Panaplex Mystery Meter!

A random Ebay find does not disappoint, and it is chuck full of high voltage neon goodness! In this video I see if I can get it working. Enjoy....
#franlab #neon #nixie
- Music by Fran Blanche -

[amtsgedicht]

That's awesome! I had the exact same Panaplex display meter. About 20 years ago, I found it in a dumpster and converted it into a thermometer for my studio using a PIC microcontroller. Unfortunately, the Panaplex eventually failed - it's probably not designed to be continuously on, but it served well as a thermometer for about 15 years. I still have an unused Panaplex meter in my box of unfinished projects. This time, I plan to incorporate a motion sensor of some sort to turn the display off when no one is around.

I have pictures of it - if you're interested, I can dig them up for you from my stash.

[WilliamDudley]

You will note that the displayed number keeps counting "down" (more negative). I think it's input bias currents charging a cap. Try shorting the input, or driving it from a power supply. I think you'll get reasonable readings that way.

[ntsecrets]

I have a Heathkit digital thermometer indoor outdoor that uses one like the left side display. It died on mine and I found what seemed to be the last one for sale online from Canada and paid like $100 for it. Works great now.

[null]

Love a good panaplex display, my datron multimeter from the 80s has one and thankfully it still looks great

[youtuuba]

A few thoughts:
- I suspect that the tabbed 3-terminal power device near the transformer will turn out to be a power Darlington transistor, acting as a series pass for the 5V logic power voltage regulator. There are enough discrete parts right next to it that would not likely be there for something like a 7805, but which WOULD be needed to bias the Darlington's base to about 6.2V (subtract the two 0.6V base-emitter drops in the Darlington, and you end up with 5V at the second emitter); only a couple resistors and a Zener are needed, plus the Darlington, to comprise a 5V regulator. This kind of circuit was VERY common in equipment during the narrow time frame when Panaplex displays were popular, and at the same time equipment of that narrow era did not usually use 3-terminal regulator ICs such as the 7805. That, and that style of heatsink tab was not found on any 7805 I have ever seen.
- Unless the product here is broken in some way, that drifty display is almost certainly due to the product's analog input(s) being unterminated, and this floating and picking up whatever electrostatic field is nearby. I very much doubt that it has anything to do with the product not being shielded or the PCB not having a ground plane. Look at ANY other DMM circuit of this era, and you will see no shielding and no ground plane; they are not needed. This product will probably work if Fran reverse engineers the Intersil 7117's inputs back to the card edge connector and either shorts the input(s) or applies an appropriate stable DC voltage to it/them.
- SOMETHING has to make the high DC voltage for the Panaplex displays, and I suspect it will be a secondary besides the one being used for the 5V logic power supply. There don't appear to be enough, and the right kind, of components on the PCB to make an oscillator and such necessary to boost a lower supply up to the necessary HV supply. The transformer only needs to have a HV winding plus a capacitor and a diode forming an "Villard Doubler" that would nicely supply just the kind of pulsating HV DC that Panaplex displays just love to work from.
- Because this product appears to be an example of a "universal" digital panel meter (and many companies use one very nearly identical to this for use in industrial plants, control panels, etc), I would expect to see pins on the card edge connector for signal ground, analog signal input, and decimal point position selection. Often, there might be several analog signal inputs, each for a different signal span, so the various connector points would subject the input signal to various degrees of attenuation. Because the Intersil digital meter ICs (7106, 7107, etc; and probably including this 7117) had TWO analog inputs (they were differential and the IC could take the ratio of the two signals and display it (times a multiplier); by using the two inputs in various combinations, the product could be very flexible in regard to what kind of signal could be accepted to drive the display.
- If the 7117 is anything like the 7106 & 7107, it internally generates a couple more voltages, which are useful for all sorts of clever circuit configurations, including being conditioned and then feeding one of the two analog inputs on the IC. There don't appear to be many discrete components on the PCB, so the 7117 might be doing almost all of the work conditioning the input signal.

[maccamcdermott]

Ooh, cool find! Such a lovely glow from these displays

[McTroyd]

It's the display half of a 2000-count multimeter. That's kinda cool. 200 mV swing would make it easier to use than an analog meter, I suppose.

