Yet another Bush DAC90A

[ms660]

How would I do that?

If there's to much oscillator signal on g1 of the mixer that prevents you seeing the RF on g1 then kill the oscillator by shorting out the oscillator section of the tuning gang.

LO (sometimes aka Lo) = Local Oscillator.

Lawrence.

[Mike. Watterson]

i.e. a short wire with croc clips from tag on fixed vane to chassis.
However likely you don't need to do that.
It is though amazing how much L.O. leakage there is on heptodes & octodes RF in grid. Usually the Triode - Hexodes (which a DAC90a uses) are not so bad.

[G8UWM-MildMartin]

Would connecting the 'scope to the RF tuning capacitor with the set unplugged from the mains work?

[Mike. Watterson]

Probably as there is no RF stage.

Except then you aren't hearing what is happening, hence NOT shorting out the L.O. by default.

[DonaldStott]

So tonight the first thing I did was to check a couple of other radios to confirm that the interference wasn't coming from the DAC90A itself. It was audible to a greater or lesser extent on the three radios I tried so that proved that it was probably somewhere in the house.

Went round systematically and unplugged all likely suspects - TV, PC, laptop, Wi-fi router, NAS, network switch, mobile phone chargers, hi-fi and anything that could have an SMPS etc. etc - but no luck.

Powered up a Roberts R303 battery powered radio to try and locate the problem but again nothing obvious.

Had a think about anything new that we had recently acquired or anything that hadn't been turned on for a while - at that point I remembered that I had been copying some photos from a PC external drive that I use for back-ups but normally this is turned off. Surely it couldn't be that but it was still switched on (!) and it does have it's own 12V 5A power supply: -



The WARNING message amused me!

And of course we have located the culprit with the interference getting louder as we brought the Roberts radio in close proximity.

Couldn't resist a YouTube video starring a Roberts R303 in olive green and behind it the power supply for the external hard drive - apologies for the shaky camera work. It starts off with Radio 4 Long Wave (even in Scotland some of us really do like Cricket) and the power supply is turned on about 7-8 seconds in - WOW! At about 30 seconds in the power supply is turned off and it goes out fighting with a screech and a scream ...!

https://www.youtube.com/watch?v=_qPGYknIPmU

So there we have it, drama over and a noisy cheap imported power supply found guilty - a lesson for us all there.

An enormous thanks to everyone who contributed to resolving this issue and I am so glad we reached a successful outcome.

I'll attempt to get this Thread back on track tomorrow as the DAC90A is sounding pretty good - still some hash between stations but nothing out of the ordinary.

P.S. The Roberts R303 is on the ToDo list...

[Ian - G4JQT]

Well done. That's one hell of a SMPS there! I don't think I've come across one that made such a regular noise. You live and learn...

[G8UWM-MildMartin]

A no-name "Replacement AC Adapter" can sometimes be found guilty of more lethal crimes than producing RF noise.
Glad you found it, and hope you give it the appropriate punishment!

[DonaldStott]

Glad you found it, and hope you give it the appropriate punishment!

I was thinking more of entering it for one of the Forum Annual Awards, for example, "Top 10 Offenders" - is there a Prize?

[DonaldStott]

After the recent "Son of Woodpecker " diversion (thanks Lawrence) it's now time to get this Thread back on track ...

The Dropper is still running very hot but it may be that I'm just being over-cautious.

Most of my questions in Post #49 above have been answered but still not sure about the diffuser which was attached with some sort of glue that I would like to remove - recommendations please for an appropriate solvent that won't attack the diffuser itself e.g. IPA, Acetone, Lighter Fluid?

In Post #68 above kalee22 advised that I should replace the 0.003μF capacitor (C17) from UBC41 pin 2 with a 390pF or 470pF to make "the sound brighter and less muffled!" In the interest of my own education I'm not sure why replacing the I.F. By-pass capacitor with one of lower capacitance would help - I don't doubt that it would just interested in the reason why? My instinct would have been to replace C20 the Fixed tone corrector capacitor across the Primary of the Output Transformer?

In Post #69 above I proposed possibly replacing the wire wound resistors -any thoughts?

[ms660]

In Post #68 above kalee22 advised that I should replace the 0.003μF capacitor (C17) from UBC41 pin 2 with a 390pF or 470pF to make "the sound brighter and less muffled!" In the interest of my own education I'm not sure why replacing the I.F. By-pass capacitor with one of lower capacitance would help - I don't doubt that it would just interested in the reason why?

Think capacitive reactance at higher audio frequencies:

https://www.electronics2000.co.uk/ca...calculator.php

Lawrence.

