Fluke PM6306 RCL Meter Discussions + Teardown

[plesa]

In attachment is pinout.
Just to make it clear for every fixture you needs to make the short and open circuit calibration.

[BravoV]

In attachment is pinout.
Just to make it clear for every fixture you needs to make the short and open circuit calibration.

Not only that, every fixtures even factory made has it's own value, need to set accordingly when changed.

I was so happy when I saw the table below for the 1st time, meaning even for "non standard" fixtures like the "DIY type"  , it can compensate it, cmiiw.

Hell, I don't believe even a poorly made fixture as long not using overly long cable will get up to that crazy 1050 pF  , again cmiiw.

[HighVoltage]

May be someone is interested in the official prices. I saved this UK price list in 2012.
Retail price of the top of the line PM6306 was £ 8034 (GBP !!!) = 11400 Euro in today's currency.

[Shock]

So wow only one 4 wire measurement in the adapter box and every wire is shielded, so that means they must run connections to multiple sockets to support both test positions. I wasn't thinking about test frequency. Should be interesting to see how they wired it.

[HighVoltage]

Here are some detailed pictures of how the connector looks in the back and how the 2 mm banana plugs are connected to the LEMO connector.
The flat flex cable plugs in directly to the header that is visible on the PCB

[BravoV]

@HighVoltage, crazy price, I guess I'm so lucky. 

So wow only one 4 wire measurement in the adapter box and every wire is shielded, so that means they must run connections to multiple sockets to support both test positions. I wasn't thinking about test frequency. Should be interesting to see how they wired it.

Now, regarding the cables used, this particular note below from the PM6306 User Guide (highlighted in yellow and red) really made me raised my eyebrows.

That means definitely they've upgraded the cables using much better dielectric and also most important much more stable dielectric. This meter deals with like 0.01pF or 0.01uH resolution and with frequency up to 1 Mhz, I guess every pico Farad counts.

Thats why I'm so interested how they built that adapter box, the cables type used, the cabling layout, the shieldings/soldering and etc. Also inside the Lemo as it uses at least 4 coaxial cables, too bad its locked.

[HighVoltage]

Now I took the Adapter 4542A apart.
Not a big surprise, just that the quality of the details are really great.
I am the first and original owner of this adapter. I think I must have bought it in the late 90's from Fluke USA.
It seems this one was never re-batched with the Fluke name on it.

[BravoV]

Now I took the Adapter 4542A apart.
Not a big surprise, just that the quality of the details are really great.
I am the first and original owner of this adapter. I think I must have bought it in the late 90's from Fluke USA.
It seems this one was never re-batched with the Fluke name on it.

HighVoltage, thank you so much for the photos, really appreciate it. 

Ok, now the cables are spotted, Teflon coax it is, cmiiw. 

Inside the box, I can see also each shield from every coax cable are terminated close at the solder joint (green arrows) and heat shrunk.



I'm not going to make an assumption here, so questions for you HighVoltage :

- Is that black ground wire (pointed by yellow arrow)  is connected to the Lemo's metal body ? Measure with DMM please.
- Does pin 1 (as it is circuit's ground) also electrically connected to that black black wire ?  Measured at the Lemo's pin 1.
- Its just a plain non coaxial cable right ?

[HighVoltage]

NO, there is no connection to the outside of the LEMO plug.
This black cable goes to a guard, I am guessing a shielding for the complete cable.

But if you plug the cable in to the PM6306 and measure the same resistance, it is 100 kOhm
It seems that one of the resistors we see on the flat flex cable is a 100k Ohm resistor to connect the housing of the LEMO to the housing of the adapter.
 
Resistance between housing and LEMO  >10GOhm (not connected)
Resistance between housing and LEMO = 100 kOhm (connected)

hope this helps you.

[HighVoltage]

This is also a hiding fact:

All these original test cables have a guard plug for a 2 mm banana plug installed and you can add a ground / guard cable and connect it to the housing of your DUT. No wonder, these cables have been so expensive.

I think I have every cable made for this PM6306 in my collection. And they all have this 2 mm ground plug installed.

[Vgkid]

Interesting that they didnt connrect the shield to the chasis, but I guess it doesn't matter much for those frequencies. Unless that black wire meets up with the shield on the inside.

[plesa]

On Agilent fixtures they did it same way and in many cased there is only shield on each conductor.
Cables used un Fluke adater and pretty good with PTFE isolation and silver coated shielding, only the central conductor is too tiny and easy to tear.

If you are interested in building the test fixture adapter try to make adapter for higher voltage testing, schamatic is available in Agilent adapter documentation (page 25).
http://literature.cdn.keysight.com/litweb/pdf/16065-90011.pdf?id=1000002525:epsg:man

[BravoV]

NO, there is no connection to the outside of the LEMO plug.
This black cable goes to a guard, I am guessing a shielding for the complete cable.

But if you plug the cable in to the PM6306 and measure the same resistance, it is 100 kOhm
It seems that one of the resistors we see on the flat flex cable is a 100k Ohm resistor to connect the housing of the LEMO to the housing of the adapter.
 
Resistance between housing and LEMO  >10GOhm (not connected)
Resistance between housing and LEMO = 100 kOhm (connected)

hope this helps you.

Great ! Thanks for clarifying that black cable, its clear now. 

This is also a hiding fact:

All these original test cables have a guard plug for a 2 mm banana plug installed and you can add a ground / guard cable and connect it to the housing of your DUT. No wonder, these cables have been so expensive.

Wow ... without you revealing that hidden plug, nobody will ever know that little detail. 

Please, again .. take another close up at the red and black Test Posts ? As these have the lowest capacitance < 50 pF according to manual, intrigued by the details on how they built it.

