Philips PM3217 X deflection problem

[high_vacuum_house]

Good evening,
I am having issues with the 3217 oscilloscope. The X deflection scan length is only around 5mm in the centre of the CRT and is also running right to left. I have been looking at this and also the X POS control or X10 magnification does very little. The timebase is running correctly with a 5.4V peak to peak sawtooth waveform to the base of V1414. All of the power supply voltages have been checked and all correct.

The X POS control does work correctly and is putting the correct voltage onto the base of V1413 at the CW/MID/AC positions of its travel.

I am getting the same 5.4V sawtooth waveform on the emitter of V1414 as on the base though I am only getting 960mV peak to peak signal on a +11.8V DC offset on the collector of V1414 though the manual states it should be 8.4V.

Both X plates are receiving 106V and 110V with little sawtooth waveform on them. The manual states that it should be approx +26V

I am getting a bit lost as to where to check next as DC coupled stages can cause much confusion.

Many thanks,
Christopher Capener

[Chris55000]

Hi!

Collections of d.c. coupled cascade transistors in oscilloscopes are confusing to fault find at first go, and they don't lend themselves to conventional techniques!

There 8V are three voltage supplies to this amplifier, +38V, +180V and -180V, the +38V is surprisingly important as the X drive preamplifier V1414 will not work correctly if this is low or missing!

I mention this point again as the +38V is obtained from small–signal BAW62 rectifiers rather than the BY206 power diodes of the other supply rails, in the course of fault–findinng this supply could have accidentally disappeared due to burn–out of the BAW62 diodes!

If you're happy the BAW62s are providing the +38V, the next step is to print out the section of the main board layout diagram covering the X Amplifier layouts, using the "Fit To Page" function to enlarge it, then carefully unsolder the following transistors from the board and check them in this order:–

V1423, V1426, V1417, V1418 and V1419.

I strongly recommend one of those little Chinese Colour–Screen Component Testers of the "LCR–T1" type for dealing with multi–transistor direct coupled circuits of this type – they take a lot of guesswork out of fault location!

If any of the transistors only give a reading of one diode or a low resistance indication between any lead(s), bin them and obtain replacements!

If you get good test results from all the six transistors, leave them out at this stage and go on to all the precision value resistors dotted throughout this circuit – Philips precision resistors of this type do have a habit of going high in value or open circuit, and a red and green pen is handy for marking them off as good or bad on a copy of the circuit diagram as the case may be!

These are all E24/E96 series values which you can obtain from RS/Farnell – they will supply individual hobbyists on a pre–paid basis!

The zener diodes V1418, V1421, V1424, V1427, V1428 &, V1431 can partially or completely short to substantially alter bias voltages or cut off transistors, but a simple voltage check across each of them against the circuit diagram should show up a duff one!

Really and truly, the method of removing the active devices, the transistors in this case, then checking the passives left in circuit, is the best way you're going to locate faults in complicated mesh–works of this type – faults in this type of circuit will cause shifts in operating points of the transistors to such an extent that normal fault–finding techniques of measuring voltages can be extremely confusing!

Chris Williams

[high_vacuum_house]

Good evening,
A little progress though there are some anomalies. the +38V rail is working fine and within its limits. I pulled V1406 and 1407 as I had replacements to hand. Both tested ok and gains were both matched on my Peak semiconductor tester.

There is a confusion which is there a potential divider connecting both of their bases which should generate -8V from the -12V line. This is made up with R1402 and R1414 which are 4.02K and 8.25K. In mine they are 2.74K and 2.86K producing -6.1V and not -8V as the manual states at the base of the 2 transistors. The resistors have been flow soldered at the factory and therefore haven't been changed at some time. Is there any benefit from changing the value of the potential divider?

There are a few other (official!) mods I can see around the X amplifier which don't fit on the circuit diagram including a 330 ohm resistor in each wire to the deflection plates and some extra capacitors.

Christopher Capener

[high_vacuum_house]

After replacing V1406 and V1407 the X deflection is back as it was with 5-6mm deflection. After further investigation it appears that V1426 and V1417 are not really turning on or drawing much current at all.

On V1426 the emitter resistor goes to the -180V rail via R1442 (1.15K) and there is virtually no voltage drop across it. Its collector should sit at -85V though it is currently sitting at -5.5V

On V1417 is drawing not much current either, the emitter resistor R1441 is almost the same voltage both sides with the collector currently sitting at +175V and not at +98V

Christopher Capener

[high_vacuum_house]

Good evening,
The output transistors are completely different from the manual.

