COMBATT: THE NICAD DISCHARGER

SOME OF THE COMPONENTS MAY NO LONGER BE AVAILABLE!

It's a few years since the COMBATT discharger was first developed (pre-1993) and I guess the electronic side has moved on a bit since then but here it is in any case. Any programmers out there please forgive the poor programming in Basic but we are amateurs and have no training. If anyone takes the source code and does something better with it, let's know about it, we'd like a copy. It you decide to have a go and make one ONLY get in touch to tell me it's working perfectly (!) since I can't offer any sort of repair or information service.

YOU MAKE AND RUN THIS DEVICE AT YOUR OWN RISK, WE DO NOT ACCEPT ANY RESPONSIBILITY FOR DAMAGE TO YOUR COMPUTER THAT MAY OCCUR.

Design and program by Graham Woods and David Woods

HOW IT WORKS

The NiCad discharger is quite simple in its operation, your NiCad pack is simply discharged through a large 50 watt resistor. As the NiCad discharges, your computer 'samples' the voltage of the battery pack on load, and marks a series of red points on your screen, i.e. it draws a line. At the point when the pack voltage just dips below 1 volt per cell (4 volts for 4 cells, 8 volts for 8 cells and so on) your computer will automatically stop the discharge of the pack. In this way the NiCad pack is 'flattened' safely without going too far. Of course, you could carry out exactly the same procedure yourself with a voltmeter, a pencil and a piece of paper but you would have to monitor the discharge yourself, so why not let your computer do it for you?

The COMBATT discharger has one other important advantage - it records all those points as it plots the discharge graph on screen and saves the voltages as a file. And once you have a set of files on the same, or similar, NiCad packs you can compare them on screen or off, that is to say, monitor the performance of all your NiCad packs whether they be receiver batteries or electric flight packs. This is where the name COMpare BATTeries comes from, BTW.

WHY DO I NEED TO MONITOR NICAD PERFORMANCE?

You don't, but if you have a mind for safety, your pocket and your time, it is wise to have a model aircraft that will not crash for lack of care; batteries can suffer through misuse and second only to poor piloting as the reason for crashes (IMHO). If you fly electric, then you will naturally enough always be concerned with cell performance as well, especially if you have a large number of expensive NiCad packs. Looking at the shape of the discharge curves may also help you to identify faulty packs too.

NICAD BATTERY CYCLING

We've all done it - flown for an hour on Sunday afternoon, forgotten about the model until the next Saturday evening and then put the transmitter and receiver batteries on charge all night. They still work but as you know from various magazine articles, repeated 'topping up' of Nicads without discharging them first does them no good at all - they apparently can develop some sort of memory. My own transmitter cells suffered from this very phenomenon; in the end my tranny ran only for an hour and a half before the bleeper sounded for a flat battery when it should have gone on for much longer.

Discharging NiCad cells fully and then recharging (cycling) them can break this nasty 'memory' habit. Using the COMBATT unit enables you to cycle your cells simply and compare the results as you go through a number of charge and COMBATT discharge cycles. The performance of cells often increases after they have been cycled only one or twice, see my graphs below.

NICAD CELLS

The charge and discharge rates of NiCad cells are expressed as multiples of their nominal capacity 'C' - the hourly rate of discharge. Thus, if a cell has a capacity of 1.2 Ah then charging or discharging at a rate of 1.2 amps for one hour, the 'hourly rate' would be called 1C. Consequently, discharging at five times the hourly rate, 5C - the cell would last for only around 12 minutes and discharging at a fifth of the hourly rate, C/5 - the cell should last for 5 hours or so.

Discharging cells at these different rates results in capacities of NiCad cells departing from their nominal values. A cell discharged at a low rate will obviously last longer than one which is discharged at a higher rate but the absolute capacity of the cell also changes according to its discharge rate (and the way in which it is charged). The capacity varies so that discharging at 1C should give the cell a capacity to 100% of its nominal capacity, the higher 5C discharge rate will yield a lower capacity of and slow C/5 discharge rate may well show a higher capacity than the nominal capacity - for example 105%.

