Hum when old BSR connected to new op-amp/speaker circuit

[indigo.girl]

Hi everyone,
I have acquired an old Sobell radiogram - the valves and receiver are not working but the record deck inside spins and the cart still works (Its a turnover TC8M and I've tested it in another of my record players). I've removed the record deck and have wired it up to modern op-amp circuit and speaker so it will play again. I get audio from the speaker but there is a tremendous hum that drowns out the sound. Is there something I can add on to my circuit to stop the hum?

I have used this kit from Maplins and wired the two wires connected to the cart directly into the input positions (I have just wired up one of the two speakers for simplicity at the moment).

http://www.maplin.co.uk/p/kitronik-d...pod-lead-a55rj

Why do I get the hum? This kit is supposed to be used for amplifying and playing audio from an ipod. The autochanger deck is metal. Do I need some kind of shielding? I have used a 12V DC mains adapter (1500 mA output) to power the amp/speakers and wired the motor direct to a mains plug. Are these causing interference? The speaker is 0.5W / 8ohm.

I'm not expecting great sound from this set up - I just wanted to have a play around at seeing if I can get an old deck running again rather simply use it for spares. Any ideas much appreciated . Thanks, Nicola

[Radio tirana]

HI
Just a thought are you definitely connecting to one of the inputs and ground? or could you have wired your cartridge to the Left and right inputs with no ground connection.
My eyesight is not good enough to read any silk screened letters on the board but this could explain the hum.
Kind regards RT

[peter_sol]

The cartridge needs 500k ohms at least. The input impedance of that amp is 20K or less.
To start with you need to put a 470K resistor in series (in line) with the input of that amp.
one side of the cartridge should be connected to the deckplate via the tag strip underneath
and that should go to the screen of the amp input cable the other cartridge wire to the inner of the screened cable. look at your other decks to see if you are not sure.
Do you get the hum without the deck connected?

[indigo.girl]

Originally Posted by Radio tirana

Just a thought are you definitely connecting to one of the inputs and ground? or could you have wired your cartridge to the Left and right inputs with no ground connection.

I think that bit's OK. The black wire is connected to ground and the red wire is connected to channel 1 (channel 2 & speaker 2 is left empty) The white wire from the cable is disconnected at both ends and surplus for this set up.

[peter_sol]

You have not answered my post.

[indigo.girl]

Originally Posted by peter_sol

The cartridge needs 500k ohms at least. The input impedance of that amp is 20K or less.
To start with you need to put a 470K resistor in series (in line) with the input of that amp.
one side of the cartridge should be connected to the deckplate via the tag strip underneath
and that should go to the screen of the amp input cable the other cartridge wire to the inner of the screened cable. look at your other decks to see if you are not sure.
Do you get the hum without the deck connected?

Thanks Peter - have been testing out what you suggested...
I have connected the black wire to the deckplate via the tag strip and then to the screen position at the amp input. This has worked wonders and I get much louder audio that is audible above the hum. The hum is still there (albeit reduced) and ideally I need to reduce this further. However I have noticed that the hum tends to increase or decrease depending on how I position or hold the components (the PCB sitting on the carpet is worse than if I hold is suspended by a wire).

I have wired in a resistor in series with the red wire going into channel one of the amp input. The only one I had to hand is 47Kohm so much bigger than you suggested. This doesn't seem to make much difference with or without the resistor now I've grounded the black wire.

With the record deck disconnected the amp/speakers alone do not hum but do buzz when I touch the red wire.

[Herald1360]

Errr 47K is actually much smaller than 470K, which would explain the little difference. 470K should improve bass response at the expense of overall level.

Is there any ground to the amplifier itself in addition to the deck ground? This could make a hum loop.

[bikerhifinut]

Have you tried paralleling the inputs and playing the sound through both loudspeakers?
I'm thinking a couple of things here, firstly it should increase the perceived output from that little chip amp and secondly, i wonder if having an open circuit input on the other audio channel might be somehow causing the hum by coupling into the adjacent channel?

Peter was suggesting using a 470k resistor in series, this is to present a load resistance to the cartridge in keeping with its needs. I doubt it would alter any hum but you'll get a better match into your amp. Probably only noticeable if you used a bigger speaker with some sort of bass response?
However as it couples directly onto the top of the volume pot I wonder if you're getting a shunt effect as with even just a 47k resistor in series with the potentiometer, at full volume you'll still have a 5:1 potential divider as going by the circuit diagram you showed the pot is only a 10k device. i imagine its a fairly low sensitivity amp as its meant for use with a headphone output from an Ipod type thing?



Nice idea though, I like the idea of making a record player with a spare deck with a cheap and effective amp.

A.

[bikerhifinut]

Whats the input impedance of that TDA822 chip?
If its high enough then to me the solution is to use a 500k pot and remove the existing 10k one. You only need a single not a dual and parallel the inputs of the amp and use a pair of speakers to get the best volume out of it.
Simply sticking a large value resistor on the top of the potential divider is only going to reduce the available signal to the amp.
A.

