PM2525 Multimeter Probe

[MrAureliusR]

Just curious, on the front of the Phillips/Fluke PM2525 there's a proprietary probe connector, and without this probe you can't do 4-wire resistance measurements, which is stupid I know. I'm trying to find out more info about it, to perhaps hack my own or find an original probe. I've been googling and I can't seem to find any info about it, the service manual only mentions that the part number is PM9264/01. There's one diagram showing the port being used during calibration, and it mentions (I think) that some of the pins use I2C??

Kind of confused. I think there's 9 or 10 pins on the port. My PM2525 is all packed up in a box as I'm moving on Sunday, but once I'm settled I can crack open the case and take a closer look at the port. Anyone who owns this machine, your input is greatly appreciated! There seems to be little to no information online about this probe. Googling PM9264 brings up a radiator fan motor...

Attached is a picture of the port.

[alm]

The same probe is used in other Philips DMMs from that era. Some manuals contain the pinout. I got these two images from the PM2534 service manual.

[branadic]

I have a PM2534 and on this the connector is used for a temperatur probe, that was included when I received this multimeter.

[PA4TIM]

Indeed probably for a temperature probe, I have a Philips , but the PM9203 RF probe used the same plug (but for a much older meter PM2421)
My  PM2517E has the same plug for a tempprobe

[MrAureliusR]

I have a PM2534 and on this the connector is used for a temperatur probe, that was included when I received this multimeter.

It is used for both temperature probe and 4W resistance.

Some manuals contain the pinout. I got these two images from the PM2534 service manual.

Awesome, thanks! I'll try this configuration when I get the chance. I have a feeling it's probably as simple as that. I wonder what the I2C connection does. I recently received a copy of Vincent Himpe's Mastering the I2C Bus and so I've been trying to locate all the different I2C peripherals and chips that I own. It'll be interesting to probe the I2C connection... I can see in the service manual it's used for calibration but I'm sure there's more functionality than just that. I desperately want a Bus Pirate for times like these!

I'll probably go to my local electronics store, buy a few 8-pin DINs, and then figure it out using the above charts and/or process of elimination (AKA guessing) until I get it right. Then I was thinking about soldering four banana binding posts to the leads of the DIN connector, and heat shrinking all the cabling to have a 4-lead plug for 4W resistance. I also own a temperature probe (came with my UNI-T meter) which also has banana plugs, so if I can figure it out I'll add two differently coloured banana jacks (yellow or something) and attach them to the proper connections for temperature. It's possible (and actually now that I think about it, quite likely) that the temp probe connection is shared with the resistance leads. So maybe I'll just colour code the 4 leads, standard red/black, and then blue/green for the temp ones or something.

Does that sound like a plan?

[alm]

Note that the temperature probe (at least for the manual I looked at) is a Pt100 resistive probe. That meter is not going to work with a type K thermocouple as is often shipped with handheld multimeters and cheap thermometers.

[MrAureliusR]

Note that the temperature probe (at least for the manual I looked at) is a Pt100 resistive probe. That meter is not going to work with a type K thermocouple as is often shipped with handheld multimeters and cheap thermometers.

Yeah, I figured as much. From what I've read, it's not terribly difficult to build your own PT100 resistive probe though?

[alm]

I've never built one, I suspect it involves a very thin platina wire. I would just buy one and fit the correct connector. For best accuracy you should use 4-wire connections, although you can also short the sense and source terminals at the connector at run two wires to the probe.

[MrAureliusR]

Sorry to bump an old thread, but I've come back to needing 4-wire resistance measurements and this stupid proprietary plug is again a problem.  I still can't find any for sale on eBay or anywhere online -- does anyone who mentioned owning a similar model happen to have a probe they could sell?

I'm still thinking about just hacking one, but I'd rather try and get a real one first.

Thanks!!

[robrenz]

I am pretty sure that is just a standard 8 pin DIN connector. Readily available just about anywhere.

[MrAureliusR]

I am pretty sure that is just a standard 8 pin DIN connector. Readily available just about anywhere.

Yeah, I know it's an 8-pin horseshoe DIN, I meant the original Fluke/Phillips cable that would have come with the meters of that era. I really can't find them anywhere.

On a side note, I just did a teardown of this multimeter and filmed it, it'll be up on YouTube sometime soon. It's a pretty neat meter inside. I even was able to dump the EPROM...

