Grundig TK 5 Restoration

[DMcMahon]

Last night found the fault and thought it was fixed.

Found that the main DC HT voltage supply was jumping up and down, found that capacitor C36 (100uF) -ve end lead was touching/shorting down to the metalwork.

C36 is the reservoir capacitor for the -24 volts supply, which is only used to power the relay for the EOT sensor. In an earlier post had replaced this capacitor and also C32, both of these capacitors being secured by a metal wrap round bracket.

I think because the bracket was not a real tight fit around the capacitors and because the chassis is out of its box, that that the bottom end lead (-ve) of C36 was resting on the floor and had pushed up C36 enough so that its +ve lead (which was not fully insulated) made contact with the nearby metalwork.

Re-positioning the capacitor (also fully insulated its -ve lead and making the bracket a tighter fit by adding more packing material) so there was no short, fixed the horrible extremely loud clicks. Recording and Playback then working OK.

The -ve end of C36 connects to the -ve side of the Selenium Bridge Rectifier, which also connects to the -ve side of C32 which is the reservoir capacitor for the main DC HT supply. Slightly unusual arrangement for the 24 volts, typically the -ve side of the bridge rectifier is connected to ground and a separate bridge rectifier is provided for the 24 volts supply (when there is a 24 volt supply).

Even though everything was working, the voltages at the 2 reservoir capacitors has changed (last measured the day before). The main HT voltage has increased to around 270 volts from previous 235 and the -24 volts has decreased to -14 from previous low twenties. A few quick measurements revealed no obvious cause for the voltage change.

This morning with the TK 5 just left switched on, there have been several intermittent bursts of the same horrible loud clicks. I have found that I can easily induce exactly the same sound by very lightly tapping the EL 42 output pentode valve. Pushing/holding/waggling the valve does not induce the sound, i.e. it has to be a tap/knock (gentle).

So more investigation required.

[DMcMahon]

Investigations ongoing. For the change of HT & LT voltages have tried various including another set of reservoir capacitors and a replacement Selenium bridge rectifier with no change.

The terrible noise issue is ongoing, intermittently occurring or can be manually initiated by lightly tapping the EL 42, after various checks think it is the valve itself, just trying to prove it conclusively.

[Refugee]

I would take the EL42 out and link all the grids to the anode and apply 6v to the heater and test it with a 10K resistor and a 30V power supply.
If that fails disconnect the grids one by one while repeating the test.
You might be able to find what grid has got a loose wire on it.

[DMcMahon]

Thank you Refugee for your suggestion. I do not have a spare valve base socket which without would make the above awkward to do I believe.

I have thoroughly checked all the valve connections, including pushing and prodding with an insulated stick and cannot get any reaction. Also pulling/pushing/trying to twist the valve itself does nothing.

So I think there is no problem with connections either the wires or the valve base itself.

It is only when I lightly tap the valve glass envelope that I can easily induce the noise.

There is a second EL 42 fitted for the bias oscillator, I have been thinking of trying that valve in the output position but somewhat reluctant currently in case it gets damaged by the very large voltage signals that are generated in the fault condition.

I will have a fresh look later this morning.

[DMcMahon]

Did some more tests and 100% confident that the issue must be the valve itself.

Fitted the EL 42 from the Bias Oscillator position into the audio output position and this valve does not exhibit the noise issue, taping this valve (quite hard) just gives a bit of hiss type noise of a few volts at the anode. The bad valve when lightly tapped outputs > 600 volts peak to peak noise spikes.

Can only assume that something internal within the valve is shorting intermittently which can be induced by taping it.

Need to try the valve out to make sure that playback still works, first have to tidy up various wiring that has been disconnected during the various tests.

Have ordered a NOS EL 42.

Have one other EL 42 that I can borrow in the meantime (from my TK 819) to fit in the Bias Oscillator position to be able to check out recording again.

[Refugee]

That valve has a nasty fault. Lets hope that the output transformer has survived.

[ricard]

I had an EBC81 in a radio that was acting up, the valve behaved as if there was a bad connection in the valve holder, but cleaning didn't bring anything but temporary relief. Finally replaced the valve, problem solved. In this case I'm thinking one of the internal welded joints had loosened inside the valve.

[DMcMahon]

All back together and playback and record (only checked microphone, still to check DIN line recording) working well, so the output transformer survived the abuse.

Interestingly the HT & LT supply voltages went more abnormal the longer the bad valve was left in/powered, with the HT getting up to 320 volts and LT down to -2.5 volts

With the replacement valve (taken from the bias oscillator) and the bias oscillator valve replacement by borrowed EL 42 from my TK 819, the HT and LT voltages returned to more normal values. i.e. HT +237 volts and LT -27 volts.

Hard to know what caused the bad EL 42 to go bad, maybe just old age failure or maybe helped along by the previous shorting down of the reservoir capacitor event (I think that might actually have been a red herring).

[Refugee]

If you are really curious you could always crack the old valve open and have a careful look at the internal components.

[DMcMahon]

Yes that would be interesting.

[Refugee]

You can usually knock the pip off the top and then chip the remaining glass away over the land fill bin to reveal the interior ready for dismantling.

[DMcMahon]

I will certainly try that out of interest during a quiet period.

[Refugee]

I see a long quiet period coming up...

[DMcMahon]

A little mishap yesterday. I was getting ready to test DIN line recording and remembered that there are a couple of things I have been meaning to repair by gluing for a long time, so thought I would do the repairs first.

