Help wanted developing idea for a all in one driver test bed

[ChaosLord]

I would like to help anyway I can. Even if it's pitching in on cost of first batch.

 

[Cyparagon]

What's wrong with mini grabbers?

Here is the resistor I use...

That's pretty pricey and dissipates a lot of heat, which contributes to innacuracy. I don't see any kelvin sense option either. You may want to consider a lower value or just use an ammeter instead.

 

[djQUAN]

What is for somebody to loan me one of all of the major players in drivers so I can map the pinholes out. Hint hint for anybody watching this..... I've got a few currently but probably need about five or six others as well

You can simply ask for the PCB cad files of the drivers and get the measurement off of those.

What's wrong with mini grabbers?

I don't think they can handle much current. In my experience, under 1A is fine, a few amps and they melt.

 

[Cyparagon]

What minigrabbers are you using? I put 10A through them sometimes. Only for 10 seconds or so, but they come out fine. I know they're not designed for high current, but neither are pogo pins (or the 26awg everyone is so fond of using around here). You can still get away with either.

 

[djQUAN]

I just use what is available so I probably bought the cheap chinese ones. But I would still trust the pogo pins before mini grabbers for a few amps of current. But that's just me.

 

[The Lightning Stalker]

This is a great idea and is how lazeerer tests the X-Drive™. Actually, he "impales" them on 0.1" gold
plated pin headers. (Don't remember where I saw this) It would make it much easier if you had to build a
whole crap load of lasers.

 

[Jmillerdoc]

That's pretty pricey and dissipates a lot of heat, which contributes to innacuracy. I don't see any kelvin sense option either. You may want to consider a lower value or just use an ammeter instead.

I'm using a 1R value, pretty much what most people use as I understand. Were you referring to the pot pictured? That ad I took a screenshot of isn't a very good picture, I had to crop it down to upload onto the forum. After doing so I realized I cropped out the picture of the resistor and cropped in the pot. What I'm referring to is the Visheay Nude Foil resistor, not the pot. Still pricey though, agreed. But less so than a micro-ohmmeter I'd have to buy to test a bunch of 1R resistors to get a perfect, or nearly so 1R.

Additionally, you mention getting an ammeter. I have a cheap one. I could be wrong here but as I understand most ammeters do essentially the same thing we do here with a resistor, it just makes the conversion for you (it has a resistor in series to the current and measures the voltage drop across a set value resistor and converts it into amps or miliamps, etc). What I also understand is most cheap ones in the sub $100 range are limited by the same issue I am attempting to overcome by using am ultra precision resistor, they tend to be inaccurate by up to 5-10%. I once asked this very same question somewhere "why not use an ammeter?" And the responses were varied. In the end the consensus was that accurate ammeters were much more expensive than using a tight tolerance resistor. I'm always open to a better solution. So far the explanations given made sens, so I opted to spend $12 on a precision resistor over a $300 ammeter.

Here's a picture of a cutaway view of the resistor below.

Hope that help clarify.
Jmillerdoc

 

[Jmillerdoc]

@djQuan- that's exactly why I asked for help, there's smarter people here than me. You are clearly one of them, excellent idea on getting the PCB files. Which brings up the next order of business, which drivers to include?

Maybe I should just start with polling those of you watching this thread. Feel free to make some suggestions.

@Adam- I like the idea of the round driver slot. What do you think about soldering two springs to the test board, a small inner one and a larger outer one. Then the driver could be pressed down to the springs for contact. Seems a little sloppy though, and it may be difficult to find the right springs for this easily. Try to think of a good way to make contact between the test board and round driver. An alternate idea would be to use a single contact point for the center of the board and a small wire with alligator clip to attach to the outer ring. Some of my round driver have holes like the square drivers but the are often tiny little holes. As Poo would say....tink, tink, tink....

 

[adambowker]

I realized while I was working on it this morning that the round driver slots I originally had wouldn't work. I had the input rings on the bottom connected to the output pads, rather than the holes on the driver that are the output.

As for connecting the round input pads to the board, I think having the copper rings on the board should be good enough for contact. I can make the inner pad on the board larger too if need be.

