Auctionstealer.com it's free and works well, its a set and forget... 10 seconds left and it bids your predetermined price. i have used it many times
good luck

IIRC, CC uses motors (not sure what kind or how big, how many) in a bath of oil (big drum actually) to load test their ESCs. Might be worth picking their brains to see if they can help setup a similar test bed. I'm sure it wouldn't be difficult or expensive to find some suitable large motors (car starter motors, etc)

Just an idea

hmm.. that's an interesting idea. i was thinking more, and really being able to setup any load on a motor should work. Are they using something like a propeller, or just a really large motor? I would think you would need a prop or something to create a load. Could get really messy tho :)

I was thinking about a prop for my above setup, but not sure how much it would mirror a car setup. Perhaps it doesn't even matter. The static tests are easier to do, I just don't know how predictive they are in dynamic apps, and its hard to create the really high loads (ie 60C) it would be nice to test. Maybe it doesn't matter... something to investigate.

It would make sense for them to test their ESCs on a real motor load to make sure it can handle highly inductive loads, but in this scenario a resistive load would be sufficient IMO.

ANY power MOSFET can do this, as long as you can take their heat out. You don't have to take care of the FET linearity, the OpAmp will do. If you need I can draw a basic schematic for yoru ref.

Someone might want to go down the road of "supply batteries" while stuff is being gathered:

I had originally thought that perhaps users might send me packs in the off season. I've bought a number of used lipos, but you never know what you're going to get so any comparisons are skewed. The shipping twice also runs the cost up to high.

A fair way might be to have the vendor/assembler send you a pack and if cycles or other tricks/tips make the performace better then so be it. Have them send the pack at whichever point they think is best knowing you may do a half dozen tests or so.

This gets rid of any discrepency for cycle life and cycle currents. I've found that a brand new pack always performs best and continues to degrade at every cycle. If they believe that 20 cycles is the best point then so be it, they send you one with 18 cycles on it.

Just trying to think ahead.

Well, despite all the less-than-ideal factors in my design, I went ahead and got 220 50w 3ohm resistors. Retail price would have been almost $400 + shipping, and I got them for ~$80 shipped. More than I wanted to spend really, but I could always use extras for various power projects.

Once they come in, I'll size up some cheap PC surplus heatsinks from BGMicro or AllElectronics to cool them. To test 2s up to 150A, I'll be using 60 resistors in parallel :oh:. Which is nice because I'll make banks of 5-6 on each CPU heatsink and just add them in the circuit depending on the max current I want to test at (and use the ESC to go from 0A up to the max value). I'll probably use some 8-10GA car audio wire for the cabling.

So, the process moves on... slowly but surely.

Great to hear an update! Keep us informed and remember what a thread is useless without... :whip:

OK, I'll bite; what are the pitchers for?

lol, ok, I get it. "Pitchers" = "Pichers" = "Pictures". It was late last night. That's my excuse anyway. :oops:

or maybe you had one too many pictures of beer last night?

lol, I rarely drink, so that's not it.

Cool Brian, I'm looking forward to seeing how it turns out. :)

A few resistors anyone? The CRT.5 in the background is for size reference. And don't mind the duct-taped body - I never said I was a good driver. :oops:

http://scriptasylum.com/forumpics/resistors_3r0.jpg

So, stage 1 complete (getting the resistors). Now for the expensive stuff (Eagletree, MMM ESC, etc)...

Keep us updated! That's awesome!

Funny how minds work alike. Here is Castle's version:

http://i37.photobucket.com/albums/e9...t/IMG_0289.jpg

We are using the new datalogging ESC (Phoenix-ICE 200) to store data from the battery discharge. 9 x 400 watt resistors, 18 x 33,000 uF capacitors and 6 x common mode chokes to keep the battery load as DC-like as possible.

Very nice PD :)

lol. Instead of 9x400w resistors, I'll be using 60x50w resistors - pretty close power-dissipation-wise. Care to donate one of those Phoenix ICE ESCs? :wink: I picked up an Eagletree 150A logger from a member here so that is taken care of, but the ICE would work just as well.

