BRT 400E - fault finding help needed

[radiogm7nva]

Hi all. This will be my first post, so give me a little space lol!

I have a GEC BRT 400 E (203280) (Inst 333)

The radio works very well, but has one annoying intermittent issue.

Fault: The signal drops in level (also noticeable on meter) and if left, will spring back after a period of time. However, the quickest way to clear the fault is, to rotate the selectivity control switch back to XTAL position 1 then, work back up again. 90% of the time, the volume will spring back to life with full gain at (IF) position 4, but not often before that position. I thought that it may be this switch but, well read on.

If I try the AVC on/off switch during the fault, the AVC will not operate. As soon as the fault clears it is fine again. Operating this switch or any other will not clear the fault. Switching the radio off and counting to 5 will also clear the fault.

If have found the following, using 2 scopes.

While in fault, there is very low signal in the whole IF section right back to the first Selectivity switch wafer and, on the leads to and from the XTAL area. So, I have assumed that the fault is not in the selectivity switch or that section. (I am not sure how to check the LO box).

On the other side of the fault, the signal can be traced in good health right up to V3 (Detector). I haven’t been able to check V3 output as yet.

So the fault seems to lie before V3, the detector, and the first section of the Selectivity Switch.

I ham now stuck. Where it goes after V3 on its way to the selectivity switch is a mystery at the moment.

Has anybody seen or heard of this fault. Any other info or ideas would be welcome.

I am not too experienced in fault finding so, please, keep it low level lol!

[cmjones01]

My BRT400 doesn't work, but at least that means I have the circuit diagram to hand...

There isn't a lot between V3 and the selectivity switch. My prime suspects would be R39, which is in V3's anode circuit, or C67 which decouples it. There's also the insides of the 'crystal filter transformer' to look at. Or could it be just dirty contacts on V3's socket?

Is the BRT400E the later version, with B7G valves? If so, apologies for my ignorance: I've only got information for the early version with Loctal valves.

Chris

[cmjones01]

Originally Posted by radiogm7nva

If I try the AVC on/off switch during the fault, the AVC will not operate. As soon as the fault clears it is fine again. Operating this switch or any other will not clear the fault. Switching the radio off and counting to 5 will also clear the fault.

Thinking again, could this be an AVC fault? For example, C78 going short-circuit would scupper the operation of the AVC switch. Though I'd have thought that would result in an increase in gain rather than a decrease. There's a lot of complexity in the AVC department.

Measuring the DC conditions around V10 under the fault condition might be instructive.

[radiogm7nva]

Hi Chris and many thanks for being at hand.

Yes this is a B7G unit. I have made a note of the C78 / R39 suggestion and the XTAL box investigation also.

Yes the AVC application is complicated.. left me with a sore head But I do notice that the volume, when in fault mode, does not alter when the AVC switch is operated.

When the fault clears, AVC 'off' gives ample, if too much, gain. The 'on' position works a treat when the fault is cleared. I have tested on either side of the XTAL box and the fault is present on both sides. That is: on the Waffer switch (13) and the variable capacitor on the other side of the XTAL unit.

the last place the signal is in good shape is at the front end of valve 3.

Going to check C78/C39 this weekend.
John

[radiogm7nva]

Update... C78 (AVC Switch) and R39 in Occ box check out fine.

[cmjones01]

Originally Posted by radiogm7nva

Update... C78 (AVC Switch) and R39 in Occ box check out fine.

Well, they might be intermittently failing - can the fault be persuaded to persist for long enough to look at the DC conditions around V3 and V10? I'll bet it's one of those faults where threatening it with a scope probe cures it.

[radiogm7nva]

Fault found.

The BRT 400 seems to suffer from valve pin/socket issues. In this case a pin on the output of valve 3 (mix / 1st detector) was giving bad contact.

By gripping the pin on the underside with insulated pliers and pulling a little on the pin a fault could be created and cleared. Cure will either be replacement of the valve socket base or a temporary fix of some type.

It is worth noting that other valves had a similar potential to cause problems. Valve 5 on this unit had problems.

