The Grundig TK149 saga part 2

[Tractorfan]

Hi,
Well, I'm running out of ideas now. I've tried:
Different valves.
Cleaned and degaussed the head.
Cleaning the record/play switch.
Cleaning the track selector buttons.
Completely by-passing said track selector buttons.
Re-terminating the screened leads to the record/play head.
Checking the leads soldered to the PCB.
Cleaning the leaf switches underneath the deack.
Checked the continuity of said leads.
'Scoping the signal on the record head which shows a strong clean bias waveform modulated by the audio.
I've even tried varying the bias to the head by twiddling the variable capacitor between the oscillator and the head.
It plays back tape recorded elsewhere OK and will erase the tape, but simply won't record.
What else could it possibly be
Cheers, Pete.

[8 Tracker]

Hi Pete,

In your previous post you said you had a strong signal on the erase head during record, but very little on the record head. Are you saying you now have a good bias signal on the record head, yet it still will not produce a decent recording?

Mark.

[Tractorfan]

Hi,
Yes, Mark. I think that was because I failed to realise that the record/play head was fed via the centre cores two screened leads, instead of one centre and the screen. Depending on whether the machine is in play or record mode; one side or the other of the head is grounded by the change over switch.
However, now that that particular penny has dropped, I get a good bias signal on the head and I can see that it's modulated with the audio. So, it all 'appears' to be tickety-boo but -- nothing.
Should the record head bias be stronger than on the erase head?
This model has a 'trick' function where the record button can be depressed while playing back to insert a recording or a silence. This kills the signal to the erase head, but strengthens the bias on the record head. Hm.
Cheer, Pete.

[ricard]

The recording bias can very well be larger than the signal on the erase head. I think it is hard to give a definite answer to that, as it is actually the current through the head that does the job, but it's much easier to measure the voltage across the head.

The fact that the bias goes up when the 'trick' button is pressed is probably just a consequence of the reduced loading on the oscillator with the erase head disconnected (or, if the disconnection of the head is compensated for by switching in a coil or resistor in the circuit, that the matching between the head and the compensation circuit is not perfect).

[SteveCG]

Just to confirm: have you checked that the earth/chassis side of the record head is actually at earth/chassis when in record mode? That is, you could indeed have a strong bias and recording signal on the 'hot' side of the head, but the other end of the head needs to be at chassis for the current to flow through the head.

[8 Tracker]

Ricard is correct about the bias level increasing during "trick" recording. I have observed the same on my machine.

A couple of questions. First, how does the recording sound? Is in faint and tinny or very muffled. Is it the same on both tracks?

These 4-track Grundigs can be a bit of a pain to get proper alignment, is the tape making full and proper contact with the head? Is the tape "wandering" up or down? What is the state of the plastic strip with felt front like that holds the tape on the head?

Mark.

[Tractorfan]

Hi,
Thanks for the explanation re: the bias getting stronger in 'trick' mode. However, it makes no difference to the recording, or lack of it.
Steve: Yes, each side of the head is grounded as and when required, so that seems OK.
Mark: The recording isn't appearing at all, despite the strong modulated waveform. I have completely bypassed the track switches to eliminate a possible weakness but hitherto, both tracks were similarly affected so, in effect, it's a quarter track machine! The tape path is clean and in good order without wander or lifting, also, the pad is OK. Therefore, I'm certain that the problem is electronic rather than mechanical.
The other day I was getting a faint crackly recording, but now none at all. It's got to be an intermittent fault somewhere, or I've done something to lose what little recorded sound there was.
Of course, in the true spirit of us forum members, I don't actually need the machine, but I'm willing to spend hours fiddling around to get it working proplerly again!
Cheers, Pete.

[SteveCG]

So we are absolutely sure that we are applying signal and bias to a clean head that is not open-circuit, and yet nothing is getting recorded. To eliminate the most unlikely then: Is the tape alright ?

[Nicklyons2]

What's the bias frequency? I'm just thinking if it's gone far too high due to say a capacitor failure, the head's impedance will be so high as to reduce the record current.

Also, I don't know what you used to clean the head but an old colleague who worked for Tandberg swore by Brasso! He reckoned it cured innumerable cases of strange frq response/alignment; he claimed it was because any oxide grains wedged across the gap forming a magnetic shunt were removed by the more brutal effect of the Brasso where mere cleaning with meths didn't shift it.

[Tractorfan]

Hi,
The DC resistance of the head is about 1K, and it plays tapes recorded on another machine clearly, so I can't see it being muck or the 'wrong' type of tape. Otherwise I would have expected the recording would be muffled, but not absent. Nor can it be open circuit as it would hum, surely.
I don't know the frequency, alas, as I never learned to 'read' an oscilloscope when I was at tech college (I must have been off that day ). But looks rock steady and the amplitude can be altered quite a bit by the variable cap on the PCB.
I'll certainly give the head a scrub and see what happens during the next wet afternoon.
Thanks for your suggestions.
Curioser and curioser.
Cheers, Pete.

[peter_sol]

The head may be so worn that it will not record but will still play back ok.

[Tractorfan]

Hi,
I didn't realise that could happen. OK, I'll whip it out and have a close look next chance I get.
Cheers, Pete.

[ricard]

Originally Posted by peter_sol

The head may be so worn that it will not record but will still play back ok.

If the head is very worn and the gap has widened, it will a) most likely be visible and b) result in a muffled playback which cannot be cured by azimuth adjustment.

Since pure record heads have a wider gap than corresponding playback heads, I would expect that playback response drops a lot faster than the record response.

So if the head is playing back properly I would think it could record without problems.

As for measuring the frequency on a oscilloscope: no bug deal, a slight amount of maths is needed. Set the horizontal sweep control so that you see slightly more than a full cycle of the bias waveform. (Leave the continuous control at it maximum (calibrated) value. Then count the number of squares in the grid on the screen that it occupies. Let's say it's 6.5 . Multiply that by the setting on the sweep control, let's say it's 5 µs per grid segment. that gives us a period time of 6.5 * 5 µs = 32.5 µs or 0.0000325 s . Now do 1 / 0.000032 s which gives you 30769 Hz (which would be a trifle low as a bias frequency if this were a real world example).