Identify Grundig Mystery FM sound. De-emphasis issue?

Restoring Grundig Mandello d//U. Experiencing a rhythmic thumping sound but only on FM. Sound goes away when bass is reduced. Thumping not in time with the bass, but has it's own rhythm. Video of the phenomenon here.

To thank the Author because you find the post helpful or well done.

Hi Keith,

your model Mandello d/U seems not being present in RMorg.

To understand the behaviour of the audio amplifier part, it would be helpful to know more about this console. Please make a new model suggestion for your Grundig Mandello d/U  with all known details (tube line-up etc) and add your pictures along with the proposal.

Do you also have the schematic? You can scan and upload it too.

Thank you for your assistance here.

Bernhard

To thank the Author because you find the post helpful or well done.

I added the model and some pictures. Unfortunately I do not have the schematic.

To thank the Author because you find the post helpful or well done.

Some background about pre-emphasis and de-emphasis may be found in "FM Pre-emphasis and De-emphasis".

Regards, Dietmar

To thank the Author because you find the post helpful or well done.

I've seen that thread. It's very technical, but I really just need confirmation that the sound in my video is or isn't an emphasis related issue and if not, what is it? Many thanks.

New model suggestion was accepted.

To thank the Author because you find the post helpful or well done.

Another thing to note: this chassis doesn't have the optional decoder unit. Could that be the issue, given that the problem occurs only on FM?

To thank the Author because you find the post helpful or well done.

It's neither to do with de-emphasis or optional decoder.

It's more likely instability caused by a faulty capacitor.

To thank the Author because you find the post helpful or well done.

Michael is right here, the thumping sound is not caused by the de-emphasis circuit (this only flatten the pre-emphased audio signal from the FM transmitter). We name this phenomenon "Motorboating".

From a first analysis of the circuit, the effect depends on an interaction between both audio channels and the power supply. Common and important for decoupling the left and right channel preamplifier from the power amp section are filter capacitors, usually electrolytic ones.

In this case, the Grundig chassis HF45(U) uses a multiple filter capacitor containing 2x 100 µF and 1x 8 µF (C88, C86 and C83) in one big aluminium can. The 8 µF is connected via a 10 kΩ resistor to the second 100 µF. If the 8 µF fails, fluctuating voltage from the push-pull final stages will pass to the preamplifier section with the double triode ECC83. A sort of reaction will start and the thumping begins. If the bass gain is decreased, it's just like adding a high pass in this "oscillating circuit" and the gain isn't  high enough to keep the motorboating going on.

What you can do is: Check all three parts of this combined electrolytic, maybe one or two of these are faulty. Capacitance loss and/or an increased ESR (equivalent series resistance) will reduce the filtering effect noticeable.

You can measue these with a capacitance meter, some also have an ESR reading in conjunction with the indicated capacitance. There is no need of desoldering these caps before measuring, but please check for residual DC before connecting the cap meter! For electrolytic filter caps, a measuring frequency of 120 Hz is advised. But 1 kHz will also do the job.

For a first test (and if a cap meter is not on hand), just connect another 8...10 µF 350V electrolytic in parallel to the 8 µF part and check if the thumping sound has gone.

Please refer also to this forum article at post #2 for a similar problem with a Telefunken Opus 55

Good luck!

To thank the Author because you find the post helpful or well done.

I had already replaced the 3-part can with individual capacitors. I just swapped out all three of those and the problem remains. Went back over my wiring of those three electrolytics and it checks-out with the schematic. One thing of note is that I didn't re-use the exact holes in the board to make those connections, but used long wires to run them directly to their next connection on the board. I can't see that making a difference but then I've been wrong before.

To thank the Author because you find the post helpful or well done.

Dear Keith,

Can you check the two grid voltages on pin 2 and 6 of the ECLL 800. There should be no positive voltage at all. Both voltages shoukd be slitely negativ, e.g. 3V but depends on the input resistance of your voltmeter.

Regards, Rolf Beckers

To thank the Author because you find the post helpful or well done.

I can try. I'm not great at taking voltage readings. Assuming I should do this while on FM? Also, it won't be hooked up to the speakers (they're all iin the cabinet); assuming that won't be a problem. Do you suspect a bad tube, or just that this is a point to identify another problem? Thanks.

Also, which ECLL 800? There are two. Thanks.

To thank the Author because you find the post helpful or well done.

I took the voltage readings from pins 2 and 6 of each ECLL 800 tube, using the 20 DC volts setting of a digital multimeter. The readings fluctuated between -.01 and .02. It did this on each pin, or with just the negative lead on the chassis. When I say fluctuated, I mean it did so in a rhythmic fashion, moving from -.01 to .02 and back again very evenly, about one second per hundreth of a point, if that makes any sense.

To thank the Author because you find the post helpful or well done.

I'm sorry, but I've no other ides where to look for this phenomena

To thank the Author because you find the post helpful or well done.

Same. Only thing I can think of is the selenium rectifier, but I never know what value resistor I need to put in with it, and of course that might not even be the problem. Is there a particular tube associated with the FM? Maybe that's bad in a way that can't be detected via the usual means.

To thank the Author because you find the post helpful or well done.

Hallo Keith,

my guess to this curious phenomenon: Check the capacitor C76, 50uF-10/21V (at the Balance potentiometer)  by soldering a similar capacitor in parallel. This capacitor at the negative feedback path of both audio chanels decouples the two audio chanels. Check also for good contact to ground of this capacitor. Otherwise an uncontroled couplig might cause the unwanted oscillations. An other - less probable candidate - might be the 250uF (or 150uF) capacitor at the cathodes of the ECLLs and ground. 

Goud luck! Regards Uwe

To thank the Author because you find the post helpful or well done.

Thanks, Uwe! C76 seems ok, but the other one (C87) was the wrong value. I must have mis-read the old capacitor because I put in a 15mfd one, instead of 250. Anyway, that didn't fix the problem entirely, but it is markedly better. I'm probably going to call it a day on this one, as other than this one problem with FM it is otherwise a good radio. Thanks everyone for their assistance. I will update here if I learn anything else.

To thank the Author because you find the post helpful or well done.

Any possibility that a bad EAF801 tube is to blame? It tested on the low side.

To thank the Author because you find the post helpful or well done.

Unlikely. The EAF801 is part of the AM/FM IF amplifier (and demodulator for AM). The view should be directed towards the power amp. If all filter caps (and values of other parts in the power amp) has been checked and turned out to be ok, both ECLL800 should be checked or replaced on trial by known good tubes.

Be careful when ckecking the ECLL800 at a tube tester. Since both G2 terminals of the power pentodes are connected internally, the g1 of the penthode not being measured shall have a negative potential of eg. -20 volts to cut-off the G2 current. Otherwise it will be overloaded! And this vice-versa when measuring the 2nd penthode.

Please read there in Post #2 (already mentioned here in post #8), that a possible reason of motorboating could be a weak tube in a push-pull system causing an unbalanced situation which is prone to unstable operation.

To thank the Author because you find the post helpful or well done.