Looks like I'm gonna have to get some caps as well. I assume by the size of yours and the value of yours that they aren't low-ESR? I suppose it's not needed with that amount of capacitance. Maybe a 0.5-1F car-audio cap would work (and probably cheaper). I imagine the hook-up spark is VERY substantial though - or are you using a no-spark resisitor?

My design didn't include chokes - maybe I should look into that as well. I could probably pick up a large toroid (or unwind a spare large transformer I have laying around) and wind one myself. What is the total inductance value of your setup?

What are you using for the throttle control signal; an actual radio, a seperate circuit (like I planned), or using some computer driven control signal? I would LOVE to have a computer-driven control signal so that I can program test runs and have it be consistent for all battery tests. So, if your setup uses a computer-driven setup, PLEASE let me know! :yes:

Thanks for your input Patrick. At least I know I'm on the right track; if it's good enough for CC, then it's good enough for me! :smile:

Also looks like setup for high powered brushed mode based on wiring

Hey Brian,
To be honest I don't understand everything I've just read but I do savy enough to comprehend that you may need test subjects ?
For whatever it's worth, I have several T-RC Packs. They are 2s&3s and all 15C with various mah capacitys. All are like new and willing to sacrifice themselves for the greater good of research .
LMK.

I appreciate that Bondo! I just want to warn everyone that may be interested in donating test subjects that there is a definite possibility that these tests will destroy the packs under test. At the very least, the packs will have a shorter lifespan.

Even so, it's not nearly done yet as I still need the following:

- MMM ESC (or any ESC that is capable of brushed mode and at least 200A).

- 500,000-1,000,000uF capacitors. Either a bunch of smaller ones paralleled, or one big car audio cap.

- Some type of heatsink. Since I'm using so many resistors, maybe just a 1.8" thick piece of aluminum with a couple fans blowing across them.

- A container and ventilation system if/when the lipos under test decide to vent/flame.

- Assembly involving some pretty heft wire (I have some 8GA and 10GA car audio wire) and a bunch of solder.

Hi Brian,

The capacitors are low ESR type, Panasonic TS-UP series, 33000uf, 35V. .014 ohms / each at 20khz.

We just threw the chokes in there when it seemed we were still getting a lot of voltage ripple from the switching on the ESC. Turns out that the voltage ripple was from ground bounce -- when we measured it differentially, the ripple went away. But, we left the chokes in anyway. :)


We are using a Castle Link to drive the controller and to log data -- the ICE can only store about 10 minutes of data at full rate. Richard and Jon cobbled up an application to drive the controller and read data while it is running.

Thanks for that info.

For the sake of cost, do you think I'll "need" low-ESR cap(s) if it is ~1F? I already have a 1F cap sitting around, but I'll get some low-ESR ones if I have to.

Unfortunately, I don't have the expertise to "cobble" programs of that nature together, so I guess I'll be stuck with the manual method. The best I could probably do is get it to generate the proper signals for the tests, but that's about it. Apps like that aren't my forte. :smile:

I know you just spent all that money on resistors but did you ever consider a motor mount, motor, and a big prop off a plane?

I did, but figured resistors were cheaper (and safer for me and the dog or cat :smile:). I have some other projects that I can use the extra resistors for, so it's not a total loss even if this doesn't pan out.

But, if anyone needs some 3ohm 50w resistors, I have about 100 I could spare. :smile:

Update:

I was going to build the throttle control circuit myself, but found the Esky EK2-0939‏ servo tester that is probably more accurate and less prone to "drifting" with temperature, age, etc, so I ordered it. Coming from China, so it'll be a week or so before it gets here. In the meantime, I noticed it needed a power supply from 6v-9v, and since the ESC might not provide a stable 6v output on the BEC at the high battery loads I'll be testing, I made a regulated 8v power supply. Nothing special, just a small transformer, a couple caps, and a LM7808 regulator. I happened to have all the parts handy so it was effectively free. :smile:

