Re: Test instrument accuracy vs stage testing during kit builds


Don Wilhelm <w3fpr@...>
 

Jim,

Precision and accuracy are two different things.
If a voltmeter is capable on showing 4 digits, that is an indication of the precision - in other words, the steps from one displayed value to the adjacent value.
Accuracy is a measure of just how true to a voltage standard any of those readings may be.

So you have lots of precision, but there is a question about accuracy of your various instruments.
You are like the man with many watches but does not know what time it is.

I do have a Precision Voltage Reference on my website that you can build and use to determine the accuracy of your various voltmeters.  Go to www.w3fpr.com and click on the article in the left column.

73,
Don W3FPR

On 6/21/2014 5:46 PM, jdcompton@... [4sqrp] wrote:
 

I may be over-thinking a problem I'm observing regarding some of my test equipment (specifically voltmeters) which I use during construction of kits and home-brew projects.  I just completed the SS-40HT and started the SS-40TX yesterday.  I had a few bogus step test readings when I was building the HT and now am "stuck" at the Step 4 Test in the TX.


I sent an email to Jim, but in the Step 4 test, I'm not getting a voltage drop.  The Step 2 & Step 3 test were satisfactory.....I think.....


Here's the rub.  I have several DVMs, including:  

Data Precision 1450 bench DVM

Wavetek BDM 35 bench DVM

2 Fluke 77 hand-held DVMs

Ideal 61-361 hand-held DVM

a couple of Simpson analog meters which belonged to my father.


My first problem was with the SS-40TX Step 3 Test where a voltage reading is supposed to be 0.2V or less than the power supply voltage.  Initially, it looked like the test was a success....until I started checking the output voltage of the "regulated" variable bench DC power supply.

Most of my voltmeters indicate that the actual output voltage is between .2 - .4 volts higher that the LED readout on the power supply itself.  (Example, when the bench supply shows 12.0 volts, the voltage at the tip of the cable measures 12.1 +/-

When I used a measured PCB input voltage instead of the bench supply voltage readout and conducted the Step 3 Test, I got 0.25 V difference.....which is above the recommended voltage difference.  Now.... was the test a PASS or FAIL?   Or is the 0.05 v excess voltage critical? And if the test FAILED, what am I supposed to do about it, assuming the build is correct at that point?


In the Step 4 TEST, I'm getting a 0.01 volt drop....supposed to be 0.2 volts.  Double and triple checked all components, solder joints, everything.....Test FAIL...  Now I'm stuck until I hear from Jim.  Looks like a problem with Q2....maybe?


Then, just because things were getting muddy, I compared the voltage readings of all 5 voltmeters using various DC power sources (batteries, bench supply, etc.).  The comparitive readings are inconsistent, but the best I'm seeing is an average of 0.03 volts difference of 4 and as much as 0.45 volts difference, usually from the Data Precision 1450.....BUT, it's not consistently the "black sheep" in the out-of-the average readings.


I'm getting paranoid.  I don't know how to test a "known" voltage source to determine which, if any, of these DVMs are accurate at any given reading, since I have no way of  knowing exactly what the voltage of any voltage source is, precisely.  And I don't know the degree of precision required when performing tests at various stages of building a kit.. 


It's very frustrating having so much money tied up in test equipment and having such inconsistent readings from multiple supposedly "precision" instruments.  And I just don't have the knowledge to calibrate the equipment or financial resources to have all the various pieces of test gear professionally calibrated.


Any comments, words of wisdom or advice?


Thanks.  Sorry for the long post!


Jim Compton, K4JDC



Join main@4SQRP.groups.io to automatically receive all group messages.