I ended up doing a rather quick'n'dirty hack of the DIN Probe socket. I happened to have an exact connector lying around (however it's not really a cable connector, more like a male panel mount) that fit (almost) perfectly in between the front panel plate and the socket when plugged in. I soldered wires to the solder cups, and put it all back together so I can easily hack into the connector.

The above pinouts seem to be correct, and the I2C bus mentioned in the service manual is on pins 5 and 6. Pin 5 seems to be SCL and pin 6 is SDA. The clock is about 70kHz. I'm going to see what I can do with my bus pirate, just curious as it's constantly repeating the same bytes over and over.

Update: Well the Bus Pirate can't seem to make heads or tails of it. I can see the data with my scope but the BP just reads 0x00 ACK 0x00 ACK over and over. Which is clearly not what's going on.

[sychowaves]

hi How did it go with the probe I found it on ebay if you manage to buy it and solve the dilemma remember to share it !! I have the same problem with mine

eBay auction: #https://www.ebay.de/itm/Philips-PM9267-Data-Hold-Probe-Digital-Multimeter-30-VAC-42-VDC-mit-Gebrauchsan-/382141655335

[mapleLC]

I've never built one, I suspect it involves a very thin platina wire. I would just buy one and fit the correct connector. For best accuracy you should use 4-wire connections, although you can also short the sense and source terminals at the connector at run two wires to the probe.

Note that the temperature probe (at least for the manual I looked at) is a Pt100 resistive probe. That meter is not going to work with a type K thermocouple as is often shipped with handheld multimeters and cheap thermometers.

Old thread

I want to buy a PT100 probe but I am only seeing 3 wire options. 

Is there a 4 wire or do I jumper something?  I would like to get the temp probe working.

[edpalmer42]

There are a few 4-wire PT100 probes on ebay.  They're rare and tend to be expensive.  I gave up looking for a reasonably-priced probe and settled for a bare sensor for my PM2534 which I believe will work with your PM2525.

I was pleased to find this one:  https://www.ebay.com/itm/223475090836 , but it looks like it's no longer available on ebay.  It looks like you could buy directly from PCSensor.com .  The model number for that sensor is PT100A4W.  You might find something similar on Aliexpress.  Not only do you need a PT100 sensor, but you need one with an alpha value (aka temperature coefficient) of 3850 (aka 0.00385).  There are lots of cheap PT100 sensors, but they typically either don't state the alpha value or they use another value.  The Philips multimeters won't read correctly if you use anything other than 3850.

I calibrated my probe with an ice water slurry and measured 0.0C.  Boiling water was a little trickier to calibrate but the error was less than 2C so I was satisfied.

Don't use a probe like this to measure live equipment.  Not only is the case made of bare stainless steel, but there's no guarantee that the case is isolated from the leads and no voltage rating for the isolation between the case and leads.

FYI, the description for the connector is "8 pin male 262 degrees".  The '262' is important.  It refers to the angle between the pins.  The more common 8 pin connector is 270 degrees.  'Horseshoe' or 'U-shaped' is also often used in the description because the pins are arranged in that shape.  In the 'wrong' connector, the pins are arranged more in a circle.

[mapleLC]

There are a few 4-wire PT100 probes on ebay.  They're rare and tend to be expensive.  I gave up looking for a reasonably-priced probe and settled for a bare sensor for my PM2534 which I believe will work with your PM2525.

I was pleased to find this one:  https://www.ebay.com/itm/223475090836 , but it looks like it's no longer available on ebay.  It looks like you could buy directly from PCSensor.com .  The model number for that sensor is PT100A4W.  You might find something similar on Aliexpress.  Not only do you need a PT100 sensor, but you need one with an alpha value (aka temperature coefficient) of 3850 (aka 0.00385).  There are lots of cheap PT100 sensors, but they typically either don't state the alpha value or they use another value.  The Philips multimeters won't read correctly if you use anything other than 3850.

I calibrated my probe with an ice water slurry and measured 0.0C.  Boiling water was a little trickier to calibrate but the error was less than 2C so I was satisfied.

Don't use a probe like this to measure live equipment.  Not only is the case made of bare stainless steel, but there's no guarantee that the case is isolated from the leads and no voltage rating for the isolation between the case and leads.