I also picked up an Eagletree V3 datalogger from another member here. It came with one USB cable, but I needed another cable, so I made one from one of a billion USB cables I have and the speaker connector from a computer motherboard:

http://scriptasylum.com/forumpics/ba...data_cable.jpg

Then, since I needed a temperature probe, but didn't want to pay for one, so I scavanged a thermistor from my pile of computer parts and attached a servo connector on the other end. Works well, is accurate, and has a quick response time. I want to bond a small piece of aluminum to it so I can get a better reading and to reduce any stress on the delicate thermistor:

http://scriptasylum.com/forumpics/ba...temp_probe.jpg

So, progress is slow, but coming along. :smile:

nice - do you wear sandles

I'm still really excited about this. Make sure to test the a123 batts! People get pissed when I say lipos are better then a123s due to voltage drop, but I don't have much to back them up.... Can't wait to see the final product Big G!

Thanks! I'm anxious to get it done too, but finances don't allow it (if they did, I'd just get a real setup).

Huh? Yeah, I have some Tevas, but I don't get it? :neutral:

HOLY COW....Impressive, and here I even thougt of oing something like this! The whole money thing is hold back though. Except I was gonna program a Microcontroller to do most the stuff, voltage current temps, plus the whole rpm idea.


Figure since I claimed I might try something like this I felt that I should post my idea just incase it may spur some ideas for others.
http://img.photobucket.com/albums/v2...enchdesign.jpg

With design it records the main pack voltage with no load, individual cell voltages, pack temp at start, and esc temp at start of test. While testing it would track the main packs total v drop and cell v drop in coralation to different applications (sudden throttle/brake, "cruise" etc) as well as peak and average current draw over a given period of time. All this info will be controlled by a PIC (hopefully most the techy heads know what they are) as well as throttle control will be handled by the pic considering, pics have a servo function with variable pulses and so on, this all will sent via usb to a pc to be debugged on the screen, which I believe theres also a command to "print" the debug values which may be similar to copy paste in which case the values would by plugged into an excel sheet then plotted on a graph. As for the voltage to signal and amp to signal it would basicly by some form of an analog to digital converter that turns voltages/amps to digital pulses sent to the pic and recorded.

A little off-topic, but this sort of drives home the need for a standardized testing and rating.

I was taking some ET graphs of a couple of vehicles and the test results are quite revealing.

First up is my CRT.5. Castle ESC, Medusa 36-50-3300, 46/14 running on an Elite 3s 30C 3.3Ah battery. Here is a screenshot of the run:

http://scriptasylum.com/forumpics/el...300mah_30c.jpg

Next up is my Slash. HobbyWing 80A ESC, Medusa 36-50-2200, 58/11 running on a Maxamps 4s1p 5Ah 20C battery. Here is the screenshot of that run:

http://scriptasylum.com/forumpics/ma...000mah_20c.jpg

All I gotta say is WOW! The Elite held 3.4v/cell at a 149A burst!!! And the average voltage is pretty darn good too! OK, it IS rated for 30C after all, so let's look at the MA pack. The highest burst is 61A, (which is 12.2C by the way) and the voltage fell to 3.12v/cell.

Of course this "test" is not repeatable, nor does it show anything conclusive, but there is a definite discrepancy here.

Brian I have a 1F cap and some heatsink material I can donate to the cause. The second pic has 50W resistors on it so it should be perfect size for your setup. Also if you need I can make a fan shroud for it as well.

Jeff

battery tests would be awesome.is there anyway to use a fluke meter?not sure how many amps they can measure.what about using automotive battery testers for the amp draw?some good ones use carbon piles.our high school had one that was adjustable.just some thoughts,i do appreciate the fact that this all is being done on a slim budget.

also,since electrical minded folk might read this,how hard would it be to build a non adjustable output driver for a brushless motor?like to power a motor on a dyno,or a starter box?

Flukes are only rated for 10A for most of them and 20 on a few models. You could use a shunt and measure higher loads, but then you have to convert milivolts to amps via a chart from the specs of the shunt

Using the Datalogger he is using will give recordable shots of exactly what's happening.

To build an non-adjustable drive for a brushless motor.

