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

beery · 12th Sep 2017, 8:04 pm

Hi all,

I've been having a go at making a substitute for the 4TP triode pentode valve that is used in the frame timebase of the Cossor 54. The orignal valve is made from unobtainium.
Luckily my set does actually have a good 4TP in it, so I have something to compare with.

David Boynes has looked into this before and found that an ECL80 is a good replacement. The problem is how to get 6.3 volts for the heater. Also, if the valve is not warm before the EHT rectifier comes on it would cause a flashover...

I decided to wind a transformer that would fit in a 7 pin valve base...

Now the ECL80 requires 6.3V at 0.3A which would mean a 1.89VA transformer which would be too big to fit in the base. However we don't need isoloation, we simply need to step up the 4V heater supply to 6.3V, so an auto transformer will do.
Oh the joy of the autotransformer... The transformer only needs a VA rating based on the voltage difference required. 6.3V - 4V = 2.3V. 2.3V x 0.3A = 0.69VA, which makes for a small enough transformer.
Note that 6.3A at 0.3A equates to 1.89VA. 1.89/4V = 0.4725A at the 4V input to the transformer. Of this 0.3A will pass into the secondary and the remainder 0f 0.1725A forms the primary current. Of course the transformer is not 100% efficient, so the primary current is in the order of 0.2A.

The laminations and the bobbin were obtained from a scrap GPO 746 dial telephone. I experimented with different ratios to work out the primay turns per volt which worked out to be about 60 (I can't remember the exact figure).
That means 4 x 60 = 240 turns for the primary. The secondary is 2.3 x 60 x 1.1 loss adjustment = 152 turns. In the end I had add more seconday turns because such small transformers are not very efficient.

The transformer then had a pancake style B9A socket attached to it and then some flying leads to enable it to be mounted into a B7 base.

The result was, it worked!!

I'm now planning to hide it in an old valve envolope...

The first picture shows the 4TP in situ.
The second shows the ecl80 sitting on the auto transformer.
The third shows the completed valve and adaptor.
The fourth picture shows the valve in situ.
The fifth picture shows the result. Note that I can't easily adjust the height as in this set it is done via resistor substitution. The poor resolution is due to a gassy CRT and the fact that the MVS-PENB sound IF valve was having a fit.

Cheers
Andy

12th Sep 2017, 8:04 pm	#1
beery Heptode Join Date: Jun 2006 Location: Ware, Hertfordshire, UK. Posts: 988	Cossor 54 4TP substitution Hi all, I've been having a go at making a substitute for the 4TP triode pentode valve that is used in the frame timebase of the Cossor 54. The orignal valve is made from unobtainium. Luckily my set does actually have a good 4TP in it, so I have something to compare with. David Boynes has looked into this before and found that an ECL80 is a good replacement. The problem is how to get 6.3 volts for the heater. Also, if the valve is not warm before the EHT rectifier comes on it would cause a flashover... I decided to wind a transformer that would fit in a 7 pin valve base... Now the ECL80 requires 6.3V at 0.3A which would mean a 1.89VA transformer which would be too big to fit in the base. However we don't need isoloation, we simply need to step up the 4V heater supply to 6.3V, so an auto transformer will do. Oh the joy of the autotransformer... The transformer only needs a VA rating based on the voltage difference required. 6.3V - 4V = 2.3V. 2.3V x 0.3A = 0.69VA, which makes for a small enough transformer. Note that 6.3A at 0.3A equates to 1.89VA. 1.89/4V = 0.4725A at the 4V input to the transformer. Of this 0.3A will pass into the secondary and the remainder 0f 0.1725A forms the primary current. Of course the transformer is not 100% efficient, so the primary current is in the order of 0.2A. The laminations and the bobbin were obtained from a scrap GPO 746 dial telephone. I experimented with different ratios to work out the primay turns per volt which worked out to be about 60 (I can't remember the exact figure). That means 4 x 60 = 240 turns for the primary. The secondary is 2.3 x 60 x 1.1 loss adjustment = 152 turns. In the end I had add more seconday turns because such small transformers are not very efficient. The transformer then had a pancake style B9A socket attached to it and then some flying leads to enable it to be mounted into a B7 base. The result was, it worked!! I'm now planning to hide it in an old valve envolope... The first picture shows the 4TP in situ. The second shows the ecl80 sitting on the auto transformer. The third shows the completed valve and adaptor. The fourth picture shows the valve in situ. The fifth picture shows the result. Note that I can't easily adjust the height as in this set it is done via resistor substitution. The poor resolution is due to a gassy CRT and the fact that the MVS-PENB sound IF valve was having a fit. Cheers Andy Attached Thumbnails __________________ www.youtube.com/user/andyvalve100