Multimeter adjustments

[cheechi]

I have a Fluke 27/FM that I bought a while back at the recommendation of GJ.

It seems to work ok for AC & DC volts but it is not accurate at all for ohms.

So now at this point, I have one or two HF freebies just in case something looks like its going to fry a meter. I have an Innova 3320 also, that seems to be ok (although to be fair I would otherwise use the Fluke to compare) it reads things in the approx range I'm expecting. 4 ohm speaker reads 4.2, 16k pickups read 16.4k, and so on.

The Fluke has a really convenient kick stand for bench work and a big enough display (and can be read from a lot wider angle) so it has value to me in that regard. However, I am not using it in a professional capacity and couldn't really justify the cost of getting a full calibration with cert. I see some are selling on ebay claiming calibrated but offering no cert, and I could get a new one if I have to but I like this old guy.

Has anyone cracked one of these open and made adjustments on their own? I was thinking of getting some resistors and borrowing another fluke and seeing if I can just work the pots for resistance to get close enough for functionality.

Is it realistic I can do this on my own or is it going to need a full calibration machine on my bench to get it done?

 

[Rookie2]

I would say no ! Calibration is not for a novice. Send it to fluke if you think you need +- 0.01% .

most problems arise in the Leads or connections.

 

[ddawg16]

Short the meter leads....then do it again pulling the meter leads out and using a short piece of wire. If there is a big difference, you meter leads are the issue.

If there is an error on resistance, most meters have an adjustment for 0...but you need to know which pot it is inside.

 

[ca90ss]

4 ohm speaker reads 4.2

That's normal, impedance and dc resistance are not the same thing.

If you really want accurate resistance measurements you need to find a meter that will do 4 wire resistance measurements.

 

[cheechi]

Comparing two sets of leads, one set of slip-on gator clips to attach to the leads, this 27/FM and the Innova 3320

Both sets of leads, when shorted and clipped to each other will come to a steady reading on the Innova. I did it that way to eliminate whatever ripple or other possibilities me holding them waiting would add.

Both sets of leads, when shorted and clipped to each other will NEVER give a steady resistance reading on the Fluke.

So if I measure actual resistor. Say the driver in my amp, it will come to a steady reading but without a way to zero it I can't know what to subtract to get the right reading.

It could be something in a connection but it certainly seems like something causing electronic noise without enough resistance for it to read its having a tantrum. And it can read anything else I need so I don't think its a broken connection port for the leads to attach or an internal short.

That's normal, impedance and dc resistance are not the same thing.

If you really want accurate resistance measurements you need to find a meter that will do 4 wire resistance measurements.

Right, I was using that as an example of saying that would be fine. 4.2 on the Innova tells me I'm right about where I need to be. 10.0 on the Fluke, on the same speaker, but it won't give me a steady zero I can subtract and can't use the delta button without a steady zero to know what I'm subtracting.

 

[Citation]

I've fixed a couple of Fluke meters. For the most part Fluke meters don't go out of calibration. If they give you a bad reading it's because something else is wrong. Typically it will be part of the safety circuitry that has been damaged. If you are decent with electronics you can try to repair the thing (ask the folks over at the EEVblog forum). If you aren't up for fixing electronics put the thing on ebay with a note about the issue. The 27 is a tank of a meter (I have one as well). Happily you can get a full schematic on line and Dave at EEVBlog has done a detailed tear down of the meter.

BTW, when testing resistance you really should test with a resistor. Speakers are a complex load. That means their resistance changes depending on the frequency of the input voltage. So they might be 3.8 at DC but 8 ohms at 1000 hz AC. Basically they aren't a good standard.

 

[cheechi]

I have the manuals and addenda and a service manual, but the service manual is (probably purposely) vague and just refers to their test equipment and the tolerances.

I will look into the EEVBlog I have read some there but never really in depth.

Oh and the speaker was me testing the speaker, and that's how I discovered this meter having issues. I had to go dig up the other because I needed to know if it's 4 or 8 ohms and got 10.0 and knew it couldn't be right being that exact, if it was high for 8 ohms. But your advice is good for those that don't know DCR vs impedance.

