EM84 magic eye does not close on FM on Philips B4X92A

I'm restoring a Philips B4X92A. After recapping and replacement of off-spec resistors, both FM and AM bands are working normally, with a loud and clear sound. The magic eye is working for AM reception, but does not move when the FM band is selected. The EM84 tube has already been replaced, as well as the 470k resistor R5 that connects pins 6 to 9/7. It lights up and appears fully open, not reacting to FM reception.

When FM is selected, I measure low (or zero) voltages in parts of the detector and AGC circuits (see diagram below). I have already measured all resistors (all within 10% in spec) and replaced a number of capacitors (the ones marked with green circles on the diagram).

There is a point where I measure 0 VDC in FM (see diagram), but I have non-zero voltages on AM (I did not write down the measurement, but can check again). It's a bit difficult for me to understand the switch (SK) diagrams, but based on continuity tests, it seems that the circuit indicated by the red line is the only one that is added when we press the FM bandswitch.

I have been in the hobby for a couple of years and so far could work my way out of most issues, but it's been several months that I'm stuck on this particular bug. Any advice is highly appreciated.

Thank you in advance!

[image]

Reageer op dit bericht

1201 keer gelezen

Hi Taciano

that’s a very clear explanation of your problem.

Since you’ve tried everything else only SK4 and the EABC80 remain as suspects.

With the unpowered radio on FM you should have continuity from pin 1 of the EM84 and the point in between R’s 18 and 21 (or 5k8 between EM84 pin 1 and EABC80 pin 7). If not the section 5/6/7 of SK4 is faulty.

If this section of SK4 seems to be working fine, the only other explanation I can think of is a bad contact or a dry solder joint on pins 1,2,3 and/or 7 of the EABC80, or a broken EABC80 that needs replacing.

If you don't have a EABC80 (yet) you could try to temporarilly clip in two solid state diodes like 1N4148 between pins 1 and 7, and between 2 and 3 (kathodes to 3 and 7) and see whether the EM84 works. If it does, replace the EABC80.

Succes! Kind regards, EJ

--
Dingen moeilijk maken is makkelijk, dingen makkelijk maken is daarentegen moeilijk.

Reageer op dit bericht

1031 keer gelezen

Hi EJ,

Thank you so much for your advice.

I have already tried a different EABC80 known to be good (purchased on Marktplaats from a supplier with access to a tube tester). Measurements still stay the same.

There is continuity between R18/R21 and pin 1 of the EM84 (almost zero ohms). The funny thing is that between the junction R11/C30 and the junction R18/R21 there is a resistance of 7.26k. I don't know where that comes from, and I suspect it comes from SK4. I have already sprayed contact cleaner inside SK4 (via a small opening, see picture below) and operated the switch vigorously many times, but resistance is always precisely 7.26k. I wouldn't expect dirty contacts be so consistent.

The problem is that it is very difficult to get access to switches (see picture) without major disassembly, which may be beyond my skill.

Any further suggestions?

[image]

Here's the original diagram again:
[image]

Reageer op dit bericht

837 keer gelezen

Just follow DC resistance path.
Its easy, just a capasitor is a break-line(no DC path)
A coil is zero ohms.

There should not be a DC-path between R11 and R18-22

--
"I haven't even turned this on to see if it works because as long as I don't it might"

Reageer op dit bericht

845 keer gelezen

Hi Maurice,

Thank you for your answer.

Testing with the radio off, when the FM button is pressed (which activates both SK4 and SK5), there is a DC path between Em84-pin1, R11, and R18/R21. Are you suggesting that there is a coupling capacitor somewhere that should be blocking DC between R11 and R18/R21, but may be shorted/leaky?

I imagine that SK4 should either connect terminals SK4-7 to SK-6 or SK4-6 to SK4-5, but never the three of them. Is this what makes you say that there should not be a DC-path between R11 and R18-22?

Do you know how should I read the SK diagrams? The bold vertical lines, do they indicate connections when the switch is ON or OFF? When the switch changes state, would the bold vertical lines move up (for example, changing the connection from 5-6 to 6-7 on SK4)? Being able to read these switch diagrams would help understanding the intended DC path.

I hope these questions aren't too silly. Thanks again!

(once again link to the original diagram for ease of reading)
[image]

Reageer op dit bericht

828 keer gelezen

If there really is a DC path between R11-C30 and R18-R21 and pin 1 of the EM, something is wrong, and you will get no moving EM.
The cause can be that SK 4 contacts both AM neg line and FM neg line in the FM position, so a short in the switch.

So recheck measurement in the FM position, the voltages on
R18-R21
R11-C30
pin 1 EM,

What is the voltage on R 21 and R 18 or over the elco in the FM detector.

Reageer op dit bericht

774 keer gelezen

Hi Ruud,

In order to check if there is a between R11-C30 and R18-R21 and EM84pin1, I've unsoldered the connection between R11-C30 and pin 5 of SK4. These are the outcomes:

1. No continuity between R11-C30 and (R18-R21 or EM84pin1)
2. When I turn the radio on with FM, I still have the same result (radio works, but EM84 does not move). Voltage at EM84 pin1 is still around 0.3VDC.