[indigo.girl]

Originally Posted by Herald1360

Errr 47K is actually much smaller than 470K, which would explain the little difference. 470K should improve bass response at the expense of overall level.

Is there any ground to the amplifier itself in addition to the deck ground? This could make a hum loop.

Whoops - I misread Peters resistor value! Thanks for spotting that There's no ground from the amp I think....

[indigo.girl]

Originally Posted by bikerhifinut

Whats the input impedance of that TDA822 chip?
If its high enough then to me the solution is to use a 500k pot and remove the existing 10k one. You only need a single not a dual and parallel the inputs of the amp and use a pair of speakers to get the best volume out of it.
Simply sticking a large value resistor on the top of the potential divider is only going to reduce the available signal to the amp.

Thanks bikerhifinut, I was thinking of paralleling the speakers at the end. I'll do that in the morning. It seems the resistance is important so I'll connect a higher value tomorrow and see how that goes. If it improves things I like the idea of putting a 500K pot in to replace the 10K. I'll report back tomorrow

['LIVEWIRE?']

The speakers should, imho, not be paralleledDoing this may damage the amplifier due to combining the outputs of the two channels. Sending a mono signal to the LH & RH inputs should be OK, though two resistors, one from each input to the signal lead from the pickup, may be needed. Swapping the exsisting volume control for a 470k or 500k one, as 'bikerhifinut' suggests, is also a good idea.

[bikerhifinut]

Yes I think Indigogirl may have misunderstood me, I did mean paralleling the INPUTS and using 2 separate loudspeakers which seem to be supplied with that little kit anyway.
Using a couple of buffering resistors certainly won't do any harm and might be beneficial as you say.
Well spotted Livewire

Andy.

[indigo.girl]

Originally Posted by 'LIVEWIRE?'

The speakers should, imho, not be paralleledDoing this may damage the amplifier due to combining the outputs of the two channels. Sending a mono signal to the LH & RH inputs should be OK, though two resistors, one from each input to the signal lead from the pickup, may be needed. Swapping the exsisting volume control for a 470k or 500k one, as 'bikerhifinut' suggests, is also a good idea.

Is it OK to leave the input channel 2 and speaker 2 unconnected like I have now? Or could it cause interference?

[bikerhifinut]

A bit off piste but I couldnt help wondering what the problem is with the original radio and amp circuit in the radiogram?
If it's trashed beyond reasonable repair then I can understand removing the turntable although I cant help thinking it might be cost effective to get a mono IC amp kit and retrofit it into the radiogram, maybe with a simple little personal radio receiver as well. Use the original loudspeaker. Then use a couple of low power lamps/led's to backlight the original tuning scale.
Just the musings of a madman..............

A.

[bikerhifinut]

Originally Posted by indigo.girl

Is it OK to leave the input channel 2 and speaker 2 unconnected like I have now? Or could it cause interference?

Its not going to do any harm, but if you do that I'd short out the unused input to kill any possible hum induced into the amplifier circuit. You're ok leaving the output open circuit.
But i feel its a waste to not use the other little speaker and get a bit more volume out of it. I dont know if its possible to get adaptors for those turnover headshells but if you could get a stereo ceramic cartridge in there you have the makings of a cheerful stereo record player that will likely sound better than the tat seen advertised in the Sunday papers? I dont know what the output of that little amp is but it would drive a pair of larger speaker units just as well and its just possible you'd get a decent SPL out of them.

Andy.

[peter_sol]

If you look at your picture 1 in post 1
You are connected wrongly at the amps input, and are you using screened cable?
You should be using screened cable with the screen to the centre pin (or hole) and the inner of the screened cable to one of the outside pins or holes depending on which channel you are using. The chips input impedance is 100K so with that circuit having 10K pot with 10K in series with it brings the total down to just under 20K so still need 470K in series to avoid shunting the pickup..

[roffe]

This amplifier has no power supply cap(!) and if fed by an adapter that puts out a not so pure DC,it hums.
Try another adapter or batteries
or simply solder a 1000uF/15V cap across + and -.

[Edward Huggins]

There has been some suggestion of using the original speaker in the Sobell Radiogram. Although we don't have the model number, it is likely to be 3 ohms and not suitable to be run off that Chip. Edward

[indigo.girl]

Some excellent comments here guys - thank you. I'll try some out today and report back later. I'm going to wire in the second speaker, connect channel 1 and channel 2 together so both speakers get the mono signal and add a capacitor across the power input....

Peter - I am using the wire supplied by the kit (it came with a 30 pin ipod connector that I've removed) - is this not suitable? What should I use instead. The old radiogram came with a wire encased in a metal mesh (the mesh being connected to ground and the inside wire connected to the input channel). Should I use this instead even though its old? Might be better than what I have used.

Thanks
Nicola