V1417, manual shows BFT45 actual transistor fitted is BF470
V1418, manual shows BF238 actual transistor fitted is BF469
V1423, manual shows BFT45 actual transistor fitted is BF470
V1426, manual shows BF338 actual transistor fitted is BF471

Christopher Capener

[WME_bill]

Philips PM3217
Taken from Mod note:OSC167.
Alternative transistor types for Amplifier Unit PCB, because transistor BF338 is obsolete, component changes:
V1417 and V1423 into BF470 (from BFT45)
V1418 and V1426 into BF469 (from BF338)
R1437 and R1439 into 100E
C1407 into 1.8pf.
For replacement in older instruments, BF338 remains available.

So I suggest that what you have is correct.

Other Mod notes relate to new Y output amplifiers. From both the /02 version and again for the /05 version, the brown look version.
A revised power suppply from /04 version.
No changes apparent for the X output amplifier.
wme_bill

[high_vacuum_house]

Many thanks Bill. I have a few more transistors on order and will see what happens.

Christopher Capener

[high_vacuum_house]

Good evening,
I think I am getting somewhere with this problem now. I have found that V1426 (BF471)is faulty. There was only 0.5V drop across the 5.1V zener diode V1424 which supplies -175V to the transistors base. Cutting one leg of the diode proved that the zener diode was working correctly, therefore a possible emitter to base leak in the transistor pushing the base voltage higher.
Removing the transistor and testing on my Peak semiconductor tester shows it thinks it is acting like 2 diodes in series.

It should be a BF469 in its place which I have on order though it has got lost in the post at present!!

Christopher Capener

[WME_bill]

PM3217 X amplifier.
The transistor types you quote shows how much manufacturing techniques have improved since 1983, when your scope was made.
The original design used BF469, rated at 250v. The substitute in place was BF471. Identical to BF469 except that the voltage rating is 300v.
At one time the higher voltage rating was a separate selection, and probably more expensive.
Now the lower voltage selection is probably almost unavailable as all those coming off the production line are now consistently at the high voltage rating.
It is a standard video output type. A search around a scrap TV could produce three suitable replacements: BF338, BF258 etc.

But now my local tip will not let me strip a board from an old TV. It has to go for "recycling", which is a misnomer. It is crushed. Then treated with acid to remove gold etc, and the rest dumped or burnt. So another batch of IC, transistors and thermistors no longer made have been lost.

Until the latest EEC Regulations, as a "trader registered at the tip" and thus complying with Health and Safety rules, I could acquire old computers or TV to get rare replacement parts out.
What experiences do other members find at their local "recycling centre".
wme_bill

[AC/HL]

We've highlighted the subject of recycling/reclamation many times.
A current thread is this one: https://www.vintage-radio.net/forum/....php?p=1210964

[high_vacuum_house]

Still no further forward. I have replaced the BF470 and BF469 transistors V1423 and V1426 and the trace is hard off the CRT now. It is getting frustrating as there are so many different values of resistors around the amplifier. R1439 to the base of V1423 should be 30.1 ohms but is 1.665K ohms. R1437 which is in the base of V1418 is 100 ohms when it should be 33.1 ohms. Could I replace R1439 with a 100 ohm resistor?

Christopher Capener

[high_vacuum_house]

Good evening,
I have made a bit of headway and found that V1422 had been replaced and wired up incorrectly. Upon testing it was ok and wired up correctly I now have the Plate X1 working correctly. The scan is now going the right way as well!!.

There is now deflection though it is stuck on the left hand side of the CRT and will not travel past the centre even with the shift control hard clockwise. The Plate X2 amplifier is completely dead with no signal on the output. The collector of V1417 is at +175V.
Transistors V1417 and V1418 have been pulled and test ok. Zener diode V1416 is working correctly and dropping 5.1 volts. I have not tested V1419 (BSX20) yet.

Christopher Capener

[high_vacuum_house]

Good evening,
I have finally fixed it!!

The last faulty component was the v1419 BSX20. It tested as 2 diodes on my semiconductor analyser. I found some at the Harwell rally this morning and replacement restored the correct X amplifier operation. Therefore all 6 final X amplifier transistors had failed and all have been replaced.

All that was left to do was to check and recalibrate the X timebase using my digital oscilloscope as a guide, give the ASTIG and GEOM pots a tweak, replace a popped and rather stinky RIFA cap across the mains and reassemble the case.

Many thanks for the help and guidance. I was determined to see this through!!

Christopher Capener