Home brew TRF

[Hunts smoothing bomb]

Hi everyone
I have recently built a 3 valve mains powered TRF medium wave reciever.
Both HT and LT supplies are derived via an HT/LT transformer, the HT using resistor/capacitor smoothing.
After testing, the set was found to perform very well. However after a few days of half an hours operation each day the set is developing serious mains hum.
The HT secondary on the mains transformer is of the centre tapped variety with an ac voltage of 270-0-270 (no load) the centre connection going to ground. The other two leads each have a 1n4001 rectifier in them which then go into a 47mF 450v smoothing capacitor.
My question is this:-
I have no data on the 1n4001 but my suspicions are that they are breaking down due to the reverse voltage against them, can anyone confirm this and tell me what diode is required and calculate the reverse voltage against them?
Many thanks

Lee

[Ed_Dinning]

Hi Lee, you surmise correctly. A 1N4001 is only rated at about 100V PIV, you really need a 1n4006 or 7, which has a more suitable PIV. You can find the data sheets via Google. It would be good practice to fit a fuse in the 0V lead from the transformer, 100mA or so should be fine. This will protect the transformer and capacitors if you have any further problems.
If you have a 270 volt output this will peak to root2 times as much (382V), and as there are 2 windings in anti-phase the PIV per diode will be twice that, 764 volts. Best to use a 1000 v device in this position.
What sort of set is it, or what is the design based on?

Best regards, Ed dinning

[mickjjo]

I think the 1N4001 is only rated at 50v , so it's not surprising you are having trouble. Use the 1N4007 instead, cheap as chips..... .

Regards, Mick.

[Biggles]

Hi, sometimes connecting a small value ceramic capacitor across (in parallel with) each diode helps. Be aware that the voltage rating on the capacitor should be more than twice the peak voltage available from the transformer winding. The value of the capacitor should be no more than a few thousand pFs. I've seen this done in commercial equipment and seems to cure the problem of "harsh" switching waveforms found with solid state rectifiers. I've just realised something else. Are you using a 1N4001? This is only rated at 50V PIV. Use 1N4007's instead, the poor 1N4001's will be under severe risk of destruction! The set will be dangerous to use in it's current state. Why not use a valve rectifier?!! Hope this helps. Whoops someone has beat me too it.

Biggles.

[Hunts smoothing bomb]

Hi everyone,

Thanks for your help, I had a sneaky suspicion it would be the diodes, I just wanted to double check.
For those interested the set is of very basic design. I havn't done alot of constructing so thought I'd start with something simple. The design is based on bits and pieces I have picked up in books along the way. It is a single tuned circuit type with reaction coil. The reaction is controled via altering the supply voltage of valve 1's anode which is a 6j5 triode this output is then fed into another 6j5 (audio amp) and then is past to a 6V6GT (audio output) I have also incorporated a 5 step tone control using a 5 way wafer switch which simply connects different value caps from the 6V6's anode to chassis it all works quite well. I would have used a valve rectifier but this set I wanted to make as small as possible so there is no room on the chassis, I am trying to recreate the small midget set look, this is why I have used octal valves.
I will post some pictures soon if anyone is interested.

Bye everyone

Lee

[paulsherwin]

As others have said, if you really used 1N4001s you're lucky to still have a mains transformer. Ed is right about the reverse voltage so you will need 800V rated diodes, not 50V If you only have 400V rated components (e.g. 1N4004s) you can use 2 in series. The first will be loaded by the forward voltage and the second by the reverse voltage.

High voltage rectifier diodes cost next to nothing, so it's worth stocking up:

http://cgi.ebay.co.uk/Diode-Rectifie...QQcmdZViewItem

Silicon diodes generally fail short circuit so it's a very bad idea to push your luck.

Best regards, Paul

[Hunts smoothing bomb]

Thanks Paul,

Rest assured, this set will not be energized again untill the diodes have been replaced with 4007's

Regards
Lee

[Biggles]

Keep up the constructional work, it's nice to know I'm not the only one who makes "old" things out of bits and pieces found lying around in the shed!

Biggles

[Hunts smoothing bomb]

Hello peeps,

Just bought 100 1n4007's off Ebay for £1.50 Inc postage
Thanks to Paul for that , just seen his link in another post

Cheers

Lee

[ukcol]

Originally Posted by Ed_Dinning

What sort of set is it, or what is the design based on?

Hi Lee

I suspect one of the things that might have been in Ed's mind when he asked the above question is what HT voltage is the radio designed to run from. Using two silicon diodes from a 270-0-270 transformer is likely to give an HT voltage too high for a 3 valve TRF.

