PM5324 Is the RF Attenuator Fried?

[MrVa1ve]

Recently, I picked up a Philips PM5324 for a reasonable price. Here's the manual in case the unit isn't familiar (https://doc.xdevs.com/doc/PHILIPS/PM...-%20Manual.pdf).

It's got a few issues that I'm busy working through. Most are easily fixable, but I'm finding it tricky to determine whether or not the variable attenuator is working as it should. It's a bronze coloured 75 ohm Preh brand and is capable of at least 60db of attenuation. The maximum output voltage across a 75 ohm load is only 50mV and a swivel counter-clockwise takes that to 50uV, which is a very small signal and not easy to measure considering the equipment at my disposal.

If it's rotated clockwise (maximum output voltage), it's fine. I get a healthy mV level signal. However, within a fraction of a turn counter-clockwise, the signal disappears from my scope. I figure the resolution of my scope isn't great, so I've also tried a micro-WattMeter I made recently with about 30db of amplification out front. Again, it picks up the signal to a small distance from maximum, but any further turns counter-clockwise and the WattMeter isn't picking anything up. My suspicion is the component is baked, but that's just a hunch and it's a sealed unit and relatively rare so I need to be sure.

So is there way to measure the signal other than with expensive or specialized equipment? I suppose I really only need to know that it's varying the signal strength at those strongly attenuated values for the time being.

[Julesomega]

Nice sig gen, I can't offer any specific advice, but if you want an RF power meter I suggest a AD8310 module from China, about 90dB in one range, prices start at 99p including a supply regulator

[buggies]

I would not expect to see anything on any of my scopes much below 5mV.
A power meter is not ideal at these levels e.g 5mV across 75 Ohms is what -
(open to correction) about 0.3 microwatts.
Personally I would just plug the output into the aerial socket of a decent commucations receiver and see if anything can be heard at smaller output settings.

[MrVa1ve]

Thanks George, clearly the Christmas has taken its toll when I missed that one as a quick test! I thought of it almost immediately after sending my initial post and I tried it out when I got back to the bench. Unfortunately, I don't have a decent communications receiver so I used a 50 ohm coax with croc endings to grab the aerial of a radio tuned to MW.

The signal is very strong with the attenuator at max and again, as with my other tests, if I turn the attenuator knob a tiny bit counter-clockwise I lose the signal. The cut-off point is very sharp I would say. I have an old valve radio I repaired recently, I might try the same test on it - feeding the test signal into the antenna port.

[MrVa1ve]

Originally Posted by Julesomega

Nice sig gen, I can't offer any specific advice, but if you want an RF power meter I suggest a AD8310 module from China, about 90dB in one range, prices start at 99p including a supply regulator

Hi Julian,

Actually, I've built a couple of watt meters recently, including a W7ZOI AD8307 variant by MrBungle (many thanks MrBungle!). The other one and one I mentioned earlier is a microWattMeter (also by Hayward, I think??). I made the uWattMeter in such a way that you can proceed it with one or more mmic type RF amplifiers (those cheap boards you get on ebay, etc).

They both seem to work quite well, but the problem is that I haven't got around to calibrating either of them yet. So, although the uwm registers the output power from the generator (and shows the sharp decline), I'm just not sure how sensitive this meter is yet. The same is true of the AD8307 based meter.

Seems like I need an accurate RF signal source to calibrate my meters. Catch 22!

[MrVa1ve]

That said, I'm making progress. I got the calibration circuit working again and damn accurate is is too! The last transistor before the meter, a BC238, had gone open circuit.

One other question for owners of this generator: how far does the smaller frequency tuning knob turn? With the unit I have the small knob only turns a short distance (maybe 5 or 10 degrees) before hitting a hard stop on either side. The manual talks of a slipping clutch, but I don't feel anything like this as I turn the small knob.

[stuarth]

The twin knobs for tuning are to give a “calibration” feature.

The outer knob drives the scale cursor, the smaller inner knob controls the tuning. The two knobs are locked together with a clutch which does not slip in normal operation, so the knobs turn as one. Adjusting the relative position of the two knobs changes (fractionally) the relationship between the tuning and the scale cursor, so the cursor and the tuned frequency can be set to exactly match at the calibration points (using the built in calibrator).

You could try testing the attenuator pot with an ohm meter. A sudden change at any position might indicate a damaged track.

Stuart

[John_BS]

You could try testing the attenuator pot with an ohm meter. A sudden change at any position might indicate a damaged track.

Seconded. It looks like a 75 ohm log pot wired with the wiper as input, so you should measure about 75 ohms to ground from the output socket. It's driven from a high source impedance. If it's faulty and assuming you can't find a replacement, I'd be inclined to fit a switched attenuator.


John

[m0cemdave]

The attenuator was a specialist item - a constant-impedance surface voltage divider type made by Preh.

Years ago I had a PM 5324 (wish I'd kept it) and I burned out the attenuator by accidentally injecting a signal at the wrong side of a blocking capacitor and applying around 150V DC. The symptoms were the same as yours.

Moral: always use an inline blocking capacitor with one of these generators.

It proved impossible to get a replacement but I was able to carefully take it apart by unpicking the seam around the middle of the can, and do a bodge repair by covering the burn, which was at one end of the track, with silver loaded conductive paint. The pot then worked OK and got the generator working again although there was a bit of a jump in level at the repair and the attenuation didn't quite follow the labelling on the panel.

I continued to search for a pot, scouring the radio rallies etc (no internet in those days) until one day a couple of years later when I was working in Vienna. There was a classic old-fashioned radio component shop (long gone) in one of the back streets and amazingly they had one in the window as part of a display of unusual components and vintage radios. * I persuaded them to sell it to me, I think for around 40 Schillings (which was £2 - their currency was easy for an old Brit to understand) and I fitted it to the generator on my return home a month or two later. I relate this tale to indicate how hard these things are to find...

