It would be nice to have 6’’ battery cable extensions to make battery (and microcontroller) placement easier.
edit:too tired to notice horrible spelling of topic
Is this for non competition use? If so, you could easily extend the port from the microcontroller.
Yes it would - This has been discussed in these forums before
Just buy some new leeds or strip the wires and make them longer. It’s that simple.
I believe this post is in reference to “competition legal” battery extensions. I believe that 721tba is quite capable of building an extension cable or modifying an existing lead, but would like to do it on a competition robot.
Game rule 15 is very clear about the fact that electrical parts are not to be modified.
But having a “game legal” battery extension cable would be quite helpful to many teams. It would allow for enhanced design possibilities, make it easier to change batteries, and reduce stress on the battery cables, which… in many cases are turned at fairly sharp angles in order to plug in to the controller.
This is a good suggestion… I’d like to see a “game legal” battery extension, too.
Up late last night, were you? Four awards at one tournament can do that, you know…
Additional ideas to make it more worthwhile:
- make it longer, rather than shorter; coil up the extra rather than needing two in series.
- make it a Y cable (or daisy chain parallel), so both a power expander and a microcontroller can be powered from the same battery. You may need to test it out first, to see which batteries can support full power on both at once.
You probably don’t want to make it much longer than necessary, since you will get some power loss. Remember the battery cable carries the full current draw of the entire robot. To reduce I²R losses, you either want the cable to be short, or heavy gauge (or both).
The point of the power expander is to allow you to use a 2nd battery. What advantage would you gain from using a power expander on the same battery as the CPU?
Cheers,
- Dean
the only reason i can think of for using one battery to power both the power expander as well as the micro-controller is to distribute the motor load more. so instead of 10 motors on 2 breakers, you would have 10 on 3, allowing for a better distribution and a lower over all mass of the robot
I remember that. Maybe thats what gave me the idea.
Do you have any interesting data about how well it works with 1 battery and two Cortex with lots of motors and high loading?
I suspect there would be differences between
the old NiCD 2AH vs the new NiMH 3AH and the new small NiMH battery.
I have the parts to make a series/parallel DVM link, so I can get a couple cheap Harbor Freight DVMs (<$3 ea) to monitor Vex battery voltage (DVM in parallel) and current (DVM in series). DVM typically only go up to 10A in series current mode though.
Regarding Deans comment on length vs voltage drop, while technically true, I wouldn’t worry about it. I found this handy link http://www.stealth316.com/2-wire-resistance.htm which shows 16 gauge wire has 4 ohms resistance per 1000 feet, and so 10A current through 1 foot would have 0.04V drop, while 10A current through 2 feet has 0.08V drop. 18 gauge wire is twice that. I don’t think the difference between 1 foot and 2foot would be measureable compared to the other sources of variation.
Yep - I threw that advice out there without crunching the numbers.
I agree that an extra foot or so would make a negligible difference.
Cheers,
- Dean
I don’t have any Cortex’s… yet… But, if you have 8 motors and Two Controller, you will pull more power than 8 motors and One Controller, until you go above the 4 Amp point of One Controller…
The differences are the Discharge Curves between NiCad and NiMH, when comparing the same size battery. Size being the mAh, not the *physical *dimensions… The new small NiMH battery is still 2000mAh, same as the old NiCd one.
More mAh, means potentially more power in a shorter time, or the same power for a longer time.
NiCads Batteries are known for being able to dump large amounts of power in a short time because of Low Internal Resistance, verses Alkaline Battries. That is why using Alkaline Batteries in your Vex Robot is a not recommended… I would assume that NiMH battries have a similar Low Internal Resistance.
According to the Wikipedia Article, Nickel-metal hydride battery, “A NiMH battery can have two to three times the capacity of an equivalent size nickel-cadmium battery.”
Check these out–> amploc Hall Effect current sensors.
I got a bunch of the 10A Zip Series, which they seem to have dropped in favor of the 25A model for the same price. These were for the ROV project I was working on last year…
These Hall Effect Current Sensors create a Voltage Divider, so you would just plug them into the Analog Input of a Vex or any other Microcontroller…
And Dean is ashamed for his assumption… 
Or at least he apologized… 
I still really appreciate Dean, even if he makes a mistake or three…
Per the wiki articles, standard discharge is C (ie 2A for a 2AH battery, 3A for a 3AH battery). Since many Vex users pull more than 2x that amount, we may be in a significantly non-linear part of the discharge curve. It would be nice to have an example discharge curve at currents that represent PIC max, Cortex Max, PE(Power expander) max,
and then PIC+PE, and Cortex+PE to see if a battery Y cable would work.
If you have a constant current load and a datalogging DVM (or vex analog port), you could document your setup and propose an experiment, and see if Vex will donate the various batteries to document the power curve, or otherwise contract you to do it.
NiCd is known to be better. Robots are more like power tools than robots they are like TV remote controls.
