.8v and 24v to run a 6v motor at a constant speed irrespective of the battery voltage. However, I build my own switchers so I don't have the expense of buying one.

 

[alband]

2x 3.6v li-polys will only give 2200mAh, and I'm not sure that would be enough.

Each drive motor draws 140mAh (apparently but that is less than a servo )
So:
2x 140 = 280mA
a couple of servos = 2000mA
that's already 2280mA.

Basically, with the previous batteries with 1400mAh, if I moved the drive motors (which I am now replacing with better ones which draw 4x as much current) the current draw was some much that the voltage dipped under load and the whole thing would cut-out.

How can I be sure that the same wont happen with 2200mAh?

I do see you point about ~7.4V would end up with less power loss than 11.1v through a converter. What about two cells in parallel (4400mAh) in series with another two? = 4400mAh with 7.4V. I'm still not sure whether the voltage will dip under load.

I can't use NiMH again, they are just so much bigger than any li-poly and very difficult to fit with a good connection and hold.

You could spread the servo load and/or keep one as a 'cleaner' supply for the PICAXE.

I don't understand (surprise, surprise )

 

[BeanieBots]

6 cell NiMh packs come in just about any size and shape you can think off. Complete with Tamya, DEANS and all other manner of standard connectors.
The most common (and hence cheapest) being made from sub 'C' cells as used by nearly all RC cars. Typically 3000mAhr.

We've already established that your load is ~6A (in round numbers).
The servos might require a supply capable of 1A but they won't draw that continuously. (unless your mechanics are really bad!).

Your drive motors therefore, pull around 3A each.
(considerably higher than your suggested 140mA. Where did that come from?).

Time to stop messing around with numbers pulled from the air and do some design work.
That means having facts and figures at the ready.
If you don't know them, go and measure them.

What current do your motors pull at full load?

You can arrange your LiPo cells in any manner you like. Parallel, series or combination. The only rule is that each 'cell' (could be several in parallel) of a series string must be the same capacity.
There is even a terminology.
5cells in parallel in series with another bank of 5 would be called a 2s5p.

On the whole, it is advised to avoid anything other than 1p because no two batteries ever have exactly the same voltage and they will charge/discharge into each other over time. When done, it should be done by the factory to ensure a good match.

 

[papaof2]

Standard AA cells with "button" ends can have less-than-reliable connections, but cells with solder tabs can be connected together with heavy wire or, if the configuration is right, by soldering the - tab of one cell to the + tab of the next cell (as is done in power tool battery packs). With good soldering, there will always be a good connection. If you want interchangeable packs, make your own with Molex connectors (polarized, high current) for the pack and the circuit board.

A soldered AA pack will deliver more power than the equivalent button-ended cells in typical AA holders because there is less series resistance in the cell-to-cell connections. Most AA holders use relatively small gauge wire for their connections - usually not something appropriate for a load that's measured in amps.

John

 

[BeanieBots]

Absolutely papaof2,
Those sprung loaded AA battery holders are no good above ~1A.
Not only are the springs highly resistive (made of steel to be springy), they get hot.
I've tried to be a cheapskate myself and had them pop their springs out the other side due to overheating and melting the plastic.

Use solder tags or ready made packs.

 

[alband]

Re: NiMH cells. Look at the previous attached picture. When I first started this tank (a good 4 years ago) I hand even heard of Li-poly, so I immediately went to NiMH as the best option. I researched the cell sizes for ages and I can assure you that the best fit of cells is pictured, and it is slightly too big. If different types are use (eg. AA size accross the hull) they either are again just to big or are an irritating quantity too small, loosing a lot of space. Basically there is no way to efficiently fit NiMH cells into it.

I did not pluck that figure out of thin air. http://www.active-robots.com/products/motorsandwheels/solarbotics.shtml
The gearboxes in this page are powered by a certain type of motor, to which a replacement is offered. They also offer an upgrade motor (the one I am planning on getting) which draws 4x the current. The original drew 35mA so this must draw 140mA. As hinted before though, this does seem wrong, but I cant measure it until I have it, and I can't ask them because they will just give the figure I gave.

So, li-polys. It is generally advisable not to put them in parallel unless the manufacturer does it, but it's ok to put them in series. Is there no way of putting them in parallel 2s2p, otherwise I am limited to (I think) 2s1p.

 

[BeanieBots]

I did not pluck that figure out of thin air. http://www.active-robots.com/product...arbotics.shtml

This is where a little common sense and experience comes in.
It's a little bit like someone claiming their car can do 5000mph. You just 'know' that they have 'made a mistake' somewhere.

The currents stated on that site are no load currents.
If what you are saying is true, then upgrading to those higher current motors would increase your run time from 15 minutes to about 5 hours.
Does that sound wrong to you? It does to me.