These attenuators were available in 75, 60 and 50 ohm versions and two different sizes - 37mm (Type 110) and 47mm (Type 65) diameter. The input is to the radial (side) screened terminal and the output from the axial (end) screened terminal. I'm not sure if the larger size would fit the PM 5324.

A 50 ohm version of this attenuator is fitted to the Marconi TF2400/1, a heterodyne converter that was made for an ancient 10MHz frequency counter. These converters are occasionally found at rallies for a few pounds (only worth buying for parts) and, although you really need the 75 ohm type, one of them might suffice if you're desperate.

I suggest trying the repair as indicated above. Be sure to draw an exploded diagram of how it comes apart so that you can reassemble it with all the bits in the correct order.

Hoping this is helpful...


* This was the shop referred to here:
https://www.vintage-radio.net/forum/...&postcount=393

[orbanp1]

RFT, that former East German electronic conglomerate, made pots like that:
https://www.ebay.com/itm/NEW-NOS-75R...YAAOSwat5bXOce

Regards, Peter

[m0cemdave]

That looks like a direct equivalent, with identical construction other than the flange where the two halves are joined (which makes it a lot easier to take apart).
One would need to check the dimensions to ascertain if it would fit in the available space.

I wonder if Preh had manufacturing facilities in both halves of Germany before the Division, as did some other companies whose products ended up being made under different brands on each side of the border.

[MrVa1ve]

Thanks for the replies so far guys and to Stuart & John for the ohm meter suggestion. That worked a treat. It showed around 22 ohms measuring between input and wiper when it should have been less than 1 ohm and somewhere between the high megaohms and open elsewhere.

This attenuator has not had a happy life, as Dave is suggesting.

So, I removed it and was busy peeling around the seam when I realized I could drill a hole and try get some contact cleaner in. So, that's what I did and it's certainly improved things but the readings are inconsistent except for the fully clockwise (maximum) position, which is now reading < 1 ohm. I'll try measuring it again tomorrow, but if it's still flaky and inconsistent I may just cut my losses and run the signal directly to the RF output. I have a home-built step attentuator I can use after the output.

I'm sure they're as rare as hens teeth Dave, the best option might be to keep an eye out for a broken one as a parts mule (assuming it hasn't suffered the same fate!)

[CambridgeWorks]

If it is the PREH one as ebay link, it looks to be same as what (Pye) Telecom, Finglas used in their (100K/455/10.7/23.455) portable IF signal generators standard equipment issue. I "used" to have a NOS one, but not a clue where it could be, if indeed it is still around. I will try to have a quick look in the next couple of days. There are maybe 10 locations amongst dozens it could be in!! Please let me know if it is NOT that type, will save me looking.
Rob

[Wendymott]

Hi MrValve..... I have a surplus PREH 75R attenuator salvaged from a Marconi Sig gen..It has X 2 BNC plugs on it with aprox 6" on one leg and 12" on the other... PM if you want it...

[MrVa1ve]

Thanks for the offer Wendy. In a stroke of unusual serendipity, it seems I actually have a 75 ohm Preh attenuator from a previous salvage. It was a UnaOhm colour bar tester and I even have the schematic I pulled at the time.

I've attached the relevant snippet from the schematic showing the attenuator and it's specified as 75 ohm/110db. This guy tests very differently in the ohm meter test. Whereas the Philips attenuator (marking code 374) tests almost short fully clockwise between wiper and input, the UnaOhm attenuator (marking code 428) measures 56 ohms in the same position. It seems 374 has a setting that has no attenuation, which seems reasonable, it allows the full signal to pass from the signal generator. I'm no expert, but is it possible the 428 has a minimum attenuation?

Anyway, I'm going to wire up the 428 and see how I get on. One things for sure, the 374 attenuator is toast.

[karesz*]

You can have it as 50 ohm attenuator too _Texscan (&Trilythic too) CA50/50_ maybe would be better if you dont youse it in video/TV systems...
As I know Ruwido had an 50 Ohm version/type too and in East-bloc was produced a 75 Ohm "Preh copy" only in GDR.

[stuarth]

My PM5324 came with a 5k pot as the attenuator, obviously replaced at some time. I swapped it for a 100ohm pot, the nearest I had to 75ohm, but the output levels aren’t right. Now I know why!

I’d be very interested in your 75 ohm attenuator, Wendy, if MrValve doesn’t need it.

Stuart

[karesz*]

Hi Stuart and MrVa1ve,
OK, Here is an original Preh 75 Ohm attenuator
rgds, Karl

[Wendymott]

Hi Stuart.... Please PM me for details. Thanks

[MrVa1ve]

Originally Posted by stuarth

My PM5324 came with a 5k pot as the attenuator, obviously replaced at some time. I swapped it for a 100ohm pot, the nearest I had to 75ohm, but the output levels aren’t right. Now I know why!

I’d be very interested in your 75 ohm attenuator, Wendy, if MrValve doesn’t need it.

Stuart

Hi Stuart/Wendy,

Apologies for not responding sooner. Wendy, I won't need that attenuator and thank you very kindly for the offer.

I swapped in my salvaged attenuator this afternoon and it seems to work just fine. I was seeing quite a bit of noise and oscillation on my DSO, even with the signal terminated in 75 ohms. However, my Hameg 60MHz scope showed a nice clear signal, not even a hint of blurring to indicate oscillation. I'm starting to suspect that my cheapo Hantek DSO has noise issues at lower signal levels.

Anyway, that threw me a little bit and I spent quite a bit of time double checking connections, etc. I'm now thinking that the old attenuator should not have measured short when fully clockwise. That's probably a feature of whatever trauma it suffered.