In order to measure a realistic set of capacities for a range of different pack voltages and cell capacities large wattage (expensive) discharge resistors would have to be tailored to suit each and every battery pack. Not only that, each resistor would have to have its own associated electronics as well and the circuit would be more costly. The COMBATT discharger gets round this problem because it is ONLY DESIGNED TO COMPARE the capacity of similar NiCad packs or the same NiCad packs over and over. For simplicity (chosen, bearing in mind the time taken for discharges) it uses one 10 ohm, high wattage resistor to discharge all your Nicads. All discharges are made through the same resistor, under the same conditions. (and hopefully the same wires and connectors)

Because of what I have written above, and the fact that batteries are not completely flattened, the stated capacity that shows on the screen will not therefore be the absolute capacity of the pack. It is a figure for you to compare the next time you discharge the same pack or a similar pack.

Let's look at an example:

Your TX Nicad may be 8 cell/600 maH - discharging through a 10 ohm resistor means the hourly rate will be 1.6C and may show quite a low reading of the nominal capacity, say for example 85% or 90% of the nominal capacity. If, however, you are discharging an 8 cell/1800 maH electric flight pack the C rate will be 0.53C and may well give a higher reading, say over 100%, of the nominal capacity for new cells. Some good packs can show capacities of well over 100%.

The way in which you charge your cells also has an impact on the 'capacity' of your NiCad packs - a long overnight charge at C/10 + 40% or 50% (the normal way RX & TX chargers charge) will often give you a higher capacity than with a fast, high current, Delta peak charger. Then again, charging electric flight cells with a temperature controlled charger, getting your cells sizzling up to 45ºC, will render a different capacity again. The temperature of the pack during charge and discharge also has its effect on its capacity. A value of battery capacity then, doesn't really mean very much considering all the variables that packs of cells are subject to. Comparisons of similar packs of cells or the same pack of cells over and over is more meaningful and the COMBATT discharger enables you to do it more easily.

RESOLUTION

When/if you have been playing with the data files (plotting graphs) you may notice that the same voltage values (or numbers) crop up time and time again. This is due to the fact that the COMBATT unit uses an 8 bit analogue to digital converter; because of the way the circuit works the NiCad packs are sampled in 60 mV (0.06 volt) steps. Greater resolution (i.e. smaller voltage steps) could be obtained if a 10, 12 or 16 bit A/D converter were used but some of these I.C.'s are much more expensive (tens of pounds each at the time of writing a couple of years ago), sometimes need an extra 12 volt power supply, and are not necessary for this simple circuit. For the same reason, the graphs of low voltage and higher voltage packs will appear slightly different on the screen. The time taken to discharge various packs will depend on their nominal capacity. The program does a calculation to divide the estimated total discharge time by 600 to give up to 600 values of sampled voltage. The sampling interval is shown on the screen.

Those of you experienced in electronics will see that not all the current from the discharging NiCad pack goes through the discharge resistor some it is used to power some the circuit itself. While the IC1 is powered from your computer's power supply, current for other devices is derived from the discharge NiCad itself. The Mosfet TR1 is of the same type as is used for older electric speed controllers, has a very low resistance and drops very little voltage and, for all intents and purposes, hardly affects the capacity readings. D3 and IC2 are there to provide some protection to your computer should you do something silly.

HOW MANY CELLS WILL THE UNIT HANDLE?

The COMBATT unit was originally designed to discharge model aircraft 4/5 cell receiver and 7/8 cell transmitter NiCad battery packs. It will, however, handle from 4 to 10 cells with capacities from 500 mAh to 1800 mAh (or higher) - a range that covers most needs. 225 mAh cells can also be discharged but the 'hourly rate' will be around 2C - although this may be too high a rate for some of these smaller NiCad cells. Not recommended for 110mAh cells either.

FILE MANAGEMENT

As mentioned earlier, the discharge data (up to 600 values) is recorded on file with the filename that you give the NiCad pack. I suggest you use the name of the model and a number. e.g.. Cherry1 for the first time you discharge the pack. The data will be saved as: A, B, C, D, E or F *:\Combatt\Cherry1.dat depending on the drive you choose. The second time you discharge this pack you would generally opt for the name Cherry 2 and the data will be saved as a completely new file. In this way similar files are grouped together in the file list.

