Mystery on my crystal set radio

[Pete_kaye]

I recently built a crystal set then improved it with a one transistor amp into a small computer speaker for easy listening .Is it normal for it to to cut out over most of the band(200m to 2000m)after tuning in to the bottom end where the 2 strongest stations are.It takes about 10 seconds to reset itself as if it has been overloaded.I can speed it up by touching the base of the BC 108 transistor which connects to the +ve end of a 1uF cap.Any advice?

[ppppenguin]

Sounds like there no bias resistor(s) for the base of the BC108 though without seeing the circuit none of us can be sure.

[Ian - G4JQT]

Without the circuit it's hard to say what the problem is, but it should be possible to overcome it. Sounds to me as if the 1uF cap is being charged by the strong signal (from the diode?) and slowly discharging through a high resistance, but without the circuit...

The biasing on the BC108 probably needs looking at to ensure there's a discharge path for the cap.

[hexenfinger]

Is it a 'self powered' one transistor amp? or have you a 3/6V supply? Have you decoupled the input of the amp? Draw and scan the diagram! Let's see! Then it's easy peasy...

[Pete_kaye]

Sorry to take so long to reply ;I couldn't find a working scanner for my sketch

[wireful3]

Pete, It is possible that the crystal diode has unsuitable bias due to the amplifier and coupling condenser. You could try a 500k resistor between the negative side of the condenser and ground.
I have heard of problems with crystal earpieces in crystal sets due to the bias produced by charge on the earpiece cutting off the detector. A resistor to give a return to ground solved this. You may have a similar effect.

[ppppenguin]

It needs a resistor from the diode cathode to ground. This is not for biasing the diode but to provide a discharge path for the capacitor. If there is not a resistor, the rectified signal will charge the capacitor and the circuit will stop working. Since it is an electrolytic, the charge will leak away after a few seconds, hence the recovery time you have observed. 470K is a reasonable value for the resistor.

Strictky speaking you also need a small capacitor from the diode cathode to ground to smooth the half cycles of RF and leave just the audio. In practice, stray capacitance will often do this quite well.

[Pete_kaye]

thanks for the idea ;I will try it when I get home.
I originally had a crystal earpiece which got overloaded one day, even with a 47K R in parallel, and now deceased but the problem was/is just the same in modern headphones and the computer speakers .it can be ~20 seconds to recover.It is speeded up by me earthing the Base with my finger .another oddity is that when my old Ariston fridge cuts in /out then the radio cuts out in a similar way.They are close to each other .I will repeat with a different position too.

[Hunts smoothing bomb]

Hi Pete,

My guess is that your fridge is close enough that when the copressor cuts in, the RF from the spark is that strong that it charges the cap in one hit causing the radio to go silent.

(You have made a crude radio controlled mute facility!)

Cheers Lee

[Pete_kaye]

While I'm at it,is it a good /bad idea to use the mains earth via a 3 pin plug as the radio 'Earth'.this certainly works with this set up .Is it safe? Is it as good as an outdoor type?Does it explain fridge effect mentioned earlier?
my aerials vary but normally i have a horizontal rectangular of multistrand wire about 50 feet around then loft space fed down an empty light switch conduit feed.Sometimes I just stretck about 8 feet of same wire around kitchen and touch it onto the kitchen sink for improvement .can anyone explain why this works as the sink is also earthed.

[Hunts smoothing bomb]

Hi Pete,

I would say that it is not a good idea to use your plug socket earth for two reasons.

1. The length of the earth cable is very long as it goes around the house picking up each socket before returning to the distribution board, this also makes it similar to a second aerial, also the earth wire is right next door to the live conductor and so is picking up a strong 50Hz signal by induction.

2. If you had an earth fault on an appliance which blew the trip or fuse and you were unlucky enough to be touching the radios earth at the same time there is a possibility you could recieve a shock for that millisecond of fault current!

15mm copper pipe driven into the ground for earth and an aerial of 20 feet or so in length (in a straight line) and as high as possible has always worked wonders for me.

Cheers Lee

[wireful3]

Pete,
It is interesting about the kitchen sink. I have not tried this myself but wonder if you are have an effective loop aerial. It seems a bit small however.
I have used an aerial in the loft with the crystal set in an upper floor and the central heating system as earth. I expected the long earth path to cause a problem but performance seemed quite good.
Obviously direct comparison is difficult as any direct earth would also be long. One suggestion is that the piping of the central heating acts as a counterpoise earth.
Crystal sets can be mysterious things!!

[flyingtech55]

The capacitor is too big surely. Would a 0.1uF not be better.

TimR

[Pete_kaye]

Thanks Fellas .putting a 1M resistor in as suggested ,did improve but not eliminate the problem.
I will try the radiator Earth next time. coupled with the sink as aerial .I cannot use an external Earth here but might try up at the allotment sometime .I might even try using my 'greenhouse in progress' as an Aerial connection. -what do you think?

Originally Posted by flyingtech55

The capacitor is too big surely. Would a 0.1uF not be better.

TimR

Yes I thought this and will try it .I was using a circuit given in PW April +May1998 I am using a different diode mind.