On the IF area, some of the valve pins were under threat from coax transiting the area and lying on the valve pins. I found some potentially bad shorts because of this.

Recommend that owners examine every section of these radios, especially valve pins, for adjacent contacts which are too close for comfort.

Through time, servicing and general examination can cause wires to be moved away from their original construction position or layout.

Cheers for all the help! John gm7nva

[cmjones01]

Originally Posted by radiogm7nva

The BRT 400 seems to suffer from valve pin/socket issues. In this case a pin on the output of valve 3 (mix / 1st detector) was giving bad contact.

Great! I'm glad you tracked it down, and it's nice to know we were all thinking about the right area.

The Loctal sockets in mine look none too reassuring as regards pin contacts, but at least everything's a bit further apart than on B7G ones...

[radiogm7nva]

Another or very similar fault developed. Dam!

The audio still cuts to a whisper after a while and touching Det/Mix valve, pin 2, ..output anyhow.. (lightly) with a previously grounded screw driver (and carefully) seems to bring the audio pounding back.

Looks like this fault is actually the same fault as previously mentioned. Having said that, after doing this screw driver touch trick, the audio stays, 99% of the time, until the radio is turned off. I ran it for 4 hours without a problem the other day. Turning on the radio again the signal seems OK for about 30 min then, goes again until the valve pin is touched.

Originally, I thought that this was a bad connection but the light touching isn't enough to disturb a dust speck! Could this be a discharge into the previously grounded/discharged screw driver?

I tried a couple of times and it works every time. Not discharging the screwdriver before using it doesn’t seem to work but, I accidentally grounded SD to chassis while touching the valve pin and zap! Back to life it all came.

Ordered a new valve in case it might be a fault in the old one but still not convinced that this is the problem.

I am convinced that the problem area is still in the V3 area though.

John

[cmjones01]

Strange. Something being that sensitive would make me suspect that the stage in question is oscillating when the fault occurs, and the tiny perturbation from the screwdriver is enough to stop it. Or there's a DC fault of some sort on a valve's grid: I spend half my working life tracking down intermittent faults due to floating gates on CMOS logic. Either leakage to a grid pin or lack of a discharge path might cause something similar in a valve circuit.

Or it's just another dirty contact...

Chris

[Biggles]

I have a BRT 402 which has lain in my workshop for quite a few years now unrepaired. It had an intermittent fault for a while until it eventually gave up altogether. The fault had similar symptoms to yours but had a loud crackle at the same time. The fault was eventually traced to a primary-secondary short in the final IF transformer which has a multiple tapped winding to alter selectivity. It may be worth checking yours for this. I gave up trying to find a replacement as they appear to be made from unobtainium! I think that the primary and secondary may be wound over each other for an internal short to occur but it's a long while since I looked at it so I may be wrong. They are a bit of a nightmare to work on as the components seem to be packed closely together under the chassis.

Biggles.

[radiogm7nva]

Thanks chaps..some good ideas to work with here. The oscillating explanation sound interesting. The other idea regarding the IF too. I have managed to obtain a spare Det/mix - valve 3 (7s7) I will pop this in, just to make sure there is nothing odd with the original valve. will be a few days before I can get back with a report. - John

[radiogm7nva]

Right! We have got to ground zero (I think)

On the Anode / output of V3 (Det/Mix) Pin2, there is 250v which is correct according to schematic and manual. This goes toward one side of the primary (Crystal Filter Transformer) via a TC6 . On a scope the signal on the valve side shows that it is unaffected during fault condition (loss of gain, but signal still audable, and meter falls to nothing, first slowly then sudden drop). Now...

On the other side of the transformer primary, after the coil wynding, the signal shows (on scope) a fault (reduced signal) when the fault decides to appear.

So, am I right in thinking that whatever is in this CFT Primary, has a problem OR is there some external component playing a card here?

All that is in this CFT on primary side is A tunable cap TC6 a fixed cap C66? and a coil or wynding.

Remember: Lightly touching pin2 with a screwdriver or rotating band selector switch 13, often 99% of the time, clears the fault. After the radio has been on for a short time the fault might not appear again for hours and hours!

The AVC is OFF but the fault appears in either position.

