Bush DAC90A problems.

[cernem1alt]

Change the Cap I shall
thanks all
Mike M1ALT

[kalee20]

Let us know how you find the change, Mike!

[cernem1alt]

Put in the .001 (1nf) for the existing C17 and I can report the sound quality to my ears is a lot brighter not so deep and much improved so well worth doing.

[kalee20]

Super!

Changing even more, to 390pF would not make such a big improvement with most broadcast stations, although if you ever use a pantry transmitter with a decent treble response, you'd find a noticeable improvement then!

[Argus25]

Originally Posted by Ian - G4JQT

The original value has far too much treble cut for modern ears, and remember that the audio as transmitted now almost certainly has less treble than 60 years ago, so maybe it sounded ok then, but not now.

Yes, often the filtering is far too heavy, but its not the only place in many vintage radios where the filtering gets overdone. By the late 1950's the RC time constant in the AM detector loads in valve radios were typically 100pF and 250k. This allows an upper max frequency response without distortion (and attenuation of high frequencies) of about 10 to 12kHz. Sets from the 1920's are terrible, often a 3meg combined with a 100pF and this gives a very muddy sound in the upper audio register. It is always interesting to look at the values and see how much attention was paid in this area too, but part of it, as you say, is the expectation of "modern ears"

[kalee20]

It's ironic that old radios have crummy treble yet transmissions at the time were not curtailed in the AF upper frequency response. Argos's point of detector filtering is dead right - and with 125kHz IF superhets you do need bigger-valued capacitors to reduce IF breakthrough anyway. Whereas post-war, we have better speakers, 470kHz IF as standard, everything we need for a response to 10kHz-plus... and the transmitters chop off at not much more than 4.5kHz!

[Ian - G4JQT]

The 9kHz channel spacing means the IF should be sharp enough not to allow adjacent-channel 9kHz hets though into the AF sections. During the day there are no adjacent stations stations serving the same area, but at night of course that's different - although not as bad as it was years ago as so many stations are leaving medium wave.

[paulsherwin]

I've always thought that the popularity of the woolly 'mellow bellow' sound in the 40s and 50s was a reaction against strong folk memories of acoustic gramophones and horn speakers. This attitude persisted for a long time amongst that generation. An uncle of mine bought quite a good Sanyo music centre in the 70s but always used it with the treble turned all the way down and the bass turned all the way up. It sounded awful but he loved it - 'listen to those bass fiddles!' he'd say excitedly.

[Argus25]

Originally Posted by kalee20

It's ironic that old radios have crummy treble yet transmissions at the time were not curtailed in the AF upper frequency response. Argos's point of detector filtering is dead right - and with 125kHz IF superhets you do need bigger-valued capacitors to reduce IF breakthrough anyway. Whereas post-war, we have better speakers, 470kHz IF as standard, everything we need for a response to 10kHz-plus... and the transmitters chop off at not much more than 4.5kHz!

Strangely, I couldn't find a readily available formula that determined the the upper audio frequency response that an AM detector could support without peak distortion. I set about calculating it and it depends on the extent of the modulation of course. With two methods I came up with, they gave 0.5/RC and 0.32/RC. However there is another way relating to detector AC and DC impedances outlined in Terman, where a formula can be resolved from his remarks; for 50% modulation it is 0.275/RC, which gives a more conservative result. I wrote an article about this a while ago. Also, out of interest, it is explained in the article what happens when the grid leak return is altered to either side of the filament, with an equivalent model of a 201A:

http://www.worldphaco.com/uploads/Ra...he_1920_s..pdf

[crusher19860138]

I replaced my 0.003uf cap with another as the sound to my ears is fine.

[cernem1alt]

Work done so far usual Caps changed new bulbs, perished wiring replaced checked my soldering all ok

Now after a while the set goes quiet where should I look? I am thinking resistors

However quick flick of the waveband switch jumps it back for a while


Mike M1ALT
"Live Chassis precautions observed "

[kalee20]

What about a slow flick of the waveband switch?

A quick flick could shock something elsewhere on the chassis which is dodgy, into recovering its contact. A slow flick (if that's not a contradiction in terms!) is unlikely to do so, except the contacts of the switch itself.

Does this happen when left on LW and on MW, or just one waveband (which?)?

[cernem1alt]

Slow flick or quick flick makes no difference, jumps it back, it is on LW where this happens.

[ukcol]

The 110pF capacitors in the IF transformers can give this symptom. I have had it with a DAC10 and heard reports of it with the DAC90A.

[Boater Sam]

Is there any screened lead around the volume control or top cap connectors that is perished inside and breaking down? Had this recently with a Murphy 104, same symptoms.

[cernem1alt]

Re wired C17 tone cap and C22 replaced wires with better screened now has case off sat on bench behaving itself listening to Radio 4 LW so we shall see.......

[cernem1alt]

Problem solved Me thinks ......a dry solder joint that is going open circuit as the set warms up and then remakes again once the set has cooled down

[cernem1alt]

On the DAC90A T1 transformer which terminals are the primary and which the secondary? am I correct in thinking that secondary wire to the speaker or am I confused,

thanks Mike

Yes, secondary to 'speaker. Even if you do get it the wrong way round all that will happen is a very low output.

[Ian - G4JQT]

The speaker (secondary) side has a much lower resistance than the primary, in case you get in a muddle!