In very general terms, the size and current capability of a motor will have little effect on the power required to move your tank at the same speed. What will be effected by a more powerful motor is the top speed and acceleration.
It's the same with cars and many fall into the myth that a larger engine uses more petrol. IT DOESN'T when working at the same power level. A two ton 1 litre car doing 30mph consumes the same petrol as a two ton 5 litre car doing 30mph. What makes the 5 litre car use more petrol in practice is the fact that it can be driven harder and thus will be driven harder. Also, a 5 litre car is likely to be heavier than a 1 litre car.

So, we come back to the very vague figure originally calculated by hippy of around 6A current consumption for your motors. Even if you did have a figure for max current, it wouldn't help because you would not be running at full load all the time. You need to MEASURE it.

You can make up a 2s2p battery pack. I simply advise against doing it yourself. Either buy the cells required to make a 2s1p pack of the same capacity or try to find a ready made 2s2p pack.

Something which bothers me a little with the data we have so far is where all this power is going.
How much does this tank weigh?

Please watch this video to get a 'feel' for the amount of power a 1800mAhr 3s1p LiPo is capable of delivering.
The heli weighs 3/4 Kg, the motor is rated at 400W and it will fly like that for about 9 mins.
http://www.youtube.com/watch?v=R5D8in1AwKY

 

[alband]

I actually have that video in my library, but just because it is immense flying. (BTW http://www.youtube.com/watch?v=IXqEY5GASoQ )

Anyway, I have no idea where the power is going. It weighs between 1/4kg - 1/2kg and is quite heavily down-geared (58:1). Tamiya 70097 Twin-Motor Gearbox Kit is the original gearbox and motors I used. The motors are vertually dead now (why I'm replacing them with better ones).
The tracks aren't that hard to move round. It also has a small laser in the barrel? - though I doubt that uses much power and it isn't on all the time.

I've posted a video on "MSN videos" of the "Mk1" version. It isn't great quality and "MSN video" isn't exactly good so I'll work on getting a better quality version onto youtube. The video is about 1.5min long and it is starting to run low by the end of it. If you go to about 1:10 you can see the LED headlamps dimming as I try to move it. I was never able to use full throttle because (I guess) it took so much power that the voltage drop was too much and it just cut out.

I've asked Active Robots how much current the motors draw and m waiting for a reply (although they haven't answered a previous question yet).

I may order 2 of the 220mAh cells and everything else I am getting from them including the motors so that I can do some testing. The P&P looks like it is related to the unit cost so if I have to place a 2nd order it wouldn't be the end of the world.

 

[BeanieBots]

Oh, it's a diddy little thing! Nicely built though.

Something's wrong if it's using all that power
Either motors are shot or battery is DEAD as apposed to flat.
At just 1/2 Kg, it should be able to get airborne and maintain flight for about 20mins with that amount of power

I have a basic wheeled robot which weighs 3Kg. It will roam around for about 6 hours on 6X 'C' 4000mA NiMh cells.

I may order 2 of the 220mAh cells

Typo, or have we downsized by an order of magnitude?

I would expect your existing 1400mA pack to give you a good few hours just based on 'gut feel'. Tracks are not as efficient as wheels.

No point asking the supplier for current specs, you need to MEASURE it.
They don't know what load you are putting on their motor.

 

[alband]

yep, very diddy and thanks.

I suspect it is the motors. When I try them with the ESC (using a PICAXE to give the servo pulses) I can only use variables between ~118-130 and they go VERY slowly.

Yes type

How exactly do I measure it when I have the motors? I've got a DVM (s'that right?) that has a few amp settings, just put it in series with the motors, right?

The youtube video is being processed a the moment. If it comes out at the same quality it went in it should be very good.

 

[BeanieBots]

Put your DVM on 10A setting. Probably need to move the lead connections as well.
Then fit in series with motor to measure current. (ideally once installed in the tank and running 'normally').
Probabaly also worth measuring current out of battery as well.

Don't forget to put your DVM back to volts when done or you'll be in for a surprise (and a new fuse) the next time you measure volts with it.

I can only use variables between ~118-130 and they go VERY slowly.

I don't understand. Why the limit of 118 to 130?
What happens outside that range?

 

[alband]

I don't understand. Why the limit of 118 to 130?
What happens outside that range?

Sorry, should have said. It's the same as when I use to try and drive the tank too fast - they just cut out. Basically I'm very sure the old motors are kaput.

Much better quality: http://www.youtube.com/watch?v=5EVQNotj8Zw
(using a friends account so if the name changes to some kind of insult or something - apologies )

 

[BeanieBots]

Hmm... something very odd going on there
Listening and obvserving the movements does NOT suggest any motor issues.
If it's cutting out at higher PWM values, it suggests that the ESC might be underated for those motors and/or a damaged (or wrong spec) battery.