 

[guy48065]

Many digital meters don't have any adjustments inside--they make adjustments in software. Many others have only a single trimmer inside that sets the basic dc accuracy (actually tweaks the ADC ) and ALL the other functions accuracy relies on the component accuracy.

The fluke 27 is an older model but I've always been impressed with its specs.
There might be a problem with yours if the resistance reading drifts. The "Rel" button zeros the initial reading (leads shorted) and all readings should only toggle the last digit, if any.

 

[royesses]

As has been said a speaker is not DC resistance. It is made up of DC resistance and AC reactance which is the property of an inductor to resist a change of current. There is also some capacitive reactance thrown in too. To test the ohms function get a new 1% tolerance resistor and check if it is in tolerance with the Fluke. You can't adjust unless you know the actual true value of the resistor and test leads.

 

[cheechi]

Yes sir. In the process of testing the speaker, I discovered the issue with the meter and tested the meter further to verify. Thanks though.

The Innova meter has a battery test feature, so while I was going into this I pulled the battery on the Fluke and tested it against other 9v I have here. I happen to have one fresh 9v reading higher than 9v, put that in in case the Fluke's was low. Still with the leads shorted I can turn off and on the meter (ohms is right next to off, how convenient)

initial: 05.7 on 300 scale. after approx 1 min ranges 7.2-7.5 on 300 scale
initial: 12.0 on 300 scale. after approx 1 min ranges 14.3-80 on 300 scale
initial: 11.8 on 300 scale. after approx 1 min OL on 30 scale
Something might have slipped on that third, I redid it a few mins later after resetting the gator clip
initial: 31.4 on 300 scale. after approx 1 min ranges 31.7-34.0 on 300 scale

All without touching it other than to turn the knob, on a fresh battery. I get the impression either the numbers will just climb forever or decrease forever if I left it long enough. THAT is the issue I'm describing. I can't get it steady, to zero it, to start using it for resistance, impedance, or chanters who say the word 'ohms' even.

Thank you all for the input, thanks for understanding I was not using the speaker to test the meter but the other way round.

 

[Citation]

For those who aren't familiar with the 27, it is a tank of a meter. It's a VERY rugged meter (water proof, VERY think plastic) and one with good accuracy. It's an old model, first released sometime in the 1980s. The main PCB is swapable with the last ones made perhaps 8 years ago. This meter was certified for mine use as well as by the military.

For all that, it is based around old hardware and was calibrated via trim pots and trim caps.
http://www.eevblog.com/forum/testgear/fluke-27-27fm-side-by-side-tear-down/

Cheechi, based on what you are describing you might have some damage that is causing a capacitor to charge slowly thus resulting in a constantly increasing offset in your reading. Typical failures in a Fluke meter like this would be the protection components. Either blown fusable resistors or over voltage protection components failing.

It may be worth taking a current reading. Basically put one of your other meters in mA DC mode and measure how much current the 27 is consuming when connected to a 9V batter. The 27 is rated as a very good 1000 hours of battery life. You should see something like 0.5mA of current draw from the 9V battery (more when you use the continuity beeper). If you are seeing say over 10mA then you probably have something causing a significant short. You also might take a resistance measurement of the 27's inputs when it's in resistance mode. It should be something like 10Mohm. If you get a reading much less than that (I'm not sure if your meters read up to 10 Mohm) then you can also assume something is wrong.

 

[cheechi]

shorted the leads of the fluke with a gator clip
innova leads at battery terminals of Fluke battery
on ohms, 0.00 mA to 0.01 mA
on continuity, same

changed battery, same results. Then I checked all the settings on the fluke. Got as high as 0.03mA for a brief second. Could be the Innova isn't good at testing amps, or that this thing's draw is in the micro amps. I don't know how to tell but I do appreciate the suggestion.

