Variable Mu Valve in Regenerative Receiver - UK Vintage Radio Repair and Restoration Discussion Forum

ionburn · 15th Jan 2019, 4:32 pm

I recently came across a post elsewhere in which someone was intending to use an EF183 valve in a regenerative receiver.

It is something I feel I should be able to work out, but have never really come across it (maybe because the bulk of regenerative receivers were produced before variable mu valves became common), and people often use simplar valves at very low voltages.

What are thoughts the possible pros and cons of using a variable mu valve in a regenerative receiver (ie, say using one for a one valve regenerative set)?

I did ask the question when I saw the post but no-one there has yet been particularly forthcoming, so maybe something for me to try but it would be good to find if there are any obvious issues.

Thanks

15th Jan 2019, 5:07 pm

I would say the mu range would be a bit restricted for a wide range regen. worth a try for the 'final tweak' though. Some (pre variable mu) sets used variable filament voltage for the same task. My Cossor 234 had "RF gain" done by such a method, not sure if it was ever applied to a regen stage though.

boxdoctor · 15th Jan 2019, 6:05 pm

I would expect the stage to go in and out of oscillation when the feedback exceeded the losses in the stage. This may occur as the -ve grid bias at any instant will depend on the feedback status of the stage. To put it another way which is hopefully more clear, when the stage is oscillating, the grid will be more negative than when it is not, this may cause sufficient loss of gain to make oscillations cease. The gain rises due to the reduction of the signal on the grid, causing oscillations to re-commence, and the whole cycle is repeated ad infinitum. Commonly called squegging, I believe.
Greater regenerative coupling would probably result in continuous oscillations.
No doubt there is some literature somewhere on this somewhat arcane topic. Tony.

Skywave · 15th Jan 2019, 6:07 pm

I've always been led to believe that for use as a regenerative detector, one should always choose a valve with a low mutual conductance, such as the 6BR7.

Al.

VT FUSE · 15th Jan 2019, 6:19 pm

In the specific case of the 1960 intro EF183 Frame Grid Vari-mu RF pentode,screen control of regenerative gain is the recognised way to go,I do expect that the EF183 would be pretty much in its element at an Anode potential of 12-24V.

A good example is the beginner type 1T4 / 1L4 b7g regen receiver,if a throttle cap or variometer is used to control regeneration,the best results are obtained with the sharp cutoff 1L4,if screen control via a potentiometer is used,then the vari Mu 1T4 is the better performer.

I am in total agreement with Skywave in that low Mu valves in general make for a more controllable regenerative detector,one of the best behaved ones that I built 20+ years back used a 12B4A series regulator Triode with a Mu of 6.6 and running 45V Anode.

15th Jan 2019, 6:19 pm

All valves have a bit of variable mu about them, perhaps that's why low mutual conductance ones where preferred for regen use?

G6Tanuki · 15th Jan 2019, 7:46 pm

The amount of grid-bias generated by self-rectification as a regen receiver slides gently into controlled oscillation will be miniscule compared with the amount of DC (quite a few volts) which you'd be needing to push a variable-mu valve significantly into its variable-mu region.

Over the years I've built loads of regen receivers using both normal and variable-mu valves, and never noticed any difference. One of my best had no grid resistor/capacitor so there was no way the grid could ever develop an oscillation-derived bias.

ionburn · 16th Jan 2019, 7:30 pm

Thanks for your input everyone. I may put together a few test circuits and see how I go. The only setup I have at the moment is a low Mu valve, Russian 1j29b of 2.5mA/V. I have tested this with a ferrite rod coil setup, which was a little harsh to use, and an air cored coil which was very sweet and controllable. It does indicate to me that direct comparison of valves may not be easy as different valve characteristics may be difficult to compare unless other circuit parameters are variable, but it's worth a go. Apart from EF183's I have several different pentodes for comparison.

G6Tanuki · 16th Jan 2019, 7:52 pm

Some of my best regen receivers have used the likes of the 1S5/1T4/DF33-type battery-valves which have a gm of something well below 1mA per volt.

Equally, I've used the EF91 which has a much higher slope. As you've found, making sure the reaction is smooth is the really important thing. I like reaction-control by varying the screen-grid voltage.

