Philips PM3055 scope repair

[cheeseit]

Scopes are addictive..  Less than a year ago I finally bought my first scope and a few days ago my fifth arrived, a 60MHz PM3055. I believe that it was a pretty good deal at €33 with two Testec LF-series 150MHz probes. It is working but a few things needs to be fixed and it needs a cleaning. I plan to clean the somewhat yellowed buttons in peroxide too.

Service manual here.

I found this thread among others and am using a few of user vtl's images but the image below is my scope. Apparently there are a few versions of this model, some 50MHz only, and I think I saw or read somewhere that the one I have is the gray PM3055/01 (not entirely sure about the /01), which is 60MHz, with Philips written on the screen bezel and Fluke on a sticker on the top cover and on labels in the back. Date codes on IC's are from '94.



The LCD backlight bulb and graticule illumination bulb are both broken and, worse, the power switch does not latch. When I got it was stuck in the powered-on position and once I took the top cover of I saw why; the extension shaft for the switch is held back by a twist tie tied around a capacitor! What a bodge. I hope that I am able to fix the latching mechanism in the switch because I'm sure finding a replacement will be difficult.



I really like the mechanical engineering in this scope, it seems to be made for servicing with clips holding the boards down and stands that hold them once they are out. (see vtl's thread for more) The front unit with the 8052 processor and LCD is only held by two screws and has only a single flatcable connection. Once out the processor board clips out and so does the LCD.







One thing that is a bit worrisome is a high pitched whine from the power supply on power up that slowly fades away over a couple of minutes. I still haven't taken the power supply board out but it is coming from the input section that has a few of those explosion-waiting-to-happen clear epoxy Rifa caps - could they be the culprits? Anyway, they are more than 20 years old so I plan to replace them just in case.
Edit: the input section is right next to the power socket that has an EMI filter built-in, could that be the source of the noise?



I've found a suitable bulb for the graticule illumination locally but still need to find a source for the backlight bulb. Backlight must be 12V since that is what is sourced to the LCD unit (refer to SM page 109-114) but I'm not sure of the current rating - any ideas?

[tronde]

If you can't fix the switch, start here
http://www.elv.de/schadow-netzschalter.html

Spend some time with the datasheet they link to.

Most likely you will find something that will fit, but maybe you must transfer the mechanical support from the old switch.

[mij59]

Looks like a fairly common mains switch, could be this one www.conrad.nl/nl/potentiometer-service-gmbh-typ2-netschakelaar-drukschakelaar-250-vac-25-a-2x-uitaan-vergrendelend-1-stuks-708817.html
The noisy power supply could be caused by bad cap's , check the capacitors on the secondary side of the power supply.
Backlight bulbs are so retro, why not use white led's ? 

[rf+tech]

Hi cheeseit,

Nice scope and yes, a good price.

I picked up a PM3335 three years ago for $20US, with multiple issues. Including the power/mains switch that would not latch on. The culprit was a tiny plastic "stud" about 0.5mm diameter that was broken off of the plastic cam follower. This piece toggles through a maze-like path molded into the actuating plunger. Without the stud, there was no toggle action and no latching.

On this side of the pond, I was not able to find a replacement switch. I managed to repair the part by first solvent welding the stud in place, then fixtured the part in a milling machine, located the center, drilled a hole through the stud and base and finally pressed in a pin turned from brass. The stud is not round, but closer to a square cross section with rounded edges and required a 10x magnifier to find the proper orientation prior to solvent welding. I own both a Sherline mill and Sherline lathe - antique clock restoration is another hobby - so while this kind of repair may seem extreme, it was not much of a challenge.

Perhaps you will have better luck finding a replacement switch, since the 3055 and switch were both made on your side of the pond.

In the interest of a long service life, I replaced every aluminum electrolytic in the power supply right away. There is a brief, 1 second whistle/whine at turn-on that rises in frequency and vanishes. No extended whine as you are experiencing. Your photo shows the same power supply as used in the 3335 - it is a DC-DC converter and the whine comes from the transformer, just above the mains switch. This is a sign the transformer is struggling under load, and the culprit will likely be electrolytics.

The PM3335 CRT scale illumination used a 24 Volt midget flange base CM8176 lamp. The new lamp was not nearly bright enough for my liking, so I retrofitted a Luxeon 2700K 5630 LED mounted on a brass "bullet" heatsink pressed into a CM midget flange base, which fits the original socket. The LED is sourced by a new variable constant current supply that is controlled by a zero to +10 Volt signal from the original front panel scale illumination control.

