Amplifier circuit query - UK Vintage Radio Repair and Restoration Discussion Forum

Neil Purling · 28th Jan 2006, 1:24 am

Here is a image of an amplifier I have made.
The first valve is a EF91 pentode with the other grids wired so as to be a triode. The second is the pentode part of a ECL80.
It is not booming out with the 80mv input from my turntable, it's still fairly quiet.
Any alterations you'd suggest?
Those that have made the regenerative electron coupled TRF circuit might want to express a opinion as to what sort of audio output (p-p) one can wring out of the detector

Alf · 28th Jan 2006, 1:38 am

Am I missing something here, there appears to be no anode load resistor on the first valve

?

paulsherwin · 28th Jan 2006, 2:12 am

Yes Neil, if that is drawn correctly it won't work at all well. Connect the cathode of the triode directly to -ve, and add a 10M resistor between the grid and -ve (oh yes, add an anode resistor as well, but I'm assuming that's a mistake

)

Good luck, Paul

Sideband · 28th Jan 2006, 8:16 am

Neil.

If that circuit IS as you have drawn it then the triode is wired more like a cathode follower......and it won't work very well the way you have it connected anyway. Rewire the valve as Paul suggested and use something like 100K to 220K in the anode.

A cathode follower has the anode wired direct to HT and the output is taken from the cathode. The cathode is wired to chassis with a high value resistor (like a 100K) but with no bypass capacitor (otherwise most of the audio would be lost).

Rich.

Neil Purling · 28th Jan 2006, 8:35 am

I'll use 220K for the anode and 10 Meg between the grid and the chassis of the triode.
If one tried several values of grid resistance between 1M and 10M. What is likely to be the effect? A change in gain?
This is all probably blindingly obvious to you lot but I am learning as I go along (with your help).

quantum · 28th Jan 2006, 10:46 am

As it appears that you are trying to use cathode bias for your triode driver stage (as evidenced by the 150 ohm resistor and 47uf cap) I'd be inclined to use a value nearer 1 meg for your grid load, rather than 10 meg - you only need that value if you are using grid bias, and a higher value might result in 'grid blocking.'. I'd also be inclined to increase the value of the cathode resistor of the triode driver as well, to 1k to 2.2k or so.

As far as gain is concerned, if you are driving your amp with 80mv then you won't get massive volume - most two stage amps driven from a cartridge require 200mv - 1v or so to get room filling volume.

paulsherwin · 28th Jan 2006, 1:17 pm

Get rid of the cathode resistor and cap and use grid bias. The value of the grid bias resistor is traditionally 10M but it isn't too critical and any resistor over 1M should work.

It's difficult to estimate how loud it will be from an 80mV ceramic cartridge. A line level output of around 1V (typical CD player phono socket) should drive it into distortion at maximum gain though.

Good luck, Paul

Neil Purling · 28th Jan 2006, 2:43 pm

Late last night I posted a circuit with some glaring errors I ought to have spotted,except I was knackered at the time...

Here is a revision based on what is really there and represents the current state of play at lunch time today.
An attempt to introduce a 220K resistor between the anode of the triode brought back the motorboating. Even a 10K/4.7uf decoupling combination wouldnt stop it.

ppppenguin · 28th Jan 2006, 2:50 pm

There's still no signal path to the output stage! The cap that's meant to be coupling signal to the control grid of the pentode is connected to HT+ where there is no signal. Not intentionally anyway. There must be some kind of anode load resistor for the triode.

quantum · 28th Jan 2006, 3:15 pm

Well, also as it stands the cathode resistor on the triode is still too low, you could increase it to 2.2k (or even higher) or, as Paul suggests, use grid bias and connect the cathode straight to chassis and use a grid bias resistor of 4.7-10 meg or so. If you do this and have a anode load on the triode plus decoupling, you shouldn't get any motorboating - if it continues check the wiring layout (for possible positive feedback) and make sure the interstage cap feeding the pentode's grid is OK, as well as the cap across the primary of the output tranny. You could also try reducing the triode anode load, down to 47k or so, and see if that helps (that reduces the gain, so may help reduce the tendency for positive feedback). If you use cathode bias on the EF91, try removing the cathode bypass cap (and, perhaps the cathode bypass cap on the ECL80's pentode as well) as this will introduce degenerative negative feedback which might also help reduce the tendency to positive feedback.

