10 ma to .01 ohm resolution DVM adapter

Thread Starter

#12

Joined Nov 30, 2010
18,224
I designed it, Tracecom built it, he says it works.

One (9) volt battery and a MAX6035 create a constant current of 10.00 ma.

Your meter on a 200 millivolt scale will show numbers like 199.9 which means 19.99 ohms. Good for those 2 watt emitter resistors like, .15 ohms or .27 ohms.

You can change to the 2000 millivolt range which will show 1999 and mean 199.9 ohms.

The chip needs 4.75 volts, so the high limit is about 245 ohms, showing 2.45V on a 20 volt scale. You can get to about 425 ohms (4.25 volts) if you remove the LED.

Using only one battery and 3 electronic parts makes this fairly inexpensive.

You can build the adapter or just clip your voltmeter on to the leads of the Device Under Test. Just be sure to place your meter connections closer to the body of the resistor than the current driver alligators.

 

Attachments

Wendy

Joined Mar 24, 2008
23,415
I think this is a pretty handy accessory myself, measuring small resistors is always a pain and very error prone. Nice job.

If the chip is sending 10ma, it seems almost any low current regulator should work. I think a LM117 (related to the LM317) can handle 10ma, though I might be mistaken.
 

THE_RB

Joined Feb 11, 2008
5,438
Nice idea! I've always used an adjustable DC supply and Avometer (ammeter) set to 0.1A or 1AQ etc, and measured with a digital voltmeter on the resistor legs. Having it all in one device is great.

Re your schematic, maybe I am reading it wrong but to me it looks like the sense terminals are only connected to the coax shield, and not connected to the resistor under test (which is on the coax centre conductor)? Or do all 4 aligators clip onto the resistor?
 

WBahn

Joined Mar 31, 2012
29,978
If you do actually use coax, I would recommend putting the current leads in one and the voltage sense leads in the other. They way the diagram shows it you have two large loop antennas that can pick up common-mode noise. You would gain the most by using twisted pair for the voltage sense leads.
 

tracecom

Joined Apr 16, 2010
3,944
Kelvin connections were new to me, and I redrew #12's design to make it easier for me to understand; that schematic is attached. As you can see from the photo, I did not use coax in the mechanical design. The PCB layout is quite simple; a graphic of that is also included. There is no on/off switch. When the DUT is connected, the LED lights; when the DUT is not connected, there is no current flow and therefore no battery drain.

To make the adapter easier to handle, I used double sided tape to attach the PCB assembly to the back of the battery holder (as shown in the attached small photo.)
 

Attachments

Last edited:

Wendy

Joined Mar 24, 2008
23,415
Another thought, between the LED drop and the voltage regulator, this would make a good LED tester, if you reverse bias the LED it wouldn't blow.
 

Thread Starter

#12

Joined Nov 30, 2010
18,224
The coax idea was a remnant from something tracecom saw on the internet. I tried to incorporate his idea, but the final version ended up without any coax.

The MAX6035 has a temperature drift of 25ppm/C and 1/2 % initial accuracy. The LM317 has a temperature drift of .7%/165 C and 4% initial accuracy. If I didn't slip a digit, that's 42.4 PPM/C.

Relevant to you? You decide.
 

Thread Starter

#12

Joined Nov 30, 2010
18,224
Isn't 0.7% at 165'C still a 0.7% error? ie will still be 0.7% for every one of those degrees?
No. 1/2 doesn't mean 1 at 2. It means 1 divided by 2.
.7%/165 C doesn't mean .7% at 165 C. It means .7% per 165 degrees of change in temperature.
 

ScottWang

Joined Aug 23, 2012
7,397
No. 1/2 doesn't mean 1 at 2. It means 1 divided by 2.
.7%/165 C doesn't mean .7% at 165 C. It means .7% per 165 degrees of change in temperature.
I did the similar way as THE_RB to measuring the resistance of wire but I didn't think about 10mA so less current what can it do as you.

I would like to introduce this method to other forums in Taiwan, what do you think if I using LM317 to replace MAX6035, do they really have a big different?
 

Wendy

Joined Mar 24, 2008
23,415
Truth, I don't think it will make much difference. The meter accuracy is probably a lot more critical, since it calibrates the current and measures the voltage.
 

ScottWang

Joined Aug 23, 2012
7,397
Truth, I don't think it will make much difference. The meter accuracy is probably a lot more critical, since it calibrates the current and measures the voltage.
I just thought in this testing circuit that the current is so small, so the affect of temprature maybe not too much.
 

Wendy

Joined Mar 24, 2008
23,415
Planning on testing something outdoors in the Sahara or Arctic? :D

Room temperature is pretty narrow for most folks.
 

ScottWang

Joined Aug 23, 2012
7,397
Planning on testing something outdoors in the Sahara or Arctic? :D

Room temperature is pretty narrow for most folks.
Maybe you should ask Kevin Lin - Taiwan's Ultra-marathon Hero, he has been ran cross to the Sahara and Antarctica, and almost die there.

He is a strange person but he got a strong will, he didn't win the 42Km marathon in Taiwan, but won the world championship Polar ultramarathon, if he bring any equiment with him then must be care about the ultra weather and big range temperature.

The weather in Taiwan comparing with USA then Taiwan is a warm place, but I think still have some places out there in the USA is hotter than Taiwan, I got an experience long time ago, a testing bingo machine made by wood which was the orignal design in Mit USA, when we assembled and packaged and tested ok, but when it shipping to the Las Vegas, something happended was that the wood has a little Split in many parts, when the engineer checked the machine and found it was split because the weather is too hot in Las Vegas, so the glue can't take it any more, Taiwan is a island climate, a lot of moisture in the air, so could keep the glue together, but the Las Vegas is too dry, so it should be changed to another glue.
 

Thread Starter

#12

Joined Nov 30, 2010
18,224
I think the low humidity is the real problem. The wood shrinks from lack of water in the air. No glue can stop that from happening.

PS, there is also a 1 amp current driver in he finished projects section. Also a very nice design if 1 amp is what you need.
 
Last edited:

Thread Starter

#12

Joined Nov 30, 2010
18,224
Truth, I don't think it will make much difference. The meter accuracy is probably a lot more critical, since it calibrates the current and measures the voltage.
One reason to choose the MAX chip is that the initial accuracy (1/2%) is so good that the trim pot can be designed for very good resolution. It will easily do 10.00 ma and might even do 10.000 ma. My analog design criteria is to try to get the circuit better than necessary so you don't have to consider the digit you can't see.

As a matter of curiosity, I would love to see if this can be calibrated to 10.000 ma, but I don't own a meter that good.
 
Last edited:

Thread Starter

#12

Joined Nov 30, 2010
18,224
Hey are there any design updates for this ??
Thanks
No, but there is a 1 amp driver in the Completed Projects if that will do your job better than this one.
I designed this for 10 ma so we could use (one) 9V battery. Good enough for LEDs and emitter resistors in power amplifiers. In my opinion, trying to measure tighter than 0.01 ohm is not necessary, but you can re-design it for 100 ma if you use bigger batteries.
 
Top