While I was testing the battery amps I figured I'd pull the fuses just in case. Tested each for continuity & resistance on the innova;
buss bbs1 0.8-0.9 ohm
buss ktk-15 0.2-0.4 ohm

Then I tested each of the fluke's leads to make sure it wasn't some continuity issue there
0.4 ohms on the red
0.3 ohms on the black

1.528 Mohm when i short red to red black to black, and both meters agree within .001 Mohm however I noticed I had (red in com black in ohms) on the fluke, so when I take the leads and put them (black in com red in ohms) on the fluke I can't get a reading on the innova. Maybe it's a bad COM port?

 

[Citation]

Sorry, I made a mistake. When the meter is on and set to ohms you can't get a good resistance reading because the meter is creating a voltage across the terminals. The result is the other meter will get a messed up resistance reading. I just checked a few measurments on my 27/FM.

When set to DCV I measured (with a second meter) 11 Mohm across Pos and Com. When set to ohms I measured 0.82V across the same terminals. That said, part is that you are seeing a higher than actual resistance reading and the offset seems to be relatively constant across all ranges (constant like 50 counts in each range). That constant offset failure tells me the problem is damage to something and most likely the input protection parts. If you can get a refund from the seller for selling a non-working unit, go for it. I'm almost certain a few of the EEVblog guys could ask just the right questions to pinpoint the issue.

One more note: The Innova meter has a somewhat limited resistance measurement range. It probably cannot measure things over 10 Mohm. The input resistance of a good multimeter is typically at least 10 Mohm so many low buck ones simply can't read that high. Regardless, I'm certain you are dealing with a damaged meter. Don't take this as a knock against Fluke. Sometimes the thing that caused the Fluke to stop working would cause a regular meter to literally explode.

 

[cheechi]

When set to DCV I measured (with a second meter) 11 Mohm across Pos and Com
that's what I get

When set to ohms I measured 0.82V across the same terminals.
0.788 v


Retried with the leads instead of direct to the terminals, same readings.

I'm sure the Innova is just as good as the $30-40 I spent on it. I really appreciate your guidance in this. In your opinion is it damaged beyond what can be calibrated? Its cal ran out in 08 but I didn't know that until after it arrived to me.

And I'm awaiting a response from an ebay seller in picking up a second 27/FM. This one was bought too long ago (don't remember where) to talk about a refund. Maybe someday I will be able to look at a new(er) but I like everything about the physical brickiness of this guy that once I got it I got turned off the style of more modern meters.

 

[Virgil Cain]

Putting aside the AC reactance, do not expect a 4 ohm speaker to read "4 ohms". Likewise do no expect a 10k resistor to read "10k".

Everything is made with manufacturing tolerances. Your 4 ohm speaker is probably built to something like 10% tolerance. Unless you specify a 0.1% tolerance resistor, you're probably going to get either 5% or 1%.

Unless you have dedicated equipment to calibrate an instrument, you have nothing with the required accuracy to calibrate against. Don't try it.

 

[firworks]

My recommendation would be to make an account and go post over on the EEVBlog forum's Test Equipment board. I'm sure they can walk you through any testing and either fixes or calibration if needed. They are nuts about multimeters over there.

http://www.eevblog.com/forum/testgear/

 

[Citation]

This is certainly something that would need to be fixed vs just adjusted. As a side note, calibration is actually just the check to make sure the meter is still in spec. "Adjustment" is what you do if the unit is out of spec. Anyway, Fluke meters should basically never need adjustment if they aren't damaged. I had a late 80s version of this meter that was still spot on when compared to a recent Fluke 87V. I would guess this one is repairable by a hobbyist. If you don't think you are up for it, list it on eBay with a description of the problem and someone will buy it.

I've had three versions of this meter. They sure are breasts and they used to be very cheep. The 8025A and B models are almost the same meter. The 27FM is TRMS while the other versions aren't. I think I paid $17 for one of them. Anyway, if you are at all inclined, try fixing it. If not, get another or one of the small box Flukes (23, 73, 79 etc). All great no BS meters.