15th Jan 2019, 4:32 pm	#1
ionburn Heptode Join Date: Dec 2014 Location: Lincoln, Lincolnshire, UK. Posts: 583	Variable Mu Valve in Regenerative Receiver I recently came across a post elsewhere in which someone was intending to use an EF183 valve in a regenerative receiver. It is something I feel I should be able to work out, but have never really come across it (maybe because the bulk of regenerative receivers were produced before variable mu valves became common), and people often use simplar valves at very low voltages. What are thoughts the possible pros and cons of using a variable mu valve in a regenerative receiver (ie, say using one for a one valve regenerative set)? I did ask the question when I saw the post but no-one there has yet been particularly forthcoming, so maybe something for me to try but it would be good to find if there are any obvious issues. Thanks

15th Jan 2019, 5:07 pm	#2
Guest Guest Posts: n/a	Re: Variable Mu Valve in Regenerative Receiver I would say the mu range would be a bit restricted for a wide range regen. worth a try for the 'final tweak' though. Some (pre variable mu) sets used variable filament voltage for the same task. My Cossor 234 had "RF gain" done by such a method, not sure if it was ever applied to a regen stage though.

15th Jan 2019, 6:05 pm	#3
boxdoctor Rest in Peace Join Date: Sep 2010 Location: Ripley, Derbyshire, UK. Posts: 785	Re: Variable Mu Valve in Regenerative Receiver I would expect the stage to go in and out of oscillation when the feedback exceeded the losses in the stage. This may occur as the -ve grid bias at any instant will depend on the feedback status of the stage. To put it another way which is hopefully more clear, when the stage is oscillating, the grid will be more negative than when it is not, this may cause sufficient loss of gain to make oscillations cease. The gain rises due to the reduction of the signal on the grid, causing oscillations to re-commence, and the whole cycle is repeated ad infinitum. Commonly called squegging, I believe. Greater regenerative coupling would probably result in continuous oscillations. No doubt there is some literature somewhere on this somewhat arcane topic. Tony.

15th Jan 2019, 6:07 pm	#4
Skywave Rest in Peace Join Date: Jun 2006 Location: Chard, South Somerset, UK. Posts: 7,457	Re: Variable Mu Valve in Regenerative Receiver I've always been led to believe that for use as a regenerative detector, one should always choose a valve with a low mutual conductance, such as the 6BR7. Al.

15th Jan 2019, 6:19 pm	#5
VT FUSE Hexode Join Date: Aug 2010 Location: Malvern, Worcestershire, UK. Posts: 343	Re: Variable Mu Valve in Regenerative Receiver In the specific case of the 1960 intro EF183 Frame Grid Vari-mu RF pentode,screen control of regenerative gain is the recognised way to go,I do expect that the EF183 would be pretty much in its element at an Anode potential of 12-24V. A good example is the beginner type 1T4 / 1L4 b7g regen receiver,if a throttle cap or variometer is used to control regeneration,the best results are obtained with the sharp cutoff 1L4,if screen control via a potentiometer is used,then the vari Mu 1T4 is the better performer. I am in total agreement with Skywave in that low Mu valves in general make for a more controllable regenerative detector,one of the best behaved ones that I built 20+ years back used a 12B4A series regulator Triode with a Mu of 6.6 and running 45V Anode. Last edited by VT FUSE; 15th Jan 2019 at 6:28 pm.

15th Jan 2019, 7:46 pm	#7
G6Tanuki Dekatron Join Date: Apr 2012 Location: Wiltshire, UK. Posts: 13,995	Re: Variable Mu Valve in Regenerative Receiver The amount of grid-bias generated by self-rectification as a regen receiver slides gently into controlled oscillation will be miniscule compared with the amount of DC (quite a few volts) which you'd be needing to push a variable-mu valve significantly into its variable-mu region. Over the years I've built loads of regen receivers using both normal and variable-mu valves, and never noticed any difference. One of my best had no grid resistor/capacitor so there was no way the grid could ever develop an oscillation-derived bias.

16th Jan 2019, 7:30 pm	#8
ionburn Heptode Join Date: Dec 2014 Location: Lincoln, Lincolnshire, UK. Posts: 583	Re: Variable Mu Valve in Regenerative Receiver Thanks for your input everyone. I may put together a few test circuits and see how I go. The only setup I have at the moment is a low Mu valve, Russian 1j29b of 2.5mA/V. I have tested this with a ferrite rod coil setup, which was a little harsh to use, and an air cored coil which was very sweet and controllable. It does indicate to me that direct comparison of valves may not be easy as different valve characteristics may be difficult to compare unless other circuit parameters are variable, but it's worth a go. Apart from EF183's I have several different pentodes for comparison.

16th Jan 2019, 7:52 pm	#9
G6Tanuki Dekatron Join Date: Apr 2012 Location: Wiltshire, UK. Posts: 13,995	Re: Variable Mu Valve in Regenerative Receiver Some of my best regen receivers have used the likes of the 1S5/1T4/DF33-type battery-valves which have a gm of something well below 1mA per volt. Equally, I've used the EF91 which has a much higher slope. As you've found, making sure the reaction is smooth is the really important thing. I like reaction-control by varying the screen-grid voltage.