Is it better to build packs based on runtime or average voltage?
12-29-2003, 01:31 PM
#2
Re: Is it better to build packs based on runtime or average voltage?
Originally posted by AWOLsoldier
Inquiring minds want to know.
Inquiring minds want to know.
12-29-2003, 01:43 PM
#3
Really, you should try to get them as close as possible, both runtime and voltage, and IR. If yer running stock, I would go IR, voltage, then runtime.......Mod, flip it around and go with runtime first.
And runtime is an issue in modified.........
Later EddieO
And runtime is an issue in modified.........
Later EddieO
12-29-2003, 01:44 PM
#4
Well, I would say runtime-TO a degree.
Lets compare two fictional packs.
#1
1.170 430 sec
#2
1.174 402 sec.
Which pack would I choose? I would choose #1. I say this because the voltage is being averaged over a much longer time. This tells me that at say 3 minutes, pack #1 will more than likely have higher actual voltage than pack #2. Think of it this way, which pack has higher actual voltage at 415 seconds?? Obviously #1, cause #2 is allready toast. Its interesting, in my experience, even at 20 amps, the average voltage #'s are not THAT much higher (they are higher, but not a whole lot) This is because the cell spends so much time discharge BELOW the average, that is pulls the average down with it.
The GP cell has interesting discharge characteristics. Towards the end of their discharge, they drop much more rapidly than Nicd's do. This makes them show a higher average because the cell can be going along nicely at 1.16-1.15 towards the end of its discharge (say 380 seconds for #2) then in the last 20 seconds the voltage drops off dramatically. NiCds' seemed to maintain the ability to put out SOME voltage under high load, where as GP3300 losses its ability to produce current under high load, but will take an hour on the discharge tray. when you take a NiCD down on the 20 amp lights, then put it in the tray, its pretty much out of juice, but the GP seems to maintain the ability to produce voltage under a small load, but hook up the lights and NOTHING.
I do think that GP3300 is the best batt ever for regular RC car racing.
Lets compare two fictional packs.
#1
1.170 430 sec
#2
1.174 402 sec.
Which pack would I choose? I would choose #1. I say this because the voltage is being averaged over a much longer time. This tells me that at say 3 minutes, pack #1 will more than likely have higher actual voltage than pack #2. Think of it this way, which pack has higher actual voltage at 415 seconds?? Obviously #1, cause #2 is allready toast. Its interesting, in my experience, even at 20 amps, the average voltage #'s are not THAT much higher (they are higher, but not a whole lot) This is because the cell spends so much time discharge BELOW the average, that is pulls the average down with it.
The GP cell has interesting discharge characteristics. Towards the end of their discharge, they drop much more rapidly than Nicd's do. This makes them show a higher average because the cell can be going along nicely at 1.16-1.15 towards the end of its discharge (say 380 seconds for #2) then in the last 20 seconds the voltage drops off dramatically. NiCds' seemed to maintain the ability to put out SOME voltage under high load, where as GP3300 losses its ability to produce current under high load, but will take an hour on the discharge tray. when you take a NiCD down on the 20 amp lights, then put it in the tray, its pretty much out of juice, but the GP seems to maintain the ability to produce voltage under a small load, but hook up the lights and NOTHING.
I do think that GP3300 is the best batt ever for regular RC car racing.
12-29-2003, 02:29 PM
#5
i think the question is more about the importance placed on matched #'s versus high numbers. ya'know?
i would think that matching based on runtime, or capacity would be more beneficial since each cell would want to peak at or near the same time, leaving no cells with a partial charge. would the repercussion to matching this way would be that the pack may only stay matched for a short while since the voltages of each cell may push charge back and forth among its neighboring cells? i dunno.
total average voltage is just that. add em up, divide, and that's your average pack voltage. if i had a choice between a matched 1.16/400sec, and six cells that were between 1.16 and 1.19, all at 400sec, i'd choose the mismatched votage pack since the average would be near 1.175 (middleground) and the pack would still dump hard and steep.
it just seems clear to me that if you have cells all over the place for capacity, you'll never run with a full charge regardless of what extreme voltages the cells may have, unless you charge each cell individually. what's the point of high numbers if the cells don't get charged completely. the first 20 seconds of the discharge contribute heavily to the numbers. w/o that initial section of the discharge curve to average, you'd be surprised how bad a pack will look.
this is a good thread. and i sure hope more people, with more knowledge on the subject than i have, stop in.
i would think that matching based on runtime, or capacity would be more beneficial since each cell would want to peak at or near the same time, leaving no cells with a partial charge. would the repercussion to matching this way would be that the pack may only stay matched for a short while since the voltages of each cell may push charge back and forth among its neighboring cells? i dunno.
total average voltage is just that. add em up, divide, and that's your average pack voltage. if i had a choice between a matched 1.16/400sec, and six cells that were between 1.16 and 1.19, all at 400sec, i'd choose the mismatched votage pack since the average would be near 1.175 (middleground) and the pack would still dump hard and steep.
it just seems clear to me that if you have cells all over the place for capacity, you'll never run with a full charge regardless of what extreme voltages the cells may have, unless you charge each cell individually. what's the point of high numbers if the cells don't get charged completely. the first 20 seconds of the discharge contribute heavily to the numbers. w/o that initial section of the discharge curve to average, you'd be surprised how bad a pack will look.
this is a good thread. and i sure hope more people, with more knowledge on the subject than i have, stop in.