Thank you so much for all your troubles helping me here. 

[BravoV]

If you are interested in building the test fixture adapter try to make adapter for higher voltage testing, schamatic is available in Agilent adapter documentation (page 25).
http://literature.cdn.keysight.com/litweb/pdf/16065-90011.pdf?id=1000002525:epsg:man

Thanks, another great resource on these RCL test fixtures build. 



Edit :

Hopefully this weekend I have time to do the teardown and post it's gut photos.

[HighVoltage]

Wow ... without you revealing that hidden plug, nobody will ever know that little detail. 

Please, again .. take another close up at the red and black Test Posts ? As these have the lowest capacitance < 50 pF according to manual, intrigued by the details on how they built it.

Thank you so much for all your troubles helping me here. 

No problem.
The red an black plugs are made by Hirschmann Germany
Actually the PM6306 was designed and built in Germany

Here are some pictures on one of my units from the inside.
Does your instrument have the DC Option? That one is very good for electrolytic capacitor measurements.

This one in the pictures has the RS232 option.
I have another one with the GPIB option but that one carries the official calibration seal and I do not want to open it.
It probably has a similar board, just GPIB instead of serial.

[BravoV]

HighVoltage, again, many thanks for the internal shots of your PM6306. I have not opened mine yet, for sure I will post it here, at least for comparison.

The red an black plugs are made by Hirschmann Germany

Really ? I did briefly check at Hirschmann web, still can't find it. Please take few shots of the plugs at various angles, this plugs will be the 1st priority for the DIY adapter as currently I can not play with this beast as I never have 2mm mini banana jack, let alone the Lemo connector. ... sigh ...

Does your instrument have the DC Option? That one is very good for electrolytic capacitor measurements.

I'm not sure, but its nice to the PM9565 DC Option board. 

Even I don't have it, still I can use the External DC bias, did you ever use this feature ? Any tips or comments regarding this feature ?

This one in the pictures has the RS232 option.

Hmm... why there are different caps (2 in the middle) in that row of 6 caps there ? Did you fix it ?

I will post the photo on my RS232 option board once I opened it for sure.

[plesa]

If you cannot find the 2mm bannana I suppose that RC batterry pack connectors will fit. To be sure measure diameter, because there are some with 2.6mm diameter
http://www.amazon.com/Neewer-Plated-Bullet-Banana-Connectors/dp/B00GTN72ZS
or
http://uk.farnell.com/hirschmann-testmeasurement/973509101/plug-2mm-red-pk5-vq/dp/497216

[HighVoltage]

Here is my collection of 2mm plugs from Hirschmann.
MC (Multi Contact) is also making some 2 mm banana plugs, but they are much more expensive.
These Hirschmann plugs are easily available in Germany, even on eBay

2 mm Hirschmann Male plug
http://www.ebay.de/itm/2-Hirschmann-Miniatur-Stecker-MST3-2mm-Bananenstecker-MST-3-rot-039675-/360774545131?hash=item53ffd6aeeb

2 mm Hirschmann Female connector
http://www.ebay.de/itm/2-Miniatur-Buchse-MBI-1-Hirschmann-2mm-Buchsen-Einbaubuchse-MBI1-schwarz-039682-/311319886252?pt=LH_DefaultDomain_77&hash=item487c1c79ac

This seller is shipping worldwide.

[HighVoltage]

For some every day use of the instrument, where I do not need the last bit of accuracy but a fast measurement and easy handling of the wires, I am using my self built test leads

2 mm Hirschmann connectors on one side
4 mm Banana Plugs by MC (gold plated) on the other side.
LiFY highly flexible test lead cables

I have found, that this is my most comfortable way of using the instrument on an every day basis for quick and easy measurements. The 2 mm banana plugs plug directly in to the PM6306.
 
 

[BravoV]

Thanks for pointing places for sourcing the mini banana accessories, bookmarked. Also I am already in the middle of procuring them, it may take a while to arrive. Not Hirschmann quality though. 

HighVoltage, the DIY test leads looks great  . Although these will not compete with the factory made ones, just curious, have you ever verified how much different between them in term of accuracy ?

[TheAmmoniacal]

BravoV, $301 USD + shipping? Deal?

[BravoV]

BravoV, $301 USD + shipping? Deal?

LOL ... sorry , I'm falling in love with this beast now. 

Even I'm going to sell it, at least not now before I'm satisfied with quality time spent with it. 

[BravoV]

Ok, opened it, abit dusty inside, although not too horrible, going to clean it carefully there. The latest date code I could find is year 1998.

So here we are, few shots of it's gut, hope you will find it useful.

The top view, personally I love this kind of design that put all high voltage stuff at the rear, seen here it uses the green long stick at most right, connected to the front on/off push button.




View from the rear.




View from the front.




The firmware, Ver 1.4, its older version compared to HighVoltage's which is at 1.6.
I am kind of worry seing the battery, I'm guessing its for the volatile calibration data ? Sigh ... another "adventurous" battery replacement task again ? For sure I can not afford to send it to calibration lab, after looking at my buying price.




Analog front end section, no DC Option board like the one owned by HighVoltage. ... sigh ...




The RS-232 option board, looks like it provides full isolation for the serial connections.




An interesting big 22nF cap popped out in the middle analog section, wonder what is that for ? They could use a smd cap instead that big cap right ?




A bodge cap, I guess its earlier version.

[TheAmmoniacal]

Great set of pictures, thanks! I like some gear porn in the morning.

[HighVoltage]

Very nice set of pictures indeed.
So, when you turn it on, you see FW v 1.4 on the screen, right?