[INDENT]… a similarly sized NiCd battery has a slightly lower internal resistance, and thus can achieve a higher maximum discharge rate (which can be important for applications such as power tools). [/INDENT]
In my view, he did neither, nor needed to. But he set a good example with a prompt self-correction that is concise, straighforward, and no-drama.
IIRC, 2Ah, or 2000mAh, is 2000mA for 1 hour. If you pull Double that, it only lasts half (e.g. 30 Minutes) as long. The actual amount that you can pull, depends on the Internal Resistance, which varies with Age, Temperature and number of those “spikes” internally shorting things out.
Sounds like a great test… I don’t know if I am the best one to do it, but if a we published a testing standard, various people could do their own testing and contribute their results.
[INDENT]… a similarly sized NiCd battery has a slightly lower internal resistance, and thus can achieve a higher maximum discharge rate (which can be important for applications such as power tools). [/INDENT]
Luckily NiMH is almost as good, although it seems to have, half the Recharging life… Half the Physical Size is a real Plus too!!
Sorry I was being Tongue-in-Cheek with Quazar… Quazar has contributed so much, and graciously, he is a True Gentleman…
I’ll look around to see if I can scrape up a lab grade 4/8/12A constant load.
That would be good to test just the batteries in standalone mode.
4A is the current limit for the Vex0.5
8A is the current limit for the Cortex, or Vex0.5 + 4A power expander
12A is the combined current limit for the Cortex + 4A power expander
12A = 6C = 10 minutes theoretical max for any 2AH battery
12A = 4C = 15 minutes theoretical max for any 3AH battery
Due to the different physical sizes and IR chemistry differences,
I would expect significantly different curves for the old NiCd 2AH sub-C cell size large battery vs the new NiMH 2AH AA cell sized small battery.
The idea of a 2:1 battery fanout cable is practical if the new small battery can last at least 5 minutes at 6C, which is 50% of theoretical max.
Alternately, I’m brainstorming some system setup that other teams could replicate. Here is what I have so far, feel free to chip in ideas:
- User-built 3x battery fanout cable, so one battery can source 3 CPUs or power expanders.
- Each leg of fanout has a ZAP25 or some other current sensor.
- Each CPU has 5 std motors for a load.
- ?Extra PIC or other datalogging DVM to track battery voltage.
- Program each cpu to monitor the current in its own supply wire, and increase or decrease motor drive parameters to keep the current stable at 3.8A
What is a good repeatable way to provide a variable load on the motors?
– Have 5 motors geared together, and program 2 of them to drive the opposite way? 3 motors run at top speed, and the 2 have variable speed to change total load?
– 5 motors geared to a wheel, use a high-torque-levered servo to press a drag block on the wheel? (might be good for grinding the tread off of wheels…)
– 5 motors parallel geared up to a fanblade for air resistance, speed = power, sounds like least damaging, but maybe hard to replicate. Need a 15-25watt load for 5 motors, x 3 CPUs
– Just stall the motors (without damaging the gears?) and just control their inputs to get the right current drain? More motors = less strain on each motor internal gearset…
I was planning on running some tests this weekend on the full range of Vex batteries and capturing the results on the integrating power meter I have access to.
I don’t have access to a constant current load, but I plan to use a large adjustable power resistor to simulate 1C, 2C, and 3C loads. (I’d need something bigger to go beyond 3C, though.) I figured a simple resistance load would be good enough, since the current that motors draw falls off with voltage too.
The basic plan is to:
[LIST=1]
*]Charge each battery fully
*]Allow the battery to “rest” for about 10 min to get its internal temp down a bit
*]Attach it to the integrating power meter (measures V, A, W, mAh, and mWh)
*]Attach a large adjustable power resistor set to simulate a 1C, 2C, or 3C load
*]Allow battery to drain down to the suggested limit for the cell chemistry
[/LIST]
I’ll post the drain graphs as well as the relevant stats on the wiki. Note that I only have new NiMh batteries and old NiCds, so that will certainly effect the results a bit. Also, this full suite a tests will take quite a while, so I’ll probably have time to run each experiment only once, making the results more anecdotal than I’d like…
Cheers,
- Dean
[EDIT] This topic is continued on this thread [/EDIT]
Bumping thread to point out that the Cortex and PIC on/off switches are tiny and hard to reach, so a battery extension would be a great place to add a large on/off switch (similar to Tetrix 12v battery switch idea).
Here is a updated brainstorm of potential features for a battery extension.
-
Large on/Off switch
– for two batteries (competition power expanders)
– and 9v battery extension cables on same switch -
long enough to allow use of batteries as counter weights (switchblade robot uses battery weight to climb stairs)
-
long enough to allow on/off switch to be mounted in easy to reach location away from both batteries and from CPU
-
PowerPole edition for use with BEST batteries?
This is really needed as the new battery plug is even shorter.