GRAPH PLOTTING

The created data (.dat files) can be imported into many programs such as Lotus 1-2-3 or Lotus Freelance Graphics, Paradox, Excel etc. to produce customised graphs. On the other hand, if such programs are not available to you, there are various Shareware alternatives. The other way to obtain a hard copy of the graph is simply to use the PRINT SCREEN option on your own computer or use ALT + PRINT SCREEN and grab a copy of the screen to the Windows Clipboard and paste it into a graphics program.

CIRCUIT CONSTRUCTION

Making the COMBATT UNIT is quite an easy thing to do for most modellers but of course there are the usual warnings that go with static sensitive devices. Use a small, fine tipped soldering iron, not heating the components too much - do not solder all IC pins one after another! Rotate the soldering. Make sure you get the polarity and/or alignment of all components correct before soldering - see circuit diagram for pin-outs. (Clean your PCB with a scouring pad and use a 1mm drill for the holes.)

These components should be available in the UK from Maplin, Farnell, CPC, and R.S.

The large discharge resistor is mounted on the heatsink which is fixed on the outside on the top of the unit using 3mm nuts and bolts for mounting both heatsink and resistor together - use thicker wires for the connections to R R on the PCB for they may carry more than 1 amp. [Warning: The discharge resistor gets very hot with larger packs (8, 9, 10 cells) since it has to dissipate up to 20 watts.]

• The PCB soldering is simplicity itself - use the socket for the 8 pin IC - pin 1 (the round mark) should be at the bottom left hand corner (looking from the component side).
• IC2 (6 pins) has the semi-circular mark on the left (component side)
• Get the diodes D1, D2, D3 and capacitor C1 round the correct way - the black band is +ve.
• Resistor values: R1 - brown, green, orange; R2 - black, brown, orange;
• R3 - gold coloured; R4 - orange, orange, red
• TR1 has three leads - g (gain), s (source), d (drain) - observe correct polarity.
• REF1 (the smaller of the two three pin devices) has the two negative outer leads connected together.
• Use the ten PCB pins for flying leads: 6 to the Centronics 36 way, 2 to R3, 2 to +ve & -ve.
• The multiturn potentiometer needs to be wound in a few turns from the screw end so that the ratio of resistance is about 1:5

SETTING UP THE UNIT

• In order to calibrate the COMBATT discharger you will need a digital multimeter and do a simple calculation.
• When you are sure the circuit wiring is correct, connect a 4.8v NiCad pack to the unit and measure its voltage across the +ve & -ve terminals of the unit.
• Now do this simple calculation: Calculated voltage = Measured NiCad Voltage divided by 6 (for example: 5.13 /6 = 0.855).
• Next plug in your computer printer cable into the Centronics socket to power up the circuit from your PC. Do not run the Combatt program!
• Locate pin 2 of IC1 (where the centre pin of the pre-set goes). See circuit diagram.
• Put the red multimeter probe on pin 2 of IC1 and the black probe on the -ve terminal.
• Twiddle the pre-set resistor, VR1, slowly until the voltmeter reads your Calculated Voltage (above). In the case of our example above: 0.855 volts or as close as you can get to it. The resolution of the blue Cermet pre-set VR1 should enable you to get very close to your calculated voltage.
• Exact comparison of the NiCad voltage on screen with the voltage using a multimeter is not very meaningful since the screen voltage will be changing continually in +/- 0.06 volt steps. Try and get as close as possible when setting up - have a few goes at doing it.

COMBATT - THE NICAD DISCHARGER
page #2

RUNNING COMBATT

When you have finally completed your COMBATT unit please do double check all you connections, remember you will be connecting it to something worth hundreds of pounds and I cannot take any responsibility for any damage you may cause to your computer, equipment or Nicads.

There are a couple of safety devices to protect your equipment and we are only working with up to 12 or so volts, so there shouldn't be any problems.

