Ineffective tone control.

[Station X]

I've been working on an unidentified transistorised Fidelity record player, probably an HF39. The tone control is totally ineffective, no matter where the knob is set there is no change in the sound.

The circuit is very simple, a 50k pot is wired in series with a 56nF cap from the bottom of the output transformer to the negative rail. All wiring shows continuity. The pot tests good. The cap is actually a 47nF component which measures 41nF and has no electrical leakage. 56nF is not a common value so I fitted a replacement 47nF component which actually measures 47nF, but it makes no difference.

Is fitting a 10nF cap in parallel likely to make any improvement? any other suggestions?

Thanks.

[HamishBoxer]

Should make some difference.

[ms660]

Has the treble already been removed pre tone control?

Lawrence.

[Station X]

I don't know. To my ears it actually sounds quite good for a record player with no pronounced bass or treble. It just seems strange to me that the control doesn't work.

Maybe I should check the 200uF bypass electrolytic?

[ms660]

I would sig gen and 'scope it.

Lawrence.

[Radio Wrangler]

The circuit looks familiar.

I think it's straight out of a Motorola application note that got reprinted in a number of those John Markus circuit cookbook tomes.

The effective working Z at the collector of the MJE340 will be an awful lot lower than the 50k tone control pot, so any effect will be cramped right at one end, and somewhat limited by the minimum slider R of the pot.. The capacitor value will make any effect rather high in frequency.

I wouldn't expect it to have much effect.

I remember the circuit because I noticed the crudity of the bias scheme, That first transistor is a PNP that is emitter following a signal WRT ground, and its collector supply is ground! Gruesome!

David

You could try one of the free "spectrum analyser" apps* for a mobile device to see any difference. This assumes you have one that is.

*application, called an app for the hard of thinking.

[Mike. Watterson]

It looks like a transistor version of a typical valve radio tone control.
The 200uF isn't the problem.
The 100pF across the input resistor boosts treble. Check
The cap on base of 1st transistor cuts treble. Check.

The tone control only will affect the higher treble. Thus if the system lacks treble it will do nothing.

[Mike. Watterson]

Originally Posted by Radio Wrangler

I remember the circuit because I noticed the crudity of the bias scheme, That first transistor is a PNP that is emitter following a signal WRT ground, and its collector supply is ground! Gruesome!

David

The 1st transistor is purely to have a high input impedance. Hence an emitter follower, gain = 1. Input impedance roughly transistor gain x emitter loading.

[Herald1360]

Isn't the dc level the audio at the input to the first stage is centred on about 5.5V?

The circuit is crude, but with only a volt maybe of audio it should work OK.

The input Z won't be particularly high, emitter follower or not since the emitter load of the first stage will be dominated by the input R of the second stage, only about 55 ohms.

[Radio Wrangler]

Nope.

AC coupled input and just a 200k resistor to deck to set the DC on the base. Before the input has swung one Vbe negative the transistor will be in saturation and the input impedance will also be very low.

Running with such low Vce, the current gain will be drastically reduced, too. Typically data sheets give Hfe with 5 or 10v Vce, not 0.7. And the low Hfe limits the extent to which the emitter follower improves the Z of the subsequent stage.

David

[Herald1360]

Just looking at the circuit and the typical voltages voltages thereon...... MJE340 emitter is at 5.5V so its base is at 6.2Vish as is MPS6517 emitter and so its base will be back at about 5.5V which will be dc level at amp input. MPS base current through the 390k will let this happen. Input base signal dropping by any amount will simply cause the emitter to follow it down so no saturation as far as I can see.

The emitter resistors for each device will give a degree of dc negative feedback tending to stabilise the operating point(s). Not brilliantly well defined but good enough for the application. An interesting bit of cost engineering (apart maybe from the need for an output transformer) I reckon.

Just off for a play in Simetrix to see what happens in a simulation.....

Had the play- the circuit seems to function dcwise in line with the above. It's fairly sensitive to the gain of the PNP device, but not to the point that would upset its operation as a simple amplifier.

[Trigon.]

If you don't have a simulator - a reasonable approximation using a simplified model:-

TR1 base voltage = 390k*Ib, emitter is at = 390k*Ib+0.6 (Assuming a 0.6v Vbe)

TR1 emitter is also at 90-47k*Ic

- So: 90-47k*Ic = 390k*Ib+0.6

Solving for Ib, where Ib = Ic/hFE

Ib = 89.4 /(390k + 47k*hFE)

The MPS6517 has an hFE spread of 90-180 at 2mA. - At 90 the base of TR1 will sit at ~7.5v, and at 180, ~3.9v.

(5.5v would require a gain of ~126.)

As a matter of interest, how close is that to the simulator ?

Cheers

[Argus25]

The DC bias conditions of this circuit are far too dependent on the hfe of the input transistor, this is easy to see if you imagine what happens with an imaginary very high hfe, the input transistor's emitter voltage would be very close to ground.

I think there is a missing DC negative feedback resistor from the emitter of the output transistor to the base of the input transistor, and if there really isn't one, there should be.

Both the value of the tone capacitor (too small) and the tone pot (too large) appear inappropriate for the task with the probable impedance in the collector.

Possibly, this diagram in defective in more than one way.

[Station X]

I'm overwhelmed by the response to my post and will do some testing later in the day.

No shortage of test gear here, so I can do a freq run if necessary.

All I have to add at this point is that the cartridge is an X3H high output type and is obviously working as there's no shortage of volume.

[ms660]

The schematic looks correct to me and appears to tally up with the layouts in the links below as far as I can tell.

A point to point version:

https://www.vintage-radio.net/forum/...2&d=1325456797

A PCB version:

https://www.vintage-radio.net/forum/...9&d=1531774417

https://www.vintage-radio.net/forum/...0&d=1531774417

Lawrence.

[Station X]

The PCB version is the very player I'm working on.

[Station X]

Initial findings.

CIRCUIT.

0.1uF 400V cap not fitted.
Smoothing can is 60uF + 150uF, not 32uF +32uF.
Tone control cap is 0.047uF, not 0.056uF.

Carbon comp resistors have all drifted high. Notably 680k on input measures nearly 1M.

VOLTAGES.

"HT" at O/P TX 102V

Input transistor.
C 0V
B 3V
E 3.7V

Output transistor.
C 99V
B 3.7V
E 3.2V

I'm going to change some resistors and check caps as I do so.

[Radio Wrangler]

The circuit is working because the base current of the first transistor manages to drop 3v across that 390k resistor (if it's still 390k!) About 8uA, then

The input impedance of the first transistor won't be anywhere near high enough for a piezo cartridge, hence the 680k // 100pf. They're playing the Burrows trick of using a low Z amplifier with a series termination in front of it. There's an explanation of it in the Bailey-Burrows preamp article in Wireless World from the early '70s.

that 680k is important for equalising the response of the loaded cartridge.

David

[Herald1360]

Originally Posted by Argus25

I think there is a missing DC negative feedback resistor from the emitter of the output transistor to the base of the input transistor, and if there really isn't one, there should be.

That would be positive feedback- both stages are emitter followers in that configuration.

This circuit is deceptively simple!