The circuit below will work for the rectangle drivers but we need a way to connect the round driver outputs to the round driver output pads (above them). Maybe we could attach alligator clips to those pads that you can use to connect to the driver output wires, but then you'd have to have wires soldered in the driver output holes already. Maybe that will work? Who knows... what does everyone else think?
(Don't mind the random "J6" and "J4" and "VCC1" and whatever numbers on the sketch... they pop up by themselves but they won't be on the final product)

 

[djQUAN]

I think the best for now is to include most drivers that are easily available to us. Then to narrow it further, include only those with pots as you would solder on those that needed links, jumpers or resistors anyway so why bother on not having to soldering on the holes Just my opinion.

Then as other drivers become more available or come out, we make now ones for those too.

 

[adambowker]

I added some spots for alligator clips beside the round driver output pads. These are little holes that will have small wires+clips on them. These clips will attach to the driver's output leads (meaning you'll have to already have wires soldered on the driver).

Again: ignore those little "J4" and "J6" and "VCC1" numbers floating around... they just pop up by themselves

Will that work?

 

[Cyparagon]

I'm using a 1R value, pretty much what most people use as I understand.

...I am attempting to overcome by using am ultra precision resistor

1R was okay back in the days of the 150mW DVD burner diodes. It's frankly outdated now. With a 3A drive, you're dropping most of your voltage in the resistor instead of the diodes. This makes the IV curve less like a laser diode and more like a resistor. "Good enough" for many people, but you're clearly after something better, and the simplest say to do that would be to switch to 0R1 or better yet, an ammeter.

1% (or 0.25% as they argue) is not ultra precise anyway. Ideally you want something with a very low temp-co, kelvin sense, and a trimmer for calibration.

[ammeters] tend to be inaccurate by up to 5-10%.

This one Claims to be 0.1%, measures voltage as well, and is half the price of a resistor. Don't you think it's worth checking out yourself for the potential cost savings and simplicity?

Even if it's off by a bit, you can easily adjust it by adding or removing material from the shunt.

 

[Jmillerdoc]

@ Cyparagon- that's what I love about this forum. All the smart people seem to find me and offer good advice. Certainly think it's worth checking out. Even to see how it compares to my "ultra-precision" resistor test rig. Another point I'm surprised you didn't take issue with is the power rating of the resistor. 0.6w. Well, you did inderectly I guess. But yes, the resistor is way under rated for this application. At least on paper. My thoughts on this and why I haven't fried it yet is they rate them based on much higher voltages (my guess) and the rating isn't entirely on whether or not it will fail or fry. So far I've been able to run 2.5a through it for the 5-10 seconds it takes to get my reading and I haven't fried it. Although I don't have the ability to measure accurately the resistance I would not be surprised if it has drifted away from 1R. Nevertheless, every time I buy a preset driver from DTR or Lazeerer I get an almost exact output reading as they said they set it to.

So, in the end it's obviously not the most perfected solution, it was just better than what I used to start with and I have become fond of my boutique $12 resistor test load. Placebo effect of sorts I suppose.

Your points are well received, I will have to order one of those ammeters and see how it compares to the other methods I've used.

Lastly, on the ammeters that can handle this sort of current, does the same issues we're discussing not hold true with how they are built? Meaning, I presume they use a drop resistor to measure the voltage across and just run the value through a simple calculation to give the current. If they are better than what I am using then do you suppose the drop resistor inside is a value like you suggest? 0R1 or less?

I will order one of these and someday post back here some comparison results.

Thanks again for your input, truly appreciate the angle and insights.

 

[Cyparagon]

If they are better than what I am using then do you suppose the drop resistor inside is a value like you suggest? 0R1 or less?

Not even that high. The 10A range shunt is typically 10mΩ or less as far as I know. It's basically just a length of wire.

 

[djQUAN]

You can probably get 10mohm or even 1mohm shunts from dead laptop motherboards. I got a couple from those.

Another option I did when I needed a high current shunt was use a strip of stainless steel. The resistance tempco is much less than copper (not perfect but won't be much of a problem unless it gets burning hot).

 

[IWIRE]

This is a really cool idea ! Would reduce time to set up a drive and reduce work bench clutter. I love it. When you brains get this all figured out, make a few extra so us dummies like me can get one Seriously, that would be handy as a pocket on a shirt ! Great start to a great idea ! :beer: