I got spoiled. It used to be, many years ago, that Volt Meters were expensive and not very accurate. Now that they are cheap I can’t stop buying them. They seem like such a good deal in comparison. Anyway, I rounded up a quite a few and hooked them up to see how their readings compared to each other. Since the first time I opened one of these cheaper meters up, I noticed that they seem to be built around a “DMM on a chip” That is, all the important stuff in on one I.C. with a few other parts hanging off. Of course there is the readout, switch and banana jacks. Here is a pic of an opened, typical Harbor Freight DMM.
The black dot on the right is the DMM chip. In the upper left hand corner are the 3 banana jacks. They are soldered directly to the circuit board. Notice that there is a wire soldered connecting the left most jack and the right most. This is the shunt resister that is passes the current when using the 10A mode. That wire has to pass all of the current that is being measured. Typically this shunt wire is a .01 ohm resistor and to measure current passing through, the meter measures the voltage drop across it. It will be very small.
To the right of the battery is a replaceable fuse and in the lower right corner you can see an adjustment. I would imagine that it is a global calibration setting that after the meter is assembled, a technician adjusts a voltage reading to a known source. I am guessing that adjusting this once on a certain scale will also adjust for all the other scales as the digital circuits all use this one adjustment. So it would once and done. The only other calibration would be for the 10A scale, more on that later.
With cheap meters, the mechanical components are the weakest link. The switch, battery connector, jacks, and especially the test leads.
To see how 10 different meters compared, I first measured ordinary US household voltage. In this measurement it isn’t that important to have good test leads. They are only measuring voltage, very little current will pass through them, so I used the stock test leads. Here is the pic
Included in the picture are 5 Harbor Freight Meters, 2 Sears Craftsmans, a large digit Excel, an “Ideal” and a Cal-Hawk. All meters measured within 1% of their collective mean.
However, what seems to concern us most on BLF is current measurements. Can cheap meters read current levels accurately? Well the answer is no, not with the stock test leads. As many of us know, it is VERY important to have excellent electrical connections in higher current applications such as high powered LED lights. For this reason, I did not use the cheap, junky test leads that came with these meters, but rather used 14 gauge copper wire. For this current test, I connected 14! meters in series with each other, and then to a TMart C-8 XM-L running on a laptop pull cell, a Sanyo 2200mAh unprotected. Here is a pic before the light is connected, all meters read 0.00
This next pic shows the readings of all 14 meters as the current passes through each one. Normally this light draws 3.2A on this cell, but in this case, inserted into the circuit is 14 meters and 28 physical connections. Even though I used 14 gauge wire, the 28 connections and 14 x .01 ohm shunt resistors have taken a toll.
What’s important here is that for the most part, all the meters agree with each other! What this means is that all of these meters were calibrated at the factory. Only the HF meter in the upper right corner and the large digit Excel read a little low.
In the past, when I have changed out the battery in the Harbor Frieght meter, I have noticed either notches or gobs of soldier on the shunt resistor. At the time, I assumed that the notch was put there during calibration to increase ever so slightly, the resistance of the shunt and the gob of soldier was put there to reduce it. It turns out I was probably right. What I am about to do now, I have never done before. I will attempt to adjust the current reading of the HF meter that is reading too low (1.40)
Notice that I had put a black line on the switch of the meter in the upper right corner. Here is a pic of the unmodified shunt resistor.
The current reading of this meter is too low. This meter is reading the voltage drop across the resistor according to Ohms Law, E=IxR. If the reading is too low, (E) then R must be increased. Squeezing some notches in the wire will increase resistance!
Now I test again. Now the reading is too high! Those 3 notches increased the resistance too much. Again notice the black line on the switch.
Now the resistance needs to be lowered, a bit of solder can do this. This next part wasn’t easy, it took several tries. I’m sure the technicians know just how much will do in each instance. Experience makes hard work easy.
This next pic is after some fussing and a cell change. The important thing is that the meter now reads the same as the others.
I believe what all this means is that these cheap meters can be used for our casual purposes. If I was a professional, I certainly would own and use the very best meter I could afford i.e. Fluke or some such.
BTW, please don’t get the wrong impression. I own a lot of tools and when it really matters, I get the very best. I enjoy using them and owning them. Maybe someday I will own the very best meter, but for now, these will do.
For those intersted, Here is a pic of my EICO, bought in kit form in 1966. $39.95 and it had to be calibrated after it was assembled. The calibration consisted of connecting it to a wall socket and adjust for 117V. The DC calibration was the measure a fresh Carbon-Zinc D cell and adjust for 1.65V.
This next pic is of a Digital Volt Meter that I bought in 1975 for $69.95. It also was a kit, Sabtronics Model 2000