Some 3 Phase Alternator Findings, 82 1100 A2

Well I thought that since DC motors are all over a modern whatever that it would be good enough to load up the system for a output test. The load tester I have now is more for the Automotive systems, and really has a rocker on/ off load switch and will be much more than 20AMPs. So I want something that will load the additional 10AMPs or so to get a total with the stock system load of say 8AMPs, then with the meter in series get a current fow reading from the regulator to the battery in AMPs. Hopefully getting close to the FSM rating for that bike being tested. All of this with simple low cost stuff. I'm trying to keep my Fluke, and AMP clamp out of the test procedure as it is expensive for me let alone every KZ rider out there.
If you got a load Idea I will be grateful and try it out. But at the moment I do not have an O-scope to follow wave forms. Sorry. Wish I did, as I would document that as well.

How about a string of ten 20-watt bulbs in parallel, each with it's own light switch. And maybe one 60-watt bulb on all the time.
Then you just flip the switches on as necessary.
Bulbs and light switches are cheap, but the thing will be huge.

Or if you check ebay, you might find a source for ten 10-ohm, 50-watt resistors (sometimes they come to $1 a piece). Put them in parallel on a big aluminum plate (like the gold resistors on my tester), each with its own switch.

This method is more incremental, but that just means you will have to select a few RPM points and connect them together on a graph. It may take a helper to flip the switches fast enough.

The resistor method will be safer in terms of giving you more time to react so as to not burn things out. With the bulbs, if the voltage gets away from you, when one bulb burns out, there will be a chain reaction where they will all burn out. :woohoo:

Luckily here we have a great electronics store. They have nearly everything for parts. So I could get some resistors cheap, and no shipping. They would have some larger scale pots as well. But I was still trying to come up with something usable for a test that not only is reasonably priced for a do it your-self test that everyone hopefully can follow.
Your right about the bulbs, you need a steady load and like you said if 1 goes, then most likely the others, and some damage to system components could occur.
So a constant safe load, adjustable, to follow the rise in output by the system.

Do you think for the load bank, that an addtional fan for air cooling would be needed? Rotron used to make a # CL2T2 120 volt fan that moved 500 CFM.

This would be to keep the heat in the load tester stable? So as not to interfere with the readings, right? And to keep say a resistor from popping?

I usually try to err on the side of safety when dealing with heated resistors. I also wonder if the wires will be up to the current load when testing.

Some wires have 80 C insulation, I prefer 105 C or even Teflon(if it can be obtained,probably $$$).

Looking at Lou's loadbank, maybe some computer "Muffin Fans" on a removable panel could be used to suction the heated air out of the enclosure instead of blowing it out.

Well the wire I think will be the easiest part, the stock stuff has held up along time. But you have a good point if using resistors. I will try to make up a cheap/ "China Tools" (ie. Harbor Freight) type of simple to use and hook up type of test. 1,volt meter, 1, AMP meter, or if they choose one of their AMP clamp models,and a running KZ. With total current output being the target, for a sustained amount of time, as with this there is no spec I can find.
I don't want to hurt any of my electrical doing this, like an expensive stator. But if it helps someone else who is having issues, and looks to KZR, I hope to have something that they can test with for a small amount of money. This mostly because it seems like no one wants to spend with the current economy, and I like the Idea of it.

Yeah, it's definitely a short-duty item. Anything longer than a minute at high RPMs and the resistors will get HOT.

With four 50 watt resistor, I'm right up close to the limit at max power from the stator. But then, I wouldn't expect to test it at max power for long since the engine will probably be overheating too.

I imagine a real test, I'd ride the bike around to get it up to temp then hook up the load for a short time, but measure while everything was still hot.

So are you doing a test of 1 of your systems? Single or 3 phase or both? If you are we can compare both output figures and test setups.
I got MFOLKS to suggest a heater grid, as he built these along with stabilizing capacitors, I believe, on a large scale for industrial load testing. I'm going to look into this, as China Tools has some 12v heaters cheap, but I want to be able to build the load up with a adjustment, rheostat, or something, hopefully part of the "Unit" chosen, Cheap and available.

The problem with a steady rheostat is that the rheostat has to handle a huge amount of power when it's resistance equals the load's. If you can get a 400 watt rheostat, it may work ok. It still has the same problem of having to control it continuously to maintain the voltage.

I like the heater grid idea, but a heater's resistance changes a lot as it heats up, so it may need a lot of adjusting. Bulbs heat up and stabilize pretty rapidly, but I think heater elements take more time, and thus will be slower to react to adjustments, resulting in over-correcting.

I must have gone through dozens of ideas to try to keep it super simple, but I couldn't come up with anything useable, hence the tester I ended up making is pretty involved.

Also, any load device you make needs to have a capacitor to smooth out the power. If not, then the meter's voltage reading will not be as meaningful. What i mean by that is when you use a meter to read DC voltage, it is measuring the average voltage. What we really want is the RMS voltage. Normally, as long as the DC is realtively smooth, the RMS voltage is basically the same as RMS voltage. But if the DC is heavily pulsed, the two values deviate quite a bit.

I haven't really worked with these, but wouldn't it be possible to just make a rectifier, attach a big capcitor, then hook the output of the rectifier to this thing:
www.harborfreight.com/500-amp-carbon-pil...ad-tester-91129.html

Rectifiers are simple enough to make. Here's a simple one using silicon rectifiers. But selenium ones wouldn't be a lot harder to make, but maybe slightly harder to get (they are preferred because they waste less power than silicon diodes).

DIY rectifier