If it's an ESC designed for aircraft, they often have a 'soft' power limit for low battery conditions which serves two purposes. One is to give the pilot advance warning of low battery, the other is to prevent excessive battery drain to avoid damaging the battery.

I think you need to do some detailed bench testing of what you already have.
Get your ESCs to drive some fixed loads such as light bulbs and find out what is going on. Observe behaviour with low input volts and high 'stick' demands.

My 'gut feel' (from the video) is a system/design failure rather than a motor failure.

 

[alband]

Well the ESC's were originally designed for blimps. On one side of the board they have a small chip (microcontroller of some kind to process the servo signal) and some discrete components, and on the other side there are two N-channel MOSFET's and two P-channel MOSFET's. I had to replace one of the MOSFET's at one time because I shorted something for too long and it fried, but I should be able to find their spec...

http://search.digikey.com/scripts/DkSearch/dksus.dll?lang=en&site=US&WT.z_homepage_link=hp_go_button&KeyWords=IRF7832-ND&x=0&y=0
http://search.digikey.com/scripts/DkSearch/dksus.dll?lang=en&site=US&WT.z_homepage_link=hp_go_button&KeyWords=+FDS6679AZCT-ND&x=0&y=0
So both 30V, >13A > 1W
I can do those test though. How do I do the light bulb bit though. How do I know what current it draws?

The motors are definitely, oh... not dead.

I just put 5V across them and they're fine.

I've also just put an LED to one of the ESC's and it reacted the same as the motors (cutting at higher values), but when I took it off the ESC, it burnt me.

 

[BeanieBots]

Replace motor with bulb. Put DVM (current setting) in series with bulb.
LED on ESC output with no resistor? Could blow ESC and/or LED

If the ESC really is rated at 30v,13A, it should be more than enough.
Sounds like they have some 'history'. Maybe duff?
Do they both behave the same?
Are they happy with an unloaded motor?

 

[alband]

What kind of bulb, miniture or real?

That will explain why it got how then.

I'm having a good look at the circuit in the ESC now. It would be unfair to hack it then post it on the web so that everyone could copy but I can PM an individuals with what I've found.

It's just the MOSFETS that are rated at 30V etc.

They do have a history of having to deal with the number of times I've shorted the batteries, but surely shorting the batteries should kiss the ESC's (although it doesn't exactly please the batteries).

Urgh, I really don't know whats going on now. I just tested them again and they seems to work up to full speed. They are very picky though. I still cant turn them use higher numbers like 225 and 75 without it cutting.

#PICAXE14M
servo 4,150
main:		readadc 0,b0
		let b0 = b0 MAX 225 MIN 75
		if b0 = b1 then main
		let b1 = b0
		servopos 4,b0
		goto main

That's the code I'm using to test them. I start with the POT in the middle (90deg) and have to slowly wind it out to about 130deg to get the motor going. If I turn it too fast it cuts out, if I got too far it cuts out, if I leave it too far for too long it eventually cuts out.

I've done some current test with the motors unloaded (although it is quite hard because of the constant cutting) and the current rises while I increase the speed until the current reaches about 0.37A (highest number I've seen on it) then the motor cuts and it fast to 0.

 

[BeanieBots]

The bulb (or resistor) should be such that it represents a similar load to the motor.

Sounds like your ESC(s) are faulty.
I'd expect them to cope with >.37A.
Besides, we already suspect a loading of ~6A, so where is that going if it's not into the motors.
Are your ESCs getting hot?
The outputs may have a duff transistor and are just shorting rather than driving the motor.
What is the current going into them compared to the current comming out?

Airborne ESCs often have overload protection. This manifests in the way you describe. It is desinged to cut power in the even of a stall of the prop. It may well be that those ESCs simply can't drive your tank motors.

So, you've shorted the batteries a few times
Could well be duff batteries then.

You will need to learn not to do that if you are thinking of using LiPo.
You'll only do it the once with them, then you WILL have learned.
(tip: have bucket of sand close at hand, and hand not so close!).

At the moment you have too many unknowns.
You need to sit down, take your time and do some quality testing.
Otherwise it's all guess work.

My opinion at the moment is that your ESCs are not suitable for the task and that your batteries are duff.
I need you to do tests and give figures to PROVE it one way or the other.
Until you can do that, there is little point going any further.
Throwing other batteries and motors at it won't get you anywhere.

Do you still have the motors from the blimp?
If you do, (with the prop on the motor),connect it to a voltage source the same as the original and measure the current drawn.

 

[alband]

How do I know if the bulb will be a similar load?

I've tested them both again, and they now seem to be working fine. They do still cut occasionally, but only when the motor burps (if it was an engine I'd describe it as a misfire).

They aren't hot at all.