[SeanBZA]

200mV says direct input to the 1G input impedance of the Intersil chip, so it drifts with no input connected as it effectively floats, as there is only a 100n capacitor and a 1M series resistor to the input. That slot in the board likely is there foradd on boards, that provide either different voltage ranges, or most likely a way to connect to a top board, which will have the ability to have a data read out of the display, likely using some logic, and an EPROM, that is fed the 7 segment display info, and where it then is programmed so 4 bits of the output are the BCD representation of the number, and the 4 digits are sequentially read out using another 2 bits of a counter that selects each digit in turn.

Was part of the Intersil application notes of the ICL7107 to provide a digital readout of the chip info, as well as displaying it, either with an external source of the digit selects, or with the DVM providing them, along with using the conversion complete of the chip to spurt out 4 nibbles in sequence, so you got the 3 full digits, the most significant digit and polarity all as the first digit. Was a way to get a stable ADC input to a MCU or minicomputer before the availability of low cost dedicated DVM chips, using the stable Intersil ratiometric measuring system. Lots of ATE systems were built on this, often using a few meters, and external preselectors to do range switching, AC conversion, and current shunts. Resistance a dedicated display as you had to do extra work to the chip to get resistance ranges, so the add on board was different, to do the range selection with reed relays, to select resistors for them.

[bradlevy]

The buffer/inverter logic chips are CMOS, so low power, and the ICL7117 pulls less than 2ma excluding load on the outputs. The DI-210 chips draw about 0.5 ma/channel on the input side. So 2 ma for the 7117, 12 ma for the 24 channels of the DI-210s, a couple of ma for the inverter chips not including their load which is the inputs of the DI-210s, already counted, is a total of only 15 or 16 ma needed from the logic supply. So you don't need much of a heat sink on voltage regulator for that.

[trainliker]

It's fun to look at a board that was layed out with tape as I once did a LOT of that. However, I was more fussy about the neatness. Some of the work here is slightly crude, but of course, that doesn't affect function in a board like this. They probably used "Bishop" brand pads and tape products. The tape can be curved, but not much as can be seen by the wide trace where they did it in sections. I can tell this was done with separate front and back arts with the curvable tape. They also had a scheme with blue and red tape that you put on the SAME sheet which insured perfect registration. The two final artworks were made on film with a camera using filters so that each image had all the pads (black being used) but one showing the red only and the other showing the blue only. But it was a mylar tape and not curvable at all so all of the corners had to be right angles or maybe little 45 degree angles. Also, if you had to make a change, it was VERY annoying because a trace you needed to work on typically had many other traces laid down OVER it that you had to remove all those and re-apply later. We didn't use that more than about once or twice and went back to black tape and multiple sheets using registration pins. When PC layout software became available, I think it took us about a millisecond to decide to get it.

[fanofoldfans]

Looks like the Heathkit digital clock I built in the late 1970's with a display like that. Cool

[gartnl]

Tape and rub-on stickers. I still have a sigarbox full of them, next to the box with memories of a passed youth. We were so proud we could make our own PCB's than.

[WRMD]

I believe that Panaplex were popular in older aircraft like an MD-90 and the Citation X. Steveo1Kineveo's previous TBM 850 had them in it.

[Jeroen]

The expansion connector could be for a BCD logging output.

[hopelessnerd]

Neat! I loved those Intersil chips. I still have an ICL7106 I was going to build something with. Doesn't look like that's gonna happen... I have an LED version that looks very similar to what you have.

[Capturing-Memories]

That's crazy to have high voltage pins on top of each other, imagine you pull the card out under load and the pins touch each other.

[dhpbear]

8:45 - A strikingly bad choice of adjacent pins, considering the rest of the design is superb!

[trainliker]

It looks like you need a few card edge connectors for testing. Yes, there are a vast number of configurations and spacings, but I suspect most of the old stuff you look at might be the 3.96 mm spacing (if I remember that right) and you can usually use one that too long, or a double sided even if board has only single sided fingers.

[cosmicraysshotsintothelight]

That case looks like die cast magnesium to me. The big screw lugs on the back are meant to hold some force. The shot of the case shows the back wall, and it looks like metal as well. It is probably meant to have the read contacts in a closed loop setting. Being open leads and out of the case, it can pick up a lot. Especially considering the range it declares.

[grimninja]

i had like a dozen of these, and may still have one or 2 meters that are extermly similar to this one . i got them in a dump of all off spec e- junk i trash picked about 15 years ago.
didn't realize they were any thing special other then basic early vfds so i threw most of them out after holding on to them for 10 years ;( pretty sure a damaged nixe tube bench meter went with them too... may have 1 or 2 somewhere buried deep in storage tho.