UK Vintage Radio Repair and Restoration Discussion Forum - View Single Post

Ed_Dinning · 31st Mar 2021, 9:18 pm

Hi Folks, after digesting the above thread and doing a little calculating on my own I believe I can produce suitable transformers that while not being "drop in" should be easy to fit.
To save confusion I have defines as follows:

Amplifier Input transformer, a ferrite device wound on a pot core that is part of a tuned circuit to resonate at 15KHz and reject the 50Hz 1/2 waves thumping through the valve.
Good insulation pri sec and needs "tuning " to the oscillator for best results.
LA1224 is a pair of cores with adjuster and is fitted onto a baseboard PCB. This is a ferrite good to 750KHz and a mid range Al (T/nH) of 295.7. It is 21mm dia.

Working backwards from the measured figures this gives a Pri of 18.34turns of 0.25mm wire and a Sec of 569turns of 0.12mm wire. These give a good fill of the bobbin.
2 layers of type 56 transformer tape is good for more than 2KV.
Turns ratio 31:1

These agree with Davids figures closely.
Other cores can be substituted easily; I may go with the Mullard FX series as I have some of them.
Probably better sticking to a similar material to A13 and the ue of 100, if an increase of size then adjust turns for the same L

Oscillator output transformer, a laminated device about the size of a very small valve portable or TV blocking osc type that is untuned and looks to be a simple design
It is a 2:1 step down ratio
Assuming a max swing of say 4v RMS, something like an EI 35 core (35 * 30mm lam)
(centre limb 9 * 6mm); Stalloy material running at 0.6 Tesla well away from saturation, and an interleaved construction as Dc is blocked by C6.

We get a pri of about 2 turns. Clearly not an ideal figure to wind, so we can either use a much smaller core or we can drop the flux density.
The bobbin has plenty of space so simply go to 200 turns pri and 100 turns sec.
From the circuit constants the primary current will be of the order of 1mA, so 0.15mm wire will give a good fill.
This gives a transformer that is overkill but is easy and cheap to manufacture.

Your comments please folks. I'll then try and get some wound up if anyone wants to try them in their tester. I'll probably do the oscillator output one first as it is the most empirical design

Ed

31st Mar 2021, 9:18 pm	#148
Ed_Dinning Dekatron Join Date: Jan 2004 Location: Newcastle upon Tyne, Tyne & Wear, UK. Posts: 8,195	Re: AVO 163 amp board ~ transformers Hi Folks, after digesting the above thread and doing a little calculating on my own I believe I can produce suitable transformers that while not being "drop in" should be easy to fit. To save confusion I have defines as follows: Amplifier Input transformer, a ferrite device wound on a pot core that is part of a tuned circuit to resonate at 15KHz and reject the 50Hz 1/2 waves thumping through the valve. Good insulation pri sec and needs "tuning " to the oscillator for best results. LA1224 is a pair of cores with adjuster and is fitted onto a baseboard PCB. This is a ferrite good to 750KHz and a mid range Al (T/nH) of 295.7. It is 21mm dia. Working backwards from the measured figures this gives a Pri of 18.34turns of 0.25mm wire and a Sec of 569turns of 0.12mm wire. These give a good fill of the bobbin. 2 layers of type 56 transformer tape is good for more than 2KV. Turns ratio 31:1 These agree with Davids figures closely. Other cores can be substituted easily; I may go with the Mullard FX series as I have some of them. Probably better sticking to a similar material to A13 and the ue of 100, if an increase of size then adjust turns for the same L Oscillator output transformer, a laminated device about the size of a very small valve portable or TV blocking osc type that is untuned and looks to be a simple design It is a 2:1 step down ratio Assuming a max swing of say 4v RMS, something like an EI 35 core (35 * 30mm lam) (centre limb 9 * 6mm); Stalloy material running at 0.6 Tesla well away from saturation, and an interleaved construction as Dc is blocked by C6. We get a pri of about 2 turns. Clearly not an ideal figure to wind, so we can either use a much smaller core or we can drop the flux density. The bobbin has plenty of space so simply go to 200 turns pri and 100 turns sec. From the circuit constants the primary current will be of the order of 1mA, so 0.15mm wire will give a good fill. This gives a transformer that is overkill but is easy and cheap to manufacture. Your comments please folks. I'll then try and get some wound up if anyone wants to try them in their tester. I'll probably do the oscillator output one first as it is the most empirical design Ed