I suspect these do have that which could be what is causing them to cut, but the cut is induced by the "burp", or if I hold the axle still (know as stalling?).
@BB check your PM regarding this, I've reverse engineered the ESC to the best of my abilities, but it wouldn't be fair to post the findings (I can send to anyone else if they request).

Yep, I have, but the reason has always been the shoddy wiring job I did because I had so little space. I should have more space in this version (I've shrunk the gearbox considerably + li-poly are smaller than equivalent NiMH). So There shouldn't be any shorts this time. However I do have a basin near by!

Test results:

DVM in series with motor:
~0.15A when idling; stalls a lot here so can't get a steady figure
0.23A at max speed; stalls eventually when motor burps
~0.37 when accelerating; figure varies a lot and is never more than 0.5A but it stalls most when accelerating.

DVM in series with ESC's supply (+):
roughly 0.02A more in each case.

Whenever the motor burps and, by chance, it doesn't cut, the reading flickers to about 0.4A.

How do I check the batteries? They are all at ~1.3V with no load, and charge fine.

The ESC's were DESIGNED for blimps (hence the reasoning behind a small, bi-directional ESC). I didn't hack them from a blimp (not that I wouldn't do that ).

 

[BeanieBots]

This is begining to sound like the ESCs aren't up to the job.
There was no "attached" in your PM but the description sounds like a self resetting overload protection circuit.
The FETs are cabable of more, so there might be scope to modify for higher current but I would need a full circuit to advise further.

Can you get hold of a spec for the ESCs or do you have a product number?
Seems they are only good for around 300mA but you require >0.5A.

BTW, DON'T put a flareing LiPo in your basin. Any water will make it flare even more. Use sand. (lithium + water = fire!).

With that level of loading, even running contiuous, you should get about 2hrs from your 1400mAhr battery. So this is suspect too.

Fully charge the battery. Put a load on it (resistor or bulb or combo) such that it pulls about 0.5v.
Keep an eye on the voltage.
Stop when the volts gets to 1v per cell.
Note how long it takes. (should be ~3Hrs).

Remind me, how many cells are you using at the moment?
You need at least 5 for the ESCs to work properly.

 

[alband]

Sorry, I forgot to attach it. It was only a layout of the FETs so I'll try to get a full schematic together.

The ESCs are home made by a guy in the US (posted the link a bit ago) so there are only the slight bits of info from that site which I've already quoted and unfortunately there isn't a datasheet.

Lilac flame?

I had already remove the cells and guts to start again before I started this thread but I agree they should last give only that much draw (even when quadroupled to include the second motor and the two servo motors). The only other things that used amps in the previous version were two SMT 08Ms and four LEDs.

I'll test those this evening but they can't be much more than a few mA?

6 cells.

 

[alband]

I've finished the schematic in a diptrace file, but there is now way of sending it to you?

 

[BeanieBots]

Send it by e-mail.
I don't have diptrace so please convert to jpg or similar prior to sending.

 

[alband]

This is the page for the ESC again: http://www.homefly.com/proddetail.asp?prod=e8

 

[BeanieBots]

Alband, that link states it's 8A.
Assuming your input voltage is within spec, (you never answered that one), and your DVM is not over an order of magnitude out on current, then it's very simple.
Faulty ESCs. Send them back.

 

[alband]

7.2V.

It only has a 10A current option and it displayed the numbers given.

To america?

Can you tell which bit of the circuit is the current protection bit? It might then be possible to remove it?

Check your PM.

 

[BeanieBots]

If you'd like to post/send it as .jpg or something I can open (not .dsn) then I'll have a look.

7.2v might be the nominal voltage of your battery but is it still within the ESC input voltage spec when the ESC is loaded to the point of cutting out?

PS. Nothing new in my PM
PPS. Got your e-mail, will have a look and get back to you later.

EDIT:
Had a good look. (strange way of drawing an H-Bridge, but never mind).
Those 3 pin devices are level shifting transistors for driving the P channel FETs.
I can see NO overload protection circuitry.

You can follow the current path yourself quite easily. (a bit easier with a more conventional layout).
From line in, through P type FET, through motor, through N type FET down to ground.
Same path except alternate FETs when motor is driven the other way.
There is NO current sense, hence no overload protection.

Any overload in the past will have destroyed one or more of the FETs.

What were the results of your battery test?

 

[alband]

You're right!

I check the voltage an every time it cut it went bellow 4.9V (I think my DVM is a little optimistic).

Anyway, I switched to a higher supply and it works almost fine.

The motor still "burps" and very occasionally it cuts from a "burp".
Any ideas what could be causing these burps. If its the motors then I would definitely buy the better ones.

The current draw (at the power supply) is still no more than before so I don't know where the current is going. The headlights take up less than 0.4A so they aren't suspect. I highly doubt it was the small PICAXE circuits.