12-29-2003, 02:44 PM
#6
Tech Adept
Good point Seaball,
I connect my discharger to my PC, or you could measure the voltage every 3 or 5 sec. and take notes.
I start at 0 and stop at 320 sec, because I race 5 min.
Build a graphic on excel and put the curves together.
The bigest AUC (area under curve) will be your best pack.
I connect my discharger to my PC, or you could measure the voltage every 3 or 5 sec. and take notes.
I start at 0 and stop at 320 sec, because I race 5 min.
Build a graphic on excel and put the curves together.
The bigest AUC (area under curve) will be your best pack.
12-29-2003, 05:43 PM
#7
gearless,
i do the same thing with a cheap digital multimeter from radio shack. it has a serial 232 output and the program can be set to take readings every second. i've got lots of graphs that don't mean much.
it's quite nice. and now i can acutally get some kind of ir #s too with my integy 16x5v6.0. i don't really know how the integy #'s compare to actual ir printed on the cells, but from a relative standpoint, it lets me see how my packs are doing w.r.t. time, and each other.
it's a good idea to record down to 5.4v or so, and let excel do the averaging over the window that you want. i always like to keep as much data as i can for future analysis, especially for those times when foresight escapes me.
oh yeah, and who said there's no place for calculus in the rc world??
i do the same thing with a cheap digital multimeter from radio shack. it has a serial 232 output and the program can be set to take readings every second. i've got lots of graphs that don't mean much.
it's quite nice. and now i can acutally get some kind of ir #s too with my integy 16x5v6.0. i don't really know how the integy #'s compare to actual ir printed on the cells, but from a relative standpoint, it lets me see how my packs are doing w.r.t. time, and each other.it's a good idea to record down to 5.4v or so, and let excel do the averaging over the window that you want. i always like to keep as much data as i can for future analysis, especially for those times when foresight escapes me.
oh yeah, and who said there's no place for calculus in the rc world??
12-30-2003, 02:23 AM
#10
Tech Adept
Hi Lee82gx,
It's easy with my GFX
. With other chargers/dischargers I don't know.
But never mind, what you will need is a digital watch and a multimeter. Just take the voltage readings every 3 (if you are enough fast) or 5 sec. Wrote them down and build a excel graphic (time vs voltage or inverse if you want).
The AUC is build on excel also.
I match to voltage because is voltage that runs the motor, and with those good GP3300 I don't have runtime issues.
To avoid one cell peak early then the others is a little difficult.
But bear in mind that even with matched to runtime after several cycles the cells are no longer well matched, so you will end with some cells peaking early then others. I have no problems with this, my cells have more cycles then I could remember and they continue to be strong.
It's easy with my GFX
. With other chargers/dischargers I don't know.But never mind, what you will need is a digital watch and a multimeter. Just take the voltage readings every 3 (if you are enough fast) or 5 sec. Wrote them down and build a excel graphic (time vs voltage or inverse if you want).
The AUC is build on excel also.
I match to voltage because is voltage that runs the motor, and with those good GP3300 I don't have runtime issues.
To avoid one cell peak early then the others is a little difficult.
But bear in mind that even with matched to runtime after several cycles the cells are no longer well matched, so you will end with some cells peaking early then others. I have no problems with this, my cells have more cycles then I could remember and they continue to be strong.
12-30-2003, 08:12 AM
#12
actually, i think the average voltage labeled on most cells, refers to the average voltage over a window of time, usually down to .9v/cell on a particular discharge. not the average over multiple discharges. even though some matchers cycle the cells multiple times, i believe it is only the last round of #'s that make it to the label.
12-30-2003, 10:53 AM
#14
Tech Adept
Yeap, the average voltage is the mean voltage till 0,9V per cell, in a cycle.
It's greatly affected by the first discharge secondes, but it wan't show you how a bat stands up the whole discharge.
That's why a built a graphic and compare the AUC's.
12-30-2003, 12:25 PM
#15
Average voltage in most cases is the dischage curve down to .90 volts per cell, but thats assuming the matcher is using .90...some like to use a higher cutoff voltage, that will increase the avg voltage but will decrease runtime. You don't have to worry about that with reputable matchers, but it does happen.
Later EddieO
Later EddieO