So that seems to rule out that the circuit at R11-C30 has anything to do with the problem.

These are the voltage measurements you suggested (with R11-C30 disconnected), with a strong FM station:

R18-R21: -0.3VDC
R11-C30: -0.3VDC
pin 1 EM: -0.3VDC
across C40: 0.010VDC
R18-C40: 0VDC
R21-C40: -0.004VDC

When I re-solder R11-C30 to SK4 pin 5, all the measurements are exactly the same with a single exception:
R11-C30: 0VDC (instead of -0.3VDC)

It seems the problem is somewhere in the FM detector/discriminator circuit. I have already replaced C41, C40 (twice), C38, EABC80 and aligned S21 and S22/S23. R37 and R16 measure correct. R18 and R21 measure about 10% their expected values, but that should be fine. Is there any chance that either a leaky C36 or C39 could be causing this problem?

Reageer op dit bericht

716 keer gelezen

Take the wires of the switchbock and see if you still measure that DC path.
Than examen the DC path of both circuits if you can find a leaky path to GND.

Indeed the bold blocks are the moving contacts.
They shift 1 position and all blocks mounted on 1 shaft or moving pertinax.

Similar to be seen on this bunch of switching parts.

Mostly right a single row contact with 2 times switching contacts.
Just left of that one a 3 row moving contact block.
Left of that one again the moving part.
Total on the left the stationairy parts of the switch block connecting the wires.
Shown at the sliding contact face.
[image]

--
"I haven't even turned this on to see if it works because as long as I don't it might"

Reageer op dit bericht

760 keer gelezen

Hi Maurice,

Thanks again for the detailed answer. I think I understand now how the switches work, and made a diagram that represents the state when the FM button is pressed (both SK4 and SK5 activated), see below.

Meanwhile, I've unsoldered the connection between R11-C30 and pin 5 of SK4, in order to chase any rogue DC-path, with this outcome:

1. Now there is no continuity between R11-C30 and (R18-R21 or EM84pin1)
2. When I turn the radio on with FM selected, I still have the same result (radio works, but EM84 does not move). Voltage at EM84 pin1 is still around 0.3VDC.

So that seems to rule out that a DC-path from EM84 pin 1 to R11-C30 causes the problem...

[image]

Reageer op dit bericht

696 keer gelezen

Hi Taciano

The funny thing is that between the junction R11/C30 and the junction R18/R21 there is a resistance of 7.26k. I don't know where that comes from, and I suspect it comes from SK4.

The 7k26 you’re measuring between the junction of R18/R21 might just be R21 being a bit (7%) high: on FM the junction of R11/C30 is connected to chassis ground via contacts 5/20 and 21 of SK4. So no malfunction there. Since the resistance you’re measuring is higher then the expected value and consistent there are no unintentional parallel contacts made wich would result in a lower value.

As for the reason why the magic eye is not working on FM I’m still puzzling. :-(

kind regards, EJ

--
Dingen moeilijk maken is makkelijk, dingen makkelijk maken is daarentegen moeilijk.

Reageer op dit bericht

806 keer gelezen

Its very straight forward.

Measure the voltage across C40, there must be a DC voltage.
Of course when receiving a station.
R18-21 to ratio voltage level EM84, 27//6,8, app 4::1
So if C40 voltage is 40Vdc - EM81 is 10Vdc. app.

Rectifier EM84 voltage is B4(EABC80) 3-2 diode.

AC on pin 3-2 diode rectified diode B4 and smooth by C40.
C41 is HF decopling.

Rule out , investigate, Cold /Broken Solderings.
Check from pin to pin, zero Ohm connections.

Radio -OFF.
Cut C40
Switch to FM:

Path B4-2 to B3-1 is 27k Ohm
Path B4-7 to B3-1 is 6k8 Ohm

Path B4 2<->7 is 27k +6k8 = 33k8
Don't bother some 10% deviation.

Still think B4 EABC80 must be faulty.
Take off both wires B4 2 and 7
Replace diode B4 with f.i. 1N4004 or simulair.

In that way you rule out B4.

Lets see.

--
"I haven't even turned this on to see if it works because as long as I don't it might"

Reageer op dit bericht

800 keer gelezen

The voltage across C40 is 0, says the OP
Unknown is if this is when receiving a station.
If this is the case the fault is in the FM detector. Forget SK 4 and the EM.

Did you change the EABC ?

Strange that you do hear sound on FM.

Reageer op dit bericht

662 keer gelezen

Mist that , 0V C40

Indeed that should be the start of investigation.

--
"I haven't even turned this on to see if it works because as long as I don't it might"

Reageer op dit bericht

644 keer gelezen

RSS Feed van deze thread

Nederlands Forum over Oude Radio's

EM84 magic eye does not close on FM on Philips B4X92A (Techniek Radio/TV)