Good Digital Multi-Meter in $50 Range

Why would you not compare the meter to another one if several items gave the same, out of spec readings? I had a minimum of four 4 1/2 digit DMMS on my three benches at Microdyne. They were compared at the start of each work day. If I found anything wrong with any piece of test equipment I took it to our cal lab to exchange it before I started working. Sure, I spent five or ten minutes each day, but I had the highest output and lowest return rate of any tech in the plant. i also kept the cal lab busy repairing minor problems with the test equipment, rather than waste time fiddling with some touchy test setup. For instance, the AGC boards had four power supplies, and had to be set to exactly +5.000 VDC, -5.000 VDC, + 12.000 VDC and -12.000 VDC because the total allowable error was +/-1.5 millivolt at either 0.000 VDC or +5.000 volts output.

Seems to me that a finiky design like that is a poor design and should be tossed and re-done from scratch.

Sorry, that was a bit confusing, it was actually only one item, the meter reading of which looked like a common known fault condition in that product line that had a standard repair technique. So without anyone sufficiently technical on hand to question the result, they believed the multimeter reading. That one reading with a cheap meter cost about $60,000 in actual cost, but the real cost was probably well over $100,000 when you include the loss of income from survey time, plus loss of face with the customer. That's an expensive meter reading!

Dave :)

What the hell do you know about it? It was a LINEAR 0 to +5 VDC AGC system used in a $20,000 telemetry receiver that were used in pairs in a dual diversity receiving system to track deep space satellites. The test were that critical to ensure that any pair of receivers could be used together without being aligned as a set. The design worked so well for NASA that NOAA bought a packaged system to track their weather satellites for the Wallops Island station. The military uses them to track remote controlled missiles and drones because they are so well designed. BTW, there were 14 op amps in that AGC circuit, with a total allowable error of 5000/1.5 system wide.

That tight AGC voltage system allowed a pair of receivers to track within .01 dB to let it recover data from very noisy signals. In fact, with the Doppler compensation we designed, NOAA could lock onto their birds a full five minutes over the horizon. They were considered the BEST analog telemetry system in the world at the time they hit the market, and we kept refining the design, and adding request features. The Italian government bought two turnkey systems from us, as well. One fixed, and the other in a large military type communications trailer for their efforts to launch rockets. The mobile unit also had a special color IR camera and tracking system to follow the rocket after it left the ground, till the engines shut down. A second tracking mount aimed the antenna to track the rocket. The biggest problem was that we had was the very high received signal level at launch could overload the equipment, so gain control had to be added to the LNAs without affecting the noise figure at low levels.

Yes, it was expensive, but it is a bad in house process. If there is a common problem like that, there should have been some requirement to verify the problem before spending that amount of time and money on repairs. If we ran into a problem like that it would have been verified on another bench by a second tech. I built test fixtures, wrote test procedures, and worked with every department in the company to improve our products and keep customers happy. I even diagnosed and helped someone in Antarctica troubleshoot a problem via the net, rather than them return the equipment they thought was bad and wait almost a year for it to be shipped to us to find that the problem was in their setup and cabling.

I never advocated using a $3 meter for work like that, and I have never seen a 4 1/2 digit meter for that price. We had a lot of Data precision 3 1/2 and 4 1/2 digit meters, along with some Fluke and HP 5

BTW, I put in over 10 hours overtime one night and saved the company over $500,000 US. We were going to miss a early delivery bonus from one customer, and fall short for production for the quarter because of delivery problems with some National Semiconductor ICs. I tested, aligned, and did minor rework on well over 150 circuit boards that night, and left another 150 failed ones to finish after the rush was over. I pissed off four other techs by doing so much work in that time, but they finally got over it. :) QC couldn't get to their desks the next morning because of all the completed work, and the rework ladies never did get over it. Oh well! ;-)

Nothing. We are not talking about using the $2.99 meter in a professional setting, as your bizaare example below presents. Sounds like the place was run by the three stooges.

Regarding your example: "no technical person was on hand" and determining that it was a "common fault in the product" rings an alarm bell. "Only QA were on hand" - another alarm bell. "and they said the meter was OK" - doubtful. They probably said it is OK to use that meter in the procedure, but probably were not saying that the meter was working properly. After the "repair" was done using standard procedure, the same failing meter magically "unfailed" and indicated the repair was successful? No testing of the "repair"?

Exactly.

Ed

Relax Ed, it's only a story that shows how a failing cheap meter *can* cost big dollars. Yes, the circumstances were fairly unique, but fact is it happened.

How often do you question the reading on your meter? do you measure everything twice with two meters just to be sure?

Yes, the repair was re-tested with the same failed meter and of course showed the same "fault". It then clicked to them that the meter was faulty, and the problem was quickly resolved.

Dave :)

On Wed, 05 Jul 2006 22:20:50 GMT, ehsjr wrote:

:Ken Hall wrote: :> I just dropped my 20 year old Radio Shack digital multimeter and it :> stopped working. I use it probably once a month so I want to order :> another DMM. :> :> Does anyone have a recommendation for a good basic DMM in the $50 +/- :> $10 range? :> :> Most of the ones I've found on the web don't provide accuracy :> specifications. For example, I found this one that looks like about :> what I want but no specs

I have around 5 of those $2.99 Harbor Freight meters. They get light use, but none have failed on me yet (had them around 2 years).

Aside from that I have a Metex M-3800 DMM, from around 1992 and an old Radio Shake analog range doubler, that I use very rarely. No meters have ever failed me, but I never seem to require accuracy, just a ball-park figure suffices in most cases.

I didn't even know autoranging meters were available! Tell me this: If I set it to amps, will it work for both AC and DC?

Dan

PS This thread has me wondering if I should look for a used Fluke, but I don't figure I need it.

No. Why should it?