When I first got the TK 5 and opened it up, found that a little plastic holder bracket for the Fast Wind indicator lamp was broken, I should have repaired it at the time but just wrapped insulating tape around the bulb holder to hold it in place.

So yesterday I started the repairs, removed the insulating tape (exposing live contacts on the base of the lamp holder) but got side-tracked on something else before I did the glue repair. Got back to it later and forgot I was going to repair it and switched it on to start recording test.

Of course as luck would have it, the live contacts on the lamp holder were resting on the metal work with subsequent flash/bang ! then a dead unit.

The 5V 0.4A MBC bulb is in series with one side of the motor winding, the 2A internal mains fuse had not blown, but it had blown the 1A fuse in my 3 pin plug (I use 1A fuse normally for max protection).

Replacing the fuse got the unit powered up, but no tape transport operation, found that the bulb had also blown, so now just need to find a suitable bulb, I think there is a very similar one in my TK 35.

Morale of the story here is not to delay important repairs

[julie_m]

The bulb being used as a fuse for the motor probably is long-since NLA, but you could use a fusible resistor in its place. You will need to use Ohm's Law V = I * R and the Power Law P = I * R to select a value which will be dissipating as nearly as possible its rated power in operation.

If you use a 10 ohm resistor as a temporary substitute for the bulb, you can measure the voltage V across this and then get the current I using I = V / R = V / 10. Then assume the motor resistance is large compared to the safety device, so the current won't change much with a different resistor. Just choose a fusible resistor with a suitable value to give a power dissipation at that current which is just within its rating. If the motor current increases, the power dissipated in the fusible resistor will increase twice as fast (since P = V * I and V = I * R, so P = I ** 2 * R. If the current increases to 1.1 times as much as it was before, the power will increase to 1.1 times 1.1 times as much as it was before = 1.21 times). Careful manufacture ensures the fusible resistor will fail (or, from the point of view of whatever it's protecting, succeed .....) open-circuit when its rated power dissipation is exceeded, and without sending shrapnel flying nor setting fire to anything in its vicinity.

[DMcMahon]

Thank you Julie for this useful information.

Not seen a bulb protecting the motor winding in these old units before.

Operationally the bulb is used as an indicator (shines up through a hole in top panel) to show that "Fast Wind" is selected. Fast Wind is a position on the rotary Play/Record/Stop selector and is mechanically interlocked with a Fast Wind mechanical selector which selects Fast Forward and Rewind, i.e. you cannot select FFD or RWD operation until you first select "Fast wind".

I borrowed a suitable/similar bulb (7V/0.5A) from my TK 35, it is working fine in the TK 5, but not that good as an indicator in the TK 5 because it has a blacked out curved end.

I have ordered a couple of spare bulbs.

Checked DIN line recording, all working well, for > 60 years R2R non Hi-Fi, it sounds quite good to my ears.

Now am doing a bit of cleaning up, lubricating and will check humdinger set-up before putting it all fully back together. Also trying to find some wax crayons to colour the selector switches lettering/icons.

[julie_m]

We used to use fusible resistors on a lot of products when I worked in an electronics factory .....

Note that if you are using another lamp, it's in a series circuit; so the current rating is what it's important to match against the original. A 12V/5W automotive lamp runs at about 0.4A .....

[DMcMahon]

After a good clean up, lubrication and humdinger adjustment (very minimal effect) this morning, tried to get the unit back in its case, which at the best of times is tight.

2 or 3 things were slightly fouling, making it more difficult than usual to get it in. Attempting to ease it in, resulted in the glued wooden blocks that it secures to coming loose, so have re-glued them.

While waiting for the glue to set, have had a quick look at latest acquisition a TK 14.

[DMcMahon]

In preparation for refitting the unit back in its case I thought I would attempt to clean up the green perforated plastic wrap around case, which looks very dirty.

After a lot of elbow grease with large number of various products applied with cloth or sponge not a great deal of improvement, think the dirt is ingrained into the plastic rather than it being surface dirt (or it is not dirt as such, just a degradation effect of the aged plastic material). I applied a small amount of cleaning product to a small area each time to check for its effect and to minimise mixing different chemicals.

Cleaning products tried included, water, water + liquid soap, washing up liquid, CIF, various kitchen/bathroom cleaners, Sugar Soap, WD40 (tried this as it does clean some materials well), Vinegar, Lemon juice, IPA, Meths, White Spirit, Lighter fluid, Paint brush cleaner, Turps, Cellulose Thinners, Petrol, various car cleaning products and various cleaning wipes etc.

Was interested to see if there was any easy/magic way of getting the plastic clean, my TK 14 white plastic case is particularly dirty/grubby looking, so was hoping to find an easy method of cleaning.

In the end on the TK 5 I had to resort to a non scratch kitchen sponge scourer pad together with a kitchen JIF lemon spray cleaner and after quite a bit of elbow grease got it reasonably clean but not as good as was hoping for. Was somewhat surprised how difficult it was to get it clean looking.

[DMcMahon]

Replaced the blown Fast Wind indicator/motor protection bulb with a new bulb rated at 0.6V/0.5A.

At post 56 where I said "shines up through a hole in top panel" should have read "shines up through a small red window in the top panel".

Refitted the metal securing bracket around the 2 reservoir capacitors, believe originally may have had it back to front, which allowed the bracket to protrude out somewhat, this was one of the things that made it difficult to get the unit back in its case on previous attempt.