What is the verdict...anyone?

Cheers and Merry Christmas all

John


Just read Biggles report before this posting. Looks like we have it then. Thanks!

[radiogm7nva]

Update:
Forgot to add that I checked the Secondary on the Crystal Filter Transformer. If there was a short across Primary/Secondary I think I should expect to see the 240v on the secondary? Nothing there.

[Colinaps]

What symptoms now, John? Readers here will have missed a chapter - C33 was a dead short, which meant that R17 was dropping about 260V when in any band other than 6 (LW). Now replaced with a "near enough" cap from the junk-box. There's a bit of a glitch in the track of the RF gain pot as it approached maximum - this has been bypassed with 100R for the time being to give nearly max-gain all the time. At switch-on, the radio is silent for several seconds then comes on very abruptly as the oscillator kicks into life.

It might be worth checking for leakage in ALL those naughty waxy paxolin caps - in some cases this would show up as voltage drop across the adjacent resistor when it's supposed to be out of circuit anyway.
Is the oscillator oscillating and is the voltage regulator striking (purple glow)? are the voltages about right on all the valve pins according to the book? Measure these with your AVO - your digi meter will read a lot higher in some parts of the circuit. Also be EXTREMELY careful about manhandling the switch connectors - changing wafers is a compete pus of a job - if you can actually find replacements.
Have a look at the hum control - is it making any difference now? Can you check the HT current? This could be done by measuring the voltage between pins 2 and 5 on the multi-pin connector, divide the voltage by 220 (r100 - if it's not been cooked) to give an answer in amps.
There are plenty winter nights left to finish the job!

Colin.

[Station X]

Readers here will have missed a chapter.

Why's that then?

[Colinaps]

because I've been over to drink John's Guinness!

[Colinaps]

Finally got round to it....
Rebuilt the LO valve area; modern components are a big advantage, there's very nearly room in there to work now. Everything has to be built on the valve base and fed in a bit like a ship in a bottle! L34 had to come out as well, which is not terribly easy because the threads are varnished and there's nothing really strong enough to hold the coil by.
There was one dry joint on the valve base, but that was completely invisible, let alone inaccessible in situ.

We used the circuit diagram and parts list for the later model; the oscillator section is the same though the part numbers differ.

The RF gain pot still needs replacing - fixed resistors at the moment set it to about 3/4 max, so there's a lot of unnecessary background noise in the absence of signal.

Now - touch wood - John has had the brute running all day without any dropping out

[Colinaps]

No such luck! Onwards to the mixer again.....cathode caps?

[radiogm7nva]

Fault Discovered!

After almost a year of weekends on the bench 'Colinaps' discovered the fault. The fault being, intermittent audio level changes. The volume would fall away either slowly or instantly and the meter would fall to zero.

The fault would appear at random times whither the radio was cold or hot. The audio level would drop to a whisper (meter showing this also). Two methods were found to temporarily clear the fault while searching for the cause. The first being a light touch with a screw driver at certain contact points. The second being a full rotation of the selectivity switch, returning to any setting. The fault would appear on any selectivity setting. Other controls had no effect on clearing the fault. Creating the fault was very difficult. Usually rotating the selectivity control eventually started it.

We checked all switch contacts, replaced most of the old capacitors, except the silver mica ones. Almost all the resistors were changed. Still no solution!

All the valves were looked at, and one base changed. Voltages were checked and some problems sorted.

The source of the intermittent audio level was eventually found to be inside the crystal filter transformer.

Inside this case are two capacitors and some bare wiring. It was the bare wires which, somehow, were intermittently touching the filter case, the action being, to short out to ground, anything, going through the primary side of the can.

There are other transformer cans in these sets. The layout of the wires appears to be as designed so, we can only assume these are other sets out there with this fault.

Caution: There are a few versions of schematics for this radio. Each model and its schematic, although mainly the same for each version, have different component numbers on the sheet. Make sure you have the right sheet for the right model.

Mauritron have the 'D' and 'E' version RCH 1067 (on sheet) Other sites may have the 'K' which is RCH 1180 and the early 400's seem to be RCH1001.