DISCHARGING A NICAD

• Unplug your printer from the parallel cable and plug in your COMBATT unit.
• Always use the same leads for your discharges, use low resistance (thick) wires, keep them as short as possible.
• Plug in your NiCad, get the positive and negative connections round the right way - if you have a charging socket to use for your battery connectors, so much the better.
• Run the program. Pick the drive (A, B, C, D, E, F) where you want the file to be stored.
• Choose Discharge (D). Answer the three questions about the pack and press ENTER each time, the last ENTER will start the discharge.
• Remember to enter NiCad Capacities in Ah and NOT mAh
• When the COMBATT discharger is running you should immediately one red pixel on the left of the graph screen by the axis line near the top.
• The program will have estimated the time of discharge for a perfect NiCad from the numbers you have entered: two clocks will be running - a count down and count up. These are only to be used as a guide.
• The number of cells of the Nicad pack and drive will be displayed.
• The % so far of nominal capacity will be displayed. The measured voltage of the NiCad on load will be displayed (remembering that the voltages go in steps of 0.06 v.)
• The measured capacity in mAh will be displayed.
• As the program slowly samples the maximum 600 values of battery voltage a line plot of the voltage/time curve will appear in red.
• Every so often data will be transferred to your hard drive or floppy disc and the drive light will flash briefly.
• When the on-load voltage of the NiCad just dips below the 1 volt per cell value (for two samples) the program will stop the discharge and a horn will sound.
• After the horn has stopped the time and date taken from your system clock will be displayed so make sure you have set the time and date correctly.
• The time, date, measured capacity, nominal capacity, sampling interval are added on to the end of the data file. (the nominal voltage is added as the first number in the datafile).
• If you subsequently use a graph plotting program for your data these values will either have to be removed manually or with the range settings of the graphing program. [If you don't the ends of the graph will appear in a funny way!] Use Notepad in Windows.
• If you press ESCAPE there will be a few seconds delay before it comes in, you will then be given three options - Discharge, Graph, Quit.
• Aborted discharges although stored, in part, on disc cannot be subsequently be displayed on screen and should be erased using a text editor or the program itself.

TYPICAL SCREEN SHOT OF COMPLETED DISCHARGE

The screen shot below shows the classic shape of a NiCad discharge although different types of cells (SC, SCR, SCE etc.) will exhibit slightly different shaped curves. SCR type cut-off cells show the classic (vertical) cut off with the pack voltage dropping off abruptly (see second graph below), other normal (let's say) NiCads like the one below trail off as they reach the cut off point. Some of my maltreated batteries exhibit strange behaviour as they approach the cut off voltage showing a slight recovery, this is probably due to the performance of individual cells not being matched. Nickel Metal Hydride cells have similar graphs, BTW.

DISPLAYING YOUR COMBATT VOLTAGE/TIME CURVES ON SCREEN

Choosing the option for Graphs (G) will allow you to display up to eight of your previous files simultaneously exactly in the way they first appeared on screen during discharge.

Simply type in the name or number of the NiCad pack you used to identify the discharge, press ENTER, and the line will plot on screen. To compare another Combatt file, type a second name, press ENTER, and that will appear on screen as well, and so on up to eight different files in different colours.

Other data about the discharge is also retrieved from the files and displayed on screen in the form of a table. See screen shot.

If you press RETURN (enter) on its own at any time while in the graph screen you will be able to opt for Discharge, Graphs or Quit.

The second screen shot above shows the graph page. Here I have recalled 3 discharges of the same pack (3 for clarity but it could be up to 8) and placed them all on screen together. Each discharge line and its data are displayed in a different colour. All the information you need to know is stored in the files and displayed on screen in the form of a table. In this example you can see the apparent pack recovery mentioned earlier quite clearly.

LOOKING AT FILES

Opting to inspect your Combatt Files (F) is quite a useful feature if, like me, you have a lot of discharges stored. Once you have been using your discharger for a little while, keeping track of what number you reached with different packs becomes difficult. This function simply locates all files with the .dat extension in the Combatt directory and displays them on screen with the date and file size.

