25-11-2017, 11:58 AM
(This post was last modified: 25-11-2017, 12:01 PM by Murphyv310.)
Hi Mike.
Yes indeed the coil can be across the HT especially as it is a high resistance type. It actually has some distinct advantages, it loads the HT so you don't get a huge rise in HT if the rectifier is a directly heated type, especially with indirectly heated subsequent valves. It is also easier to get the HT smoothed with a separate HT choke, so in fact a not a bad idea with the proviso that with all that wire it could be more prone to failure.
Re the intervalve TX, are you 100% its not been modified, that's the way I did a couple but some sets need a bit more drive, adding a FET will suffice. The HT can be dropped by say a 220k, decoupled to chassis via 2.2uf 300v, gate biassed to chassis with a 1 meg, signal fed via a 0.02uf Source to chassis via a 2.2k and the drain to the 220k via a 15k with audio taken from here via a 0.01uf. The grid of the output would need to be fed from the 0.01uf and also a 470k that returns to the bias feed. All this can be fitted in an intervalve TX case and one of the ferrules for mounting can be used as a common chassis connection.
Sorry I forgot to draw the circuit. A 2N3819 will do fine.
This pinched off the net gives a rough idea, https://tapeop.com/_m/photologue/photos/JFET-f1-new.jpg
I'd stick roughly to my values though.
Yes indeed the coil can be across the HT especially as it is a high resistance type. It actually has some distinct advantages, it loads the HT so you don't get a huge rise in HT if the rectifier is a directly heated type, especially with indirectly heated subsequent valves. It is also easier to get the HT smoothed with a separate HT choke, so in fact a not a bad idea with the proviso that with all that wire it could be more prone to failure.
Re the intervalve TX, are you 100% its not been modified, that's the way I did a couple but some sets need a bit more drive, adding a FET will suffice. The HT can be dropped by say a 220k, decoupled to chassis via 2.2uf 300v, gate biassed to chassis with a 1 meg, signal fed via a 0.02uf Source to chassis via a 2.2k and the drain to the 220k via a 15k with audio taken from here via a 0.01uf. The grid of the output would need to be fed from the 0.01uf and also a 470k that returns to the bias feed. All this can be fitted in an intervalve TX case and one of the ferrules for mounting can be used as a common chassis connection.
Sorry I forgot to draw the circuit. A 2N3819 will do fine.
This pinched off the net gives a rough idea, https://tapeop.com/_m/photologue/photos/JFET-f1-new.jpg
I'd stick roughly to my values though.
