FEELER: Continuously Adjustable Test Load

Hey everyone. I love doing circuit design, and I think I may have stumbled upon something great here - a continuously adjustable voltage drop test load.

This test load should be capable of driving up to 3A or more, depending on the set resistor used (I figure 3A would be a reasonable upper limit). Then, you can adjust the voltage drop (measured however, I could probably just add a measurement point for voltage drop) with a simple, 10 turn potentiometer or what have you.

The only problem here is that, when measuring low-current output drivers, the test load itself will be consuming a significant amount of power. I'm thinking a test load that one simply plugs into the wall, complete with voltage drop and current output read out on an LCD display.

I will probably make one of these for myself (it seems like a useful tool), but I wanted to see if anyone else would be interested.

EDIT:

So it won't be necessary to have an external power source of any sort - I figured out a way to make it all work a lot more simply! This should actually be revolutionary when it comes to test loads, because you will be able to precisely choose your voltage output. Also, it shouldn't be too expensive for a basic version Going to place an order for the PCB now, will have updates in a few weeks.

You've got my attention, mostly because I'm intrigued. I'm curious as to the "continuous" V drop method used. Adjustable zener? A transistor that control collector to emitter Vf by setting some base current?

Well, I have a few ideas. Working on another one now.

EDIT: So it won't be necessary to have an external power source of any sort - I figured out a way to make it all work a lot more simply! This should actually be revolutionary when it comes to test loads, because you will be able to precisely choose your voltage output. Also, it shouldn't be too expensive for a basic version Going to place an order for the PCB now, will have updates in a few weeks.

Wolfman29 said:

Well, I have a few ideas. Working on another one now.

EDIT: So it won't be necessary to have an external power source of any sort - I figured out a way to make it all work a lot more simply! This should actually be revolutionary when it comes to test loads, because you will be able to precisely choose your voltage output. Also, it shouldn't be too expensive for a basic version Going to place an order for the PCB now, will have updates in a few weeks.

Click to expand...

Why would you order PCBs before you have a proven design...:thinking:
Seems like a lot of expense before knowing the final design...


Jerry

You can contact us at any time on our Website: J.BAUER Electronics

I order a small amount of PCBs for my prototypes, because I don't have a method for testing otherwise. However, I am nearly 100% certain this design will work because several of my simulation programs are giving me the same results, which is the result I want, for this test load.

I'm not trying to rain on this parade, but have you considered:

• Test loads are usually designed to mimic the voltage drop characteristics of the actual load. Using a potentiometer to change the resistance, at a given current, will affect the voltage drop of the load unless it is compensated in some other way.
• If you don't care about this voltage change, or will compensate with diodes or something else, what is the point of the potentiometer? Why not just use the lowest-possible measurable resistor you can find instead and use selectable diodes?
• Furthermore, at a given voltage drop, and a set constant current (say you compensate), you're still dissipating the same amount of power no matter what resistance you set. What is being accomplished with this device?
• Potentiometers are not very adjustable at low resistances. Those that are are usually wire-wound, expensive, and highly inductive, which can affect a driver's operation. Also, utilizing a potentiometer in parallel with another resistor results in non-linear resistance with a very small usable range.
• A transistor can be used as a variable resistance; however, it can "artificially" limit the resistance as well, requiring extra measurement and tuning to ensure that the driver is actually doing the constant current limiting. If the transistor prevents the correct operation of the constant current output, the driver may be damaged in the process.
• Often test loads are rarely utilized for longer than a few seconds, because it is just used to set the current once, and quickly. For those operations that require longer durations such as power plots, power usage is rarely an issue anyway. Where does this device fit in?

On another note, if you're going to make some sort of advanced test load, I have a different idea that might be useful:

Create a test load current meter with an LCD. That way someone can just hook it up, power it, and not need to use any multimeter or whatever else. You can even use those voltage-sensing LCDs to avoid some sort of micro-controller.

Sounds like a voltage sink to me:



I've thought about it before, but an almost perfectly flat slope like the one given by the circuit above doesn't mimic laser diodes well. Rounding to the nearest 0.7V is good enough for me, and it offers the best approximation of an actual laser diode.

Similar ideas to the ones above.

However, Bionic, you're not seeing it right -I am not directly adjusting resistance. Rather, I am directly adjusting the voltage drop. Further, I'm a step ahead of you - already working in the design of the LCD to this test load. If I am going to make an advanced test load, I may as well pull out all of the stops, eh?

I am currently working on adding an LCD panel meter to it to read off voltage of test load (almost constant across various currents) and current through the test load.

Here's a thought, how about a voltage sink in series with a rheostat so you can select the "threshold voltage" and the IV slope as well? Still won't have the response time of a diode, but it's closer than just a voltage sink.

Hmm, this'll be an interesting project. I'll have to see how it turns out.

Hmm. That's an interesting idea, Cyp. But who uses a test load for the response time of the diode, is my question. The response time of diodes are so short that you would need an oscilloscope to see it differ from a rheostat.

Well, it can also potentially oscillate with the driver. I'm one of the few people out there that use test loads to actually perform tests instead of setting the drive current:can:. For example, if I wanted to see the startup performance of a driver, an active system like the one you propose will have no mustard to cut.

That may very well be the case. Once I get my prototype in, I will start to examine how it compares to a real laser diode.

I agree with Cyp's concerns. Response time is one thing, risk of active test load oscillation in MHz range is another thing that can lead to trouble. Most people wont even notice it without a scope...Feedback compensation may dampen oscillations, but response time may also be affected, including phase margin under various ''test load'' conditions. I am not familiar with the topologies of the boost CC drivers popular on here, but some of the designs may suffer damage if they see an open load condition during the time it takes for the test load to react due to overvoltage stress of the output switches...

Seems like a legitimate concern. I'll get back to you guys on that when I find out. I have an oscilloscope I can test it with.