FYI, the description for the connector is "8 pin male 262 degrees".  The '262' is important.  It refers to the angle between the pins.  The more common 8 pin connector is 270 degrees.  'Horseshoe' or 'U-shaped' is also often used in the description because the pins are arranged in that shape.  In the 'wrong' connector, the pins are arranged more in a circle.

Thank you.  And no thank you to whoever above said that a standard 8 pin din would work, it didn't.  Your description is correct.

Perhaps a $4 temp sensor from Xhina is easier after all?

I bought the following sensor, which now I have not opened because it must be a certain alpha.  I dont see that on this probe anywhere.  And I thought these things were standardized.  The older I get, the more money I seem to be wasting on hyper-details missed.  No wonder people give up and pray to the god of Xamazon.

Does anyone know if this probe is reliable in that regard?

The god of Xamaon says its so:

- Tolerance of PT100 Ω (Alpha = 0.003850 @ 0°C) , Class A : ± 0.06 Ω OR ± 0.15°C

[edpalmer42]

I bought the following sensor, which now I have not opened because it must be a certain alpha.  I dont see that on this probe anywhere.  And I thought these things were standardized.  The older I get, the more money I seem to be wasting on hyper-details missed.  No wonder people give up and pray to the god of Xamazon.

Does anyone know if this probe is reliable in that regard?

The god of Xamaon says its so:

- Tolerance of PT100 Ω (Alpha = 0.003850 @ 0°C) , Class A : ± 0.06 Ω OR ± 0.15°C

It's rather confusing.  The statement on Amazon is using all the right words.  Reading that, you'd think it's perfect.  But on the not-so-perfect prime website, they mix up the ideas of thermocouple and RTD (i.e. PT100) sensors and don't give that nice tolerance statement.  They actually say that this probe is compatible with "other brands thermometers that use a K-type probe".  I've never seen a 4-wire K-type sensor.  AFAIK, all K-type sensors are 2-wire.

So the bottom line is ......  .  I have no flippin' idea!

You could contact the Amazon seller and ask where he got that tolerance statement from - since it isn't on the manufacturer's website.  If he can give you a source - it might be okay.

[mapleLC]

I bought the following sensor, which now I have not opened because it must be a certain alpha.  I dont see that on this probe anywhere.  And I thought these things were standardized.  The older I get, the more money I seem to be wasting on hyper-details missed.  No wonder people give up and pray to the god of Xamazon.

Does anyone know if this probe is reliable in that regard?

The god of Xamaon says its so:

- Tolerance of PT100 Ω (Alpha = 0.003850 @ 0°C) , Class A : ± 0.06 Ω OR ± 0.15°C

It's rather confusing.  The statement on Amazon is using all the right words.  Reading that, you'd think it's perfect.  But on the not-so-perfect prime website, they mix up the ideas of thermocouple and RTD (i.e. PT100) sensors and don't give that nice tolerance statement.  They actually say that this probe is compatible with "other brands thermometers that use a K-type probe".  I've never seen a 4-wire K-type sensor.  AFAIK, all K-type sensors are 2-wire.

So the bottom line is ......  .  I have no flippin' idea!

You could contact the Amazon seller and ask where he got that tolerance statement from - since it isn't on the manufacturer's website.  If he can give you a source - it might be okay.

I guess the takeaway is that if I don't need to be too precise, this piece of China Maybe(TM) will suffice.

[mapleLC]

if anyone wants to create their own probe, here is the pin diagram.

The 4 wire probe has 2 red and 2 blue wires.  According the diagram below, should it be:

2 red wires > pin 1 and 3

2 blue wires > pin 3 and 4

Is that correct?

[edpalmer42]

That should be good.

As a double-check, the resistance between the red wires should be almost zero and the same between the blue wires.  A resistance check between red and blue will probably be around 108 - 110 ohms.  If your meter causes any heating of the probe, that value might slowly rise.

[mapleLC]

It worked fine.  At least with this probe, which is probably true for most if not all, its the same colors for each branch as shown in that photo.

Question, this thing measures temperature like old people f___. Is it really this slow or are there settings someone has dug into and knows?

[edpalmer42]

I'm not aware of any speed settings for the temperature range.  Changing the temperature of a physical object (i.e. the sensor) takes time.   

[mapleLC]

I'm not aware of any speed settings for the temperature range.  Changing the temperature of a physical object (i.e. the sensor) takes time.   

Those infrared's have made us impatient.