Sometimes restorers will use silicon diodes (hidden away) to replace a no longer available rectifier valve; when this is done they use a wire wound resistor in series with each diode to make a better job of simulating a valve rectifier. I would suggest you check what the HT voltage should be and experiment with the values of series resitors until you get it about right. Each resistor should be the same value of course; start with 100 ohms and work up as required.

[octode]

Your HT voltage is way to high for a TRF. Most regenerative detectors operated with an HT of about 60volts or less (the output stage can operate at the usual voltages tho), you might find that you have difficulty controlling the reaction or get microphonics and other undesirables.

Remember that too much reaction can turn your receiver into an effective transmitter and that the humble 6J5 can run upto 2.5 watts input!.

As for controlling the reaction by dropping the HT, I don't think thats a scheme usually used even in very early sets. With triode detectors, the usual method is either to use a small adjustable link winding over the main tuned winding, or via a small (say 150pf) variable from the tuned winding to anode.

The majority of 30's TRFs used a screened grid (tetrode) or small RF pentode for the detector, the screen voltage being varied to control the gain.

You really don't need to use reaction at all these days, one decent stage of RF amplification followed by a detector and output stage is all that is needed. With broadcast stations putting out 10's (or even 100's) of Kilowatts, you really don't have any difficulty receiving them even if you've got a mediocre aerial.

On the plus side, TRFs are great fun, they are simple to work on and can work well plus the sound quality can be superb (at least I think so) in a well designed set.

Good luck with your project

[Hunts smoothing bomb]

Thanks for all your support guys,

I must clarify though, I think there has been some confusion over operating voltages. Smoothed HT is around 325v this then passes through a 1k wire wound resistor then into a second 47mF 450v can, this drops the voltage to the correct levels for the 6V6. HT is further dropped again through several high value resistors before going through a potential divider pot, (reaction control) the wiper of this pot is then connected through the reaction coil and then straight to valve 1's anode. Believe it or not, measuring HT from valve 1 anode to chassis is only around 30v. Give it anymore and the set will go into oscillation This particuler part of the circuit came from an old 50's book I have, It's called "Radio For Boys" and is written by E.N. Bradley, some of you out there may know of it.
Here are some pics for thoughs interested.

Regards Lee

[Biggles]

You've made a nice job there. I have tried the single valve circuits from this book a couple of times without much success (probably my lack of perseverance) so I tend to go for superhets as I've always had good results with them although they are a bit more fiddly to align etc.

Biggles.

[paulsherwin]

Very neat job Lee. It certainly puts my efforts to shame.

Best regards, Paul

[Ed_Dinning]

Hi Lee, what did you do for tuning coils?

Ed

[Biggles]

Isn't that a coil attached to the side of the variable capacitor Ed?

Biggles

[Norm_Leal]

Hi Lee

Very nice job. You used ST 6J5's. These sell for more than other shaped valves.

Manufacturers don't try to make 1N4001 diodes. They go for the highest PIV possible then mark parts as to breakdown. You were lucky. When there is a demand for 1N4001 higher PIV diodes are just marked with this number.

When selecting diodes at least double the peak AC voltage. Since your transformer is 270 volts you need diodes rated greater than 764 volts. (270 x 2.828)

Norm

[Hunts smoothing bomb]

Hello guys (and gals)

The tuning coil is mass produced, don't ask me where to get this type though because last year I took shed load of NOS components from a friend of mine who aquired them from the science department of a school.
The coil has one centre tap which I havn't used. I had to wind my own reaction winding though this was done by laying the windings straight on top of the tuning coil.
I have built one of these circuits before and what I used then was 65 turns of thin enamelled copper wire on a toilet roll tube with a paper sleeve around it, tight enough so it would just slide up and down. I then wound 25 turns of the same wire on the sleeve. after connecting it all up you can slide the sleeve up and down untill you get the best reception and then seal it with glue. I had very good results indeed with this type of coil.
I will post a circuit diagram of the set I have built tomorrow so if anyone wants to build it and give feedback on how it can be improved then they can...

Buy for now

Lee

[Hunts smoothing bomb]

Here we go.

The circuit.

The coil can be constructed as mentioned earlier.

Regards Lee

[quantum]

The 0.01mf value cap connected to the grid of the first 6J5 seems rather high for a leaky grid detector - should it be 0.0001mf (100pf)? How is the selectivity - I'd imagine that with a long aerial you would have difficulty in separating the stations, unless you have an extra winding on the coil for the aerial and/or a small value cap in the aerial lead.