Fluke 89 IV slow settle in 5V range

[bdivi]

Hello,

I have recently purchased a used Fluke 89 IV. The unit seems to be in working order and in tolerance.

There is one issue however that I need help with.
When measuring DC voltage the 5V range takes a couple of seconds to settle and that is very distinct compared to all other meter ranges and also compared to my Fluke 187.
I assume that it is an issue with the front-end but I do not have schematics and have no idea where to start the diagnostics.

I will appreciate any advice or schematics (I read the circuit should be the same for 87 IV as well as 187/189).

Cheers

[retiredcaps]

I assume that it is an issue with the front-end but I do not have schematics and have no idea where to start the diagnostics.

You probably won't find schematics for the 87IV, 89IV, 187 and 189.

If you got this used and abused, you can try the following:

1) wash the pcb with IPA and let it dry thoroughly.
2) check the PTC - it should measure around 1.1kohm.
3) check the MOVs - it should measure 0L (infinite) resistance with another multimeter.
4) check the fusible resistor - it should measure around 1kohm.

[bdivi]

Thanks retiredcaps,

I did wash the board and also visually inspected it and found no apparent issues.

I will make the measurements you suggested and report back in the coming days.

Cheers

[retiredcaps]

Other suggestions.

5) Clean input jacks with IPA.
6) Inspect and possibly resolder/reflow solder on input jacks.

[bdivi]

All input protection circuitry checked - unfortunately no issues found.

I can measure however the same slow settle problem right at the ADC pin, which makes me believe that there is still something wrong in the path from input to ADC - clipping diodes, range dividers, filtering ...

Any help is appreciated

[retiredcaps]

Long shot, but is your line frequency on the 89IV setup as the same in the 187?

[bdivi]

yes, it is 50Hz  and I also tried 60Hz - no difference

[ModemHead]

When you say slow settling time, do you mean it slowly rises to the correct reading, over-shoots and falls to the correct reading, or bounces around?

I don't have a schematic for that model but I'll throw out some basic stuff based on other Fluke models, maybe it will give you ideas.  In DCV mode, the input should be coming in through a 10Meg resistor straight into the ADC.  For 5V range, I'm guessing the divider would be in 1/10 mode, meaning the controller will bridge the input via an internal switch to a 1.111Meg resistor, the other end of which is grounded (probably in common with the other values necessary for 1/100, 1/1000, etc.)  All these should be on the thick-film resistor array on the LCD side of the board.

I can't quite imagine what sort of fault there would create a time constant slow enough to observe, but hey, have a look at that divider resistor array.  If the symptom only occurs on the 5V range, then it seems to be the likely spot for something to go wrong.

[bdivi]

Thank you ModemHead,

When I say slow settle I mean it rises quickly to within 50-100mV bellow the measured value and then takes a couple of seconds to reach the final value. Just like charging a capacitor, no overshot whatsoever.

I also noticed that the meter is somewhat slow with the continuity beeper - again next to 187 that never misses.
All other modes and ranges are much quicker but when I carefully compared to the 187 side by side the 89 takes a couple of cycles to settle like charging a capacitor.

Overall it looks like an RC time constant that is larger and most visible in the 5V range.

Schematics of similar Fluke meters would help

Cheers

[ModemHead]

Schematics of similar Fluke meters would help

The first-generation 80 series have full schematics in the service manual. These models are old, but the section at the end includes "signal flow" diagrams that are instructional, and not all that different from many models.  It's too bad that service manuals and real schematics are a thing of the past. 

http://assets.fluke.com/manuals/83_85_87smeng0500.pdf

As to your problem, I'm not sure what to recommend.  Maybe check the input impedance for anomalies?  This 189 I have here measures 10.998Meg on the 5V DC range.  It will go down as the ranges go up, as the controller switches in smaller resistors for the bottom end of the divider.

[retiredcaps]

Another long shot.  Looking at modemhead's 189 pictures, I see three green 100kohm resistor which may or may not be power resistors to limit current for input protection.  You may want to verify they are indeed 100kohm?

[retiredcaps]

I got a new to me 89IV and just tried measuring a simple 1.5 AA cell.  I am noticing the same slower settle time compared to a 187.

I guess we will have to see if more 89IV owners report the same or not?

[ManateeMafia]

Do you notice any difference if you disable autoranging?