Simple version:
Use a servo tester an ESC, motor, power supply/battery. Connect them together and set the motor RPM you want. Remove the tester and measure (in Ohms) the pot for adjusting the "servo" between the middle terminal and one of the outer terminals, then the same for the middle and other terminal.

Write down those numbers and find resistors that match those 2 numbers. Remove the pot. There will be 3 holes where the leads of the pot were connected. Solder the 2 resistors into the 3 holes left by the pot and use the numbers written down earlier to determine which resistor goes on which side. You will have a lead from each resistor in the middle hole and the other lead goes into the outer hole. You will also need to have a Dual Pole Dual Toggle (DBDT) switch to be able to switch in 2 resistors of the same value half the resistance of the pot. (if the pot is a 5K you will need 2 2.5K resistors.) That way you have a neutral position and a place so the ESC can arm.

More complicated version should be handled in another thread:

Jeff

for some reason, i thought my buddies fluke measured higher?oh well.
i was looking to replace the esc with a simple max rpm driver for a brushless motor.something simple and cheap.possible other uses might be for a starter box or a dyno,or better yet,to use a brushless motor to run a comm cutter!
HA! i digress though ,it should be in another thread...

Yeah, my Fluke is fused for 10A. Gotta figure these meters are for general purpose measuring, and not for currents in the 100's of amps. You can use clamp style meters, but that's just more cost. And using any meter would require the tester to constantly jot down voltage a current measurements (so would need two meters), and that would not be close to accurate. The Eagletree measures voltage and current using a hall-effect sensor (does not drop any voltage that a shunt would) to measure up to 150A, and can record as fast as 10 samples per second (100ms). For ~$70, the ET device is hard to beat for what it does.

And as far as a non-adjustable BL motor driver, I can think of a simpler way: Use an ESC like J57 said, but use a failsafe device as the throttle driver. They generally have an adjustment pot, but it's meant to be set once.

Wow, I must have missed this post somehow! Didn't mean to ignore you. :oops:

Anyway; yeah, that would be awesome! Those heatsinks look like the PERFECT size. I intended to make the resistors lined up like you have in the pic, just a 46 more. :smile:

I got six 80mm fans for this project ($2 each at allelectronics.com), but was thinking I might put the fins side down, build a little tunnel, and put two fans (one at each end of the sink) and cool that way. This would allow an easy way to replace resistors if I blow some. Plus, it would look nice. :smile:

I also finished the power supply for the servo tester. I got a 12.6v/1A transformer from the same place. The supply actually has two outputs; one regulated for 8vDC for the servo tester (needs 6v-9v), and the other output is unregulated and puts out around 14vDC at ~750mA for the fans; perfect!

I was thinking a little though; To make this test repeatable, I'm gonna have some problems:

1) Constant current. The current through the resistors depends on voltage. I can program the ESC to output X amount of voltage, but as the battery depletes, the voltage will fall, and so will the current. So, testing a certain discharge rate (constant current) will be tough unless I constantly tweak the servo tester accordingly.

2) Pulse discharge profile. The servo tester I got has an automatic function which will sweep from min throttle to max throttle repeatedly. I can vary the speed of the sweep from slow to fast, but the profile is not "peaky" like we see in our cars. It starts at min throttle, ramps linearly up to max, sits there for a fraction of a second, ramps down linearly to the min, sits there for a fraction of a second, and the process repeats.

I wonder if I could see if Patrick will sell me one of those new ICE controllers he is using along with the custom software he is using? :whistle:

That's exactly how I have done my heatsinks and it works great I just used 1 at the end that was a 4 5/8" square fan thermister controlled (ended up just shorting it out because it just wouldn't run fast enough for the application). Mainly a fan at 1 end because it's easier to bend. I have a ACAD pic I drew up but I can't remember how to capture a screen shot(edit: nevermind). I have it setup with 2 fans on 1 end.

As far as constant current you must add in the more parrallel resistance in order to keep the current constant. You could use a window comparator to turn on a fet or relay to switch in more resistors. I have a line on some .001 or .01 ohm resistors that are like 100W or 75W same style but bigger.

Jeff