Gossen Metrawatt METRISO 5000A, Insulation Tester 5000V [Success]

[HighVoltage]

I have not seen this on on eevblog and thought I will share the repair and upgrade.

Gossen Metrawatt METRISO 5000A, Insulation Tester 5000V
It was probably made in the late 90s and seems like new but is not working.

Symptom:
Not working at all.
The 6 x MONO batteries are new.


Also Problem:
These 10 mm cables are so stiff, when you move cables, the instrument is moving.
It is horrible and not really workable.
It seems the cables have lost all the softener in the plastic.
So, my plan is to exchange these hard and stiff cables with very flexible high voltage silicone cables.

I am starting this tread, before I open the instrument.
The user manual and datasheets are available
But I can not find a service manual or the schematics

May be someone here in the community has seen a service manual?
 

[HighVoltage]

Next step, removed the big stiff cables and took the PCB out
Here is the back side of the PCB with voltage multiplier and what looks like a high voltage precision divider resistor.
Nice to see those red WIMA MKS4 capacitors.
There was a time when almost every instrument in Germany had these type of capacitors installed

[HighVoltage]

Some more pictures of the PCB back side

[HighVoltage]

Here are some pictures of the front PCB
There are actually two PCB's in this construction and the high voltage is completely separated from the control circuitry.

[HighVoltage]

It seems I have found the problem.
No voltage after the two in parallel input protection diodes.
They are marked: S4 77 GI and I can not find a datasheet
They are open in both directions.

When I hook up a power supply after the diodes, the instrument seems to work.
At least it is turning on.

I think these are low dropout germanium diodes?
May be 0.2V forward voltage drop?
Possible?

Any hints from anyone of what I can replace them with?

The only purpose seems to be polarity protection, I guess.
Any other ideas?

Thanks for any feedback.

[RobK_NL]

Damn, that thing is pornographic!

The diodes are SS14 Schottky's. http://www.digikey.nl/product-search/en/discrete-semiconductor-products/diodes-rectifiers-single/1376383?k=ss14

[olliisit]

Depend on Package Size:

S4 =  SS14 or SS24 or SS34

General Semic. / Vishay

Shotkky Barrier Rectifier

[HighVoltage]

Thank you guys, that was very helpful.
I will see what I can get locally for now and then add the correct diodes later.

[HighVoltage]

OK, I installed some new diodes and the instrument is working
I have no idea why they broke.
I did suspect some shortage in one of the tantalum capacitors but that was not the case

Now I want to modify the cables

[HighVoltage]

I looked around for some good high voltage probe tips but I could not find any.
Maybe someone here has some suggestions.

So, I decided to use the old handle and remove the cable.
The original handle is molded around the cable and the 4 mm banana plug.
I started with a 3 mm drill bit and then went up to 7.0 mm
It really turned out much better than expected.

On the MINUS side, the shielding and 4 inner connectors are connected in parallel to the instrument
But on the PLUS side, the shielding is connected to a spark gap on the tip and with a 47 kOhm resistor to the ground plane of the PCB.
I don't care about the functioning of the spark gap.

[HighVoltage]

Then I inserted the 7 mm silicone cable in to the handle
As a new 4 mm Banana plug I use a MC part and made the tip pointy
Added a small plastic sleeve
And pressed the assembly in to the handle with a small press

[HighVoltage]

Next, hooked up the cables to the inside of the instrument
Since these are 7mm cables, and the old one had 10 mm diameter, I added a silicone tubing at the outlet.
And the PLUS cable got a red piece of shrink tubing for quicker polarity identification.


 

[HighVoltage]

Fully assemble and working.
The spark is jumping good and stable at the 5000 V setting

Next is to test the voltage output in the different settings.
Hopefully all settings are correct, since I do not have a service manual with calibration instructions.

[HighVoltage]

Testing 100V, 250V, 500V
The 34461A is going to "overload" at 1000V setting
Hmm, should the Agilent meter not at least show like 5% above 1000 VDC like the 34401A?

[HighVoltage]

1000 V setting, measured with a 34401A

[HighVoltage]

And the last test,
5000V with a 1000:1 high voltage probe
Reading about 3% high

All perfectly working and I am very happy with this upgrade.
Anyone interested in one of these testers, I can only highly recommend them.

[RobK_NL]

Wow, congratulations on a fine job. That looks excellent!

I can only highly recommend them.

I'll second that. I have the 'little brother'; the 1000D (for digital) and have always found it to be a very good instrument.

[elimenohpee]

Very impressive!  The modified cables look really clean and professional.

[Vgkid]

Nice job, the green cables really stand out.

[BFX]

Very nice modification of probes

[carl_lab]

You replaced the 2 schottky diodes by non-schottky BYV96?

[HighVoltage]

You replaced the 2 schottky diodes by non-schottky BYV96?

That was for testing purpose only and it worked.
A few days later I installed new Schottky SMD diodes.

I bought a second Metriso 5000A very cheap on ebay Germany and it had the same problem.
Now I have two fully functioning 5000V insulation tester.

[MarkoET]

I recently bought cheap Metriso 5000A at unknown condition. Normally, I would first check the instrument, but this time seller was from different city and I am short on time, so I decided to gamble with this... And was losing at the beginning.

At first, Metriso acted very strange. At voltage switch position at "OFF,V" it gave 100Vdc output. Also it gave 100Vdc output at other switch position (from 100V to 1000V). At higher positions (up to 5000V) it gave no life, like it is in OFF position. So i searched at internet (not much info), came across this thread, read what HighVoltage was experiencing and decided to open the Metriso.

Long story short, I found inside instrument that voltage switch position on PCB was at some strange position. I needed to "calibrate" it myself. So I used the picture provided by HighVoltage here in thread it determinate what is the "OFF, V" position, since you cannot guess it blindly, because rotatory switch on the outside of the instrument must be inline with the circle "switch" on the PCB inside of the instrument. So, after all clicked, I put instrument back together and put it to the tested.

And it worked. At "OFF,V" it was off, at 100V switch position, it gave about 110Vdc output, at 250V pos gave about 260Vdc output, and so on. I managed to measure at 1500V switch (1591Vdc output), but for higher voltage I cannot measure the output, but I assume that it is now ok.

 

[MarkoET]

But, still it is now working fully. Even it worked at the beginning, the analog needle is now stuck. 
If I tap on it, it moved less than mm at the direction that it is supposed to move (left, if it's probes are short-circuted  and right at other case). But, it should move like this:    So have to figure out now what to do. I did notice that little zero correction regulation is not working anymore (it does not move the needle), so it might have impact on needle being stuck.

Any ideas about stuck analog needle? I mean, not fully stuck, it still moves "inch by inch" if I keep taping on instrument, but it must work normally.

[Gyro]

The Gossen is an insulation tester for electrical installations. It will always output more than the indicated voltage to guarantee that it tests the installation with a terminal voltage of at least the indicated voltage at its maximum specified leakage current (1mA according to BS7671 in the UK).

The voltage will (should) always read high to ensure that this worst case requirement is met.


Edit: Regarding the meter sticking, it sounds like mechanical impact damage especially if the zero adjust has stopped working at the same time. Check for jumped turns on the hair springs or debris in the magnetic gap around the moving coil. This is a delicate job.