[retiredcaps]

Do you notice any difference if you disable autoranging?

No difference in auto or manual range (5V range).  If it isn't obvious, this is in the 50,000 count mode.

The 89IV seems to display five to seven readings before settling on the "correct" one.  It doesn't overshoot at all, it just takes time to "ramp up" to the correct reading.  It is like smoothing mode is enabled or something?

I'll try to take a video over the next couple of days and go frame by frame so I can get the exact number of readings.

[bdivi]

That is exactly what I see as well - some sort of smoothing in software I guess because I do not see any capacitor.

The questions here is why only the 5V range ?

Fluke on the other hand responded in a mail that this slow settle should not be normal.

==========
Good Morning,

8Xiv series was replaced by the 18X series in October 2000,  the two series are very similar apart from the PC interface being slightly different.  I have asked the service team who have researched their knowledge base and there is nothing to indicate that the slower response you have noticed has been reported as an issue.

Regards
Phil Sabberton
Fluke UK Tech Support

[AndyC_772]

I got a new to me 89IV and just tried measuring a simple 1.5 AA cell.  I am noticing the same slower settle time compared to a 187.

I guess we will have to see if more 89IV owners report the same or not?

I have an 89IV, and I'm afraid the symptoms described don't apply to my meter. The continuity beeper is extremely fast, and there doesn't seem to be any real difference in settling time between ranges.

It looks as though the meter takes about 3 readings to fully settle. The first reading after connecting a voltage source is low, which is exactly what I'd expect given the fact that it's going to be averaging the voltage over the whole of the sample period. The second reading is almost spot-on, give or take a few counts, and the third is the final, stable reading which it will maintain indefinitely.

[bdivi]

Here is a video of the 89IV and 187 meters side by side.

The first part is 2V slow motion x0.25 were the difference is clearly seen - the 187 takes less than a second to stabilize and the 89 needs 3-4 seconds to do the same. You can notice the same problem when going from 2V down.

I also captured the same 2V in the mV range - no difference whatsoever. Same for 20V



Interesting !

[AndyC_772]

My 89IV behaves just like your 187.

[Robomeds]

I'm going to have to check my 89IV the next time I get a chance.  I don't recall noticing that it was slow in any range but I haven't used it that much. 

[retiredcaps]

I don't see a way to display the firmware level of the 89IV so can people please list the first 4 digits of their serial number?  Maybe Fluke made a firmware change to get rid of the "smoothing" mode?

[retiredcaps]

Here is a video of the 89IV and 187 meters side by side.

Since you already have a case/dialog with Fluke technical support, can you point them to this thread and your video so they can see the problem first hand?

The chances of your 89IV being "broken" due to a bad component is now remote since I have the same smoothing behaviour and we are both in different parts of the world and probably acquired our meters in totally different ways.

I'm now suspecting they made a firmware change?

[retiredcaps]

The first part is 2V slow motion x0.25 were the difference is clearly seen - the 187 takes less than a second to stabilize and the 89 needs 3-4 seconds to do the same. You can notice the same problem when going from 2V down.

You can also see in your video, the 89IV displays 8 readings before getting to 2.0035, whereas the 187 displays 3 readings to get to 2.0034.

At 4/sec display update (as per manual), that suggests the 89IV takes 2 full seconds (at least) to display 2.0035 and the 187 takes 0.75 seconds to display 2.0034.

The 89IV 8 readings: 0.3884, 1.5171, 1.8709, 1.9672, 1.9935, 2.0025, 2.0031, 2.0035.
The 187 3 readings: 0.8376, 2.0031, 2.0034.

PS. Great video capturing the behaviour.

[dtmf]

I have the same problem with 89IV and 5v range.
Any solution?
Fluke 87 III is 2 times faster.

[daddario]

My 89IV is also slower than, for example, my old 85III, even in 5000 count mode. But not several seconds slow and nothing exclusive for the 5V range.
What's more irritating, is the slow continuity buzzer with its silly chirping sound. I mean, it's still fast, but not quite as gripping as any other "true" 80-series I've ever used. I can without a problem persistently grate my probes together without the meter making a single beep.

[bdivi]

Continuity also seems to be slower on my 89IV compared to 187.

It looks like the 187/189 has improved over 89IV after all.

Anyway now that I know it is not my meter fault I am using it for current measurement as my second meter.