ERROR HANDLING

• There are several timed error handling routines and screens in the COMBATT program.
• In discharge mode, if you try to name a new file with a filename you have already used for a discharge the error message FILE ALREADY EXISTS will appear. Your existing COMBATT files will be displayed on screen. You then have the option of erasing such files if you wish.
• In graph mode, if you try to display a COMBATT file that doesn't exist the error message FILE NOT FOUND will appear. Your existing COMBATT files will again be displayed on screen.
• Using more than eight characters or illegal characters (* ? / " ; > : ETC) you then have the option of trying again.
• If you try to store or retrieve information on disk and there is no disk in the drive the message THERE IS NO FLOPPY DISC IN DRIVE will appear.
• If there is no directory called COMBATT on floppy or hard drives (or you chose a non-existent drive) the program comes up with the error message PATH NOT FOUND or THERE IS NO DIRECTORY CALLED COMBATT. You will then have the choice of creating a directory or changing disk drive.
• If there is any other type of error the error message UNDEFINED ERROR will appear and the program will close in five seconds. (An example would be if you tried to display an aborted COMBATT file - such files cannot be displayed for they contain insufficient information.)

In practice, you may, of course, like to initiate the first two error messages yourself in order to view all of your previous COMBATT files on the disk/drive you are using instead of going to one of the other file listing/editing facilities to be found on your computer.

NICAD ANALYSIS

So how do we interpret our Combatt discharges apart from the obvious capacity readings? There are several things to take note of, the first being the method of charging. Going right back to the beginning: there are three ways of charging a NiCad battery:

1) Fast charge - using voltage detection to stop the charge,

2) Temperature charging - a variation of 1 except that the temperature of the cells is used to abort a charge and

3) Slow or trickle charging.

Slow charging (e.g. C/10 + 50%) generally gives the best results, your cells may receive a higher charge than with a fast charge. Voltage and temperature detect fast charges can give an inferior charge so remember how the NiCad was charged when analysing your discharge graphs. Temperature also has an effect on charging. Compare Like with Like. Always use the same leads for comparison discharges for wires and connectors have resistance of their own. Try charging your NiCads in different ways, at different rates, or for different times to find the best charge. If you have other NiCads, such as for a Camcorder, CB, CellPhone, Ham radios, Scanners or Laptop give them a go too.

What does the graph of a good discharge look like? Well, first off it should be smooth and not like the one I have shown above - the one shown is obviously faulty in some way because of the shape of the curve - that little bump at the end. The second thing is to look for a graph where the discharge line plot stays above the nominal voltage dotted line for as long as possible - for at least 50% of the time; i.e. it holds up under load. It should also be as high above the nominal voltage line as possible.

If you think about, it the area under the curve is the measure of capacity. [The vertical axis, although recording voltage, is, in effect, also recording the current flow through the fixed 10 ohm discharge resistor (E = I x R e.g. 8.4 volts = 840mA x 10 ohms) and area = length times height so the area under the curve is a measure of the capacity - i.e. mA times time in seconds/60 x 60 mAh.]

If we were discharging at constant current we would be looking for a graph that reached as far across the screen as possible but we're not. In our case though, the discharge current is not constant but varies from high to low with the voltage so that we are not merely looking for a discharge that reaches as far to the right of the graph screen as possible - we are looking for a graph that covers the maximum area under the curve. We could have a situation where the capacity of two NiCad packs reads the same but one has a 'higher' line plot and yet only reaches part of the way across the screen that the other one does with a 'lower' line plot. The better pack is the one with the higher line plot.

AND FINALLY...

I hope that you do not have any problems with the construction of the COMBATT unit or the program. The unit has a small electronic component count (14) and is easy to make for experienced modellers with a small soldering iron and simple tools. More sophisticated versions, a greater number of cells and/or better resolution could have been made but it was felt that keeping the whole thing as simple as possible would mean a low price. Increasing the number of cells that can be discharged increases the problems of heatsinking and large resistors. Higher discharge rates would also bring the same problems. Lower rates of discharge would mean that packs of cells would take 'forever' to discharge. In writing our quirky program we hope almost every eventuality has been taken into account. We welcome any positive feedback you may have.

The COMBATT unit is a sensitive device and can be subject to interference under certain conditions. Use a short length, quality printer cable rather than a ribbon cable for connecting the unit to your computer. (You may, though not always, encounter problems if you use a T-switch to switch between computers and printers since cable lengths are effectively doubled.)

COMPONENT SUPPLY...                 www.maplin.co.uk            www.cpc.co.uk            www.farnell.co.uk

DOWNLOAD COMBATT PROGRAM

COPYRIGHT © 1993-2000 Graham Woods All rights reserved.