[Mike. Watterson]

The 470pF is more usual there, the 3nF sounds like a rather large value.
I don't like changing designs, I have the attitude that restore/repair should put it back the way it was originally, not try to change it
The "tone correction" caps on primary of audio transformer, or sometimes the O/P valve anode to ground are actually chosen to match the characteristics of a transformer. They are really snubbers, though do reduce HF response, usually to what it should be. Without them, or if too low in value, any transient noise from interference, crackly volume control, wavechange or tuning capacitor creates large voltage spikes that may even flash over on Anode to "g3" deflector plates or g2.
So certainly the only places to mess with frequency response are the preamp stage, g1 area on O/P valve and detector.
A schematic I have here has 300pF on a 6AT6 anode (preamp feeding O/P stage via 4nF) and 100pF plus 100pF to the volume control from the final IFT (the other side to detected by the diode in the 6AT6). It's common in valve sets for final IFT secondary to be in series between volume control and detector, but on transistor sets for it to be "earthed" and detector diode feed volume control.
The ripple of the IF has to be filtered, usually before volume control and also additionally at output of the preamp. Otherwise gain will suffer and you'll have IF feedback / motorboating and distortion. Too large a value and the higher audio frequencies suffer.

[Ian - G4JQT]

I can confirm that reducing C17 from 3nF to 470pF is well worth doing. Thirty years ago I realised the DAC90A and DAC10 were quite lacking in treble and it was not due to the IF bandwidth. It didn't take long to realise the capacitor in question here is too high a value, not only removing any IF, but also the higher audio frequencies.

No idea why Bush did this and continued exactly the same with the DAC10.

I also reduce the tone correction capacitor too on some sets - though recognise what Mike W says.

I suspect the more limited audio bandwidth of modern AM transmissions sometimes make some of the original capacitor values in the post-detector stages less appropriate than they were in the distant past.

[ms660]

Some circuit designers were quite clever back in the day, sometimes what seems like a dumb idea might not be such a dumb idea, I can't speak for the Bush DAC90a or any other particular model with a UBC41/UL41 sliding screen output valve combo but Vin at the AF amp versus power out (50mW ref.) seems to figure in the valve manufactures requirements to minimize microphony without further modification.

Lawrence.

[DonaldStott]

Originally Posted by Mike. Watterson

The 470pF is more usual there, the 3nF sounds like a rather large value.

Originally Posted by Ian - G4JQT

I can confirm that reducing C17 from 3nF to 470pF is well worth doing.

Quite a large drop in value but I'll give it a go if I can source a suitable 470pF (0.00047µF) capacitor - will a sufficiently high voltage ceramic capacitor be o.k. in this location?

[Mike. Watterson]

Yes, 250V or more ceramic is fine.
The blue 1KV disk ceramics are cheap. The value/tolerance isn't critical.

[kalee20]

Originally Posted by DonaldStott

In Post #68 above kalee22 advised that I should replace the 0.003μF capacitor (C17) from UBC41 pin 2 with a 390pF or 470pF to make "the sound brighter and less muffled!" In the interest of my own education I'm not sure why replacing the I.F. By-pass capacitor with one of lower capacitance would help - I don't doubt that it would just interested in the reason why? My instinct would have been to replace C20 the Fixed tone corrector capacitor across the Primary of the Output Transformer?

Good question! The 0.003μF capacitor is indeed there to bypass the IF.

It just happens that this value is so big that it bypasses some of the higher audio frequencies too, so you lose some treble. Changing for a smaller value means you keep your treble, but the IF is still bypassed effectively.

Your instinct to replace the tone corrector is not misplaced, you can do that too, I would suggest 0.005μF (make sure it is a high voltage type, as if the volume control is well up you can get 500V across the transformer primary). I'm my own DAC90a, I use 0.0068μF in series with 4.7kΩ, but this radio has several mods to it!

[Mike. Watterson]

I'd leave tone-correctors at original value as they are often chosen to match the transformer and are often snubbers, not really there for "tone".

[kalee20]

They are. But I spent quite a lot of time looking at the DAC90a design, the 6800pF and 4.7k does give a fairly flat frequency response, the L,R of the loudspeaker and the C,R of these values giving a constant resistive load and cancelling the HF lift break point. (The low-frequency resonance is still there of course, no better, no worse).

[Mike. Watterson]

What happens with a pulse input?
No ringing on a square wave?

Certainly there are few models where the design can't be improved. Maybe if I had two, I'd have one original and one optimal?

[DonaldStott]

Almost there - most of my questions in Post #49 above have been answered but still not sure about the diffuser which was attached with some sort of glue that I would like to remove - recommendations please for an appropriate solvent that won't attack the diffuser itself e.g. IPA, Acetone, Lighter Fluid?

Two more small questions: -

1. Where exactly is the tuning capacitor earth spring - is it located between the two sets of vanes?

2. What should be used to lubricate the tuning capacitor ball bearings - I normally use trombone slide oil sparingly on shafts, bearings etc. as it penetrates really well - a tip from Forum member boxdoctor (Tony).