Funny how I remember all of this detail, yet my memory fails me about the LCD backlight. I think I replaced this lamp with another Luxeon 2700K 5630 LED, probably running at less than 100 mA, to reduce heat dissipation. The warm white 2700K LEDs look so much like incandescent.

BTW, does the 3055 use a pair of AA cells to back up control settings and store waveforms? The original Philips branded AA cells were still in the 3335, the corrosive vapors from the leakage had eaten through copper tracks in five separate places on the logic board. Locating those took some serious effort.

RF+ Tech

[Shock]

Replace that X2 cap you're talking about, they can catch on fire under certain circumstances.

[cheeseit]

Well, it's been a year and a half but figured that I ought to update this thread since I dug the scope up a couple of weeks ago and finally did the repair. I also want to say thanks to rf+tech and mij59!



mij59: Great, that was exactly what I needed. Finding a replacement wasn't that hard after all, I just had to click the link! Fixing the latching mechanism did not pan out so I ordered one from Conrad. Thanks buddy!

Shock: There appears to be a few versions of the PSU and mine turned out not to have Rifa smoke bombs. The last image of the PSU board in my post was borrowed from another user and is a slightly different version than mine. I used that before I had taken a look at my board.

rf+tech: I took the switch apart but didn't see anything broken or missing. Sure sounds like impressive machining you did on your switch!

You were right regarding the electrolytics and the whine was gone after replacing almost every one of them, and a bit of noise that were in the trace disappeared too. Mine doesn't have that brief whistle/whine you described and I wasn't even sure that it powered up the first time until I looked at the front and saw the LCD was lit.

I started by lifting legs of a few caps and measured high ESD and decided to just replace the whole lot. Once desoldered I then measured all of them and almost all were bad, the exception being the the two high voltage 68µF cans and the two metal can 6800µF/10V axials mounted as radials. ESD measured in the hundreds of Ohms on some and a few were in kilo Ohms, and they had lost up to 90%+ capacity. At that point I got a bit impressed that it did work despite those caps, apart from the whine, and would guess that it must be a pretty good design since it tolerates such out of spec caps.

I ended using a bulb for the graticule illumination but used three LED's for the LCD lighting. I like the result and prefer the cold white for the LCD instead of the incandescent used originally.

The plastic in front of the LCD was warped and had two annoying drops of what must have been super glue. I replaced that with a cutout from a random but scratch-free plastic clamshell pack. Unfortunately the blue CRT filter has a few drops of super glue too but I can't seem to find any blue 2 mm perspex, only the standard 3 mm that everyone has. If anyone knows of a source please share, I could use some for other instruments as well.

I found something odd (to me) that I'd like an explanation to. The following image shows a series of 110 Ohm (0.6 W I think) resistors in the finay Y amp circuit that get quite hot, judging by the discolored PCB. Why this string of resistors, that apparently aren't big enough to handle the power, instead of say two 330 Ohm? I'd guess this is a possible future point of failure?

[tautech]

Thanks for sharing as I've got a 3070 (100 MHz) that needs an eyeball too.

I'd first check those 110 resistors for drift and then maybe replace them with a pair of 1W 330's but stood off the PCB some in there's room (height) to allow for it.
Obviously the designers knew there was some heat issues to address hence the holes in the PCB.
But all close together and hard on the PCB like the 110's are is not good for helping to dissipate the heat.

[Shock]

Not sure what they were going for there, just a voltage divider and spread the heat? You could also put a higher rated resistor in and alternate them between the top and bottom of the PCB with slightly longer leads and fan them out.

They appear to have dry crusty joints as well and the PCB at some point is going conductive. It seems marginal as otherwise they wouldn't have put the holes there.

If you wanted to be thorough measure resistance, voltage, current and temp, calculate power and look at the subsequent circuit to see if it's got bad caps etc that could be adding to the current draw.

[cheeseit]

I thought those joints looked dry and the board a bit too baked and was debating whether to take the board out to check and maybe replace them but decided not to. I'd hate to unnecessarily stress the tube by taking the neck board off. I'll probably do it though, just out of curiosity.
Sticking them to the bottom could maybe provide a bit more cooling I'd think but alternating between sides is a good idea. Both would be a bit of an eyesore though.