28th Jan 2006, 1:24 am	#1
Neil Purling Nonode Join Date: Jul 2005 Location: Hull, East Yorkshire, UK. Posts: 2,087	Amplifier circuit query Here is a image of an amplifier I have made. The first valve is a EF91 pentode with the other grids wired so as to be a triode. The second is the pentode part of a ECL80. It is not booming out with the 80mv input from my turntable, it's still fairly quiet. Any alterations you'd suggest? Those that have made the regenerative electron coupled TRF circuit might want to express a opinion as to what sort of audio output (p-p) one can wring out of the detector Attached Thumbnails Last edited by Paul Stenning; 28th Jan 2006 at 11:25 am. Reason: Made subject meaningful

28th Jan 2006, 1:38 am	#2
Alf Hexode Join Date: May 2005 Location: Wigan, Greater Manchester, UK. Posts: 479	Re: Opinions wanted Am I missing something here, there appears to be no anode load resistor on the first valve?

28th Jan 2006, 2:12 am	#3
paulsherwin Moderator Join Date: Jun 2003 Location: Oxford, UK Posts: 27,970	Re: Opinions wanted Yes Neil, if that is drawn correctly it won't work at all well. Connect the cathode of the triode directly to -ve, and add a 10M resistor between the grid and -ve (oh yes, add an anode resistor as well, but I'm assuming that's a mistake ) Good luck, Paul

28th Jan 2006, 8:16 am	#4
Sideband Dekatron Join Date: Oct 2003 Location: Croydon, Surrey, UK. Posts: 7,580	Re: Opinions wanted Neil. If that circuit IS as you have drawn it then the triode is wired more like a cathode follower......and it won't work very well the way you have it connected anyway. Rewire the valve as Paul suggested and use something like 100K to 220K in the anode. A cathode follower has the anode wired direct to HT and the output is taken from the cathode. The cathode is wired to chassis with a high value resistor (like a 100K) but with no bypass capacitor (otherwise most of the audio would be lost). Rich. __________________ There are lots of brilliant keyboard players and then there is Rick Wakeman.....

28th Jan 2006, 8:35 am	#5
Neil Purling Nonode Join Date: Jul 2005 Location: Hull, East Yorkshire, UK. Posts: 2,087	Re: Opinions wanted I'll use 220K for the anode and 10 Meg between the grid and the chassis of the triode. If one tried several values of grid resistance between 1M and 10M. What is likely to be the effect? A change in gain? This is all probably blindingly obvious to you lot but I am learning as I go along (with your help).

28th Jan 2006, 10:46 am	#6
quantum Retired Dormant Member Join Date: Mar 2003 Location: Birmingham, West Midlands, UK. Posts: 785	Re: Opinions wanted As it appears that you are trying to use cathode bias for your triode driver stage (as evidenced by the 150 ohm resistor and 47uf cap) I'd be inclined to use a value nearer 1 meg for your grid load, rather than 10 meg - you only need that value if you are using grid bias, and a higher value might result in 'grid blocking.'. I'd also be inclined to increase the value of the cathode resistor of the triode driver as well, to 1k to 2.2k or so. As far as gain is concerned, if you are driving your amp with 80mv then you won't get massive volume - most two stage amps driven from a cartridge require 200mv - 1v or so to get room filling volume.

28th Jan 2006, 1:17 pm	#7
paulsherwin Moderator Join Date: Jun 2003 Location: Oxford, UK Posts: 27,970	Re: Amplifier circuit query Get rid of the cathode resistor and cap and use grid bias. The value of the grid bias resistor is traditionally 10M but it isn't too critical and any resistor over 1M should work. It's difficult to estimate how loud it will be from an 80mV ceramic cartridge. A line level output of around 1V (typical CD player phono socket) should drive it into distortion at maximum gain though. Good luck, Paul

28th Jan 2006, 2:43 pm	#8
Neil Purling Nonode Join Date: Jul 2005 Location: Hull, East Yorkshire, UK. Posts: 2,087	Re: Amplifier circuit query Late last night I posted a circuit with some glaring errors I ought to have spotted,except I was knackered at the time... Here is a revision based on what is really there and represents the current state of play at lunch time today. An attempt to introduce a 220K resistor between the anode of the triode brought back the motorboating. Even a 10K/4.7uf decoupling combination wouldnt stop it. Attached Thumbnails

28th Jan 2006, 2:50 pm	#9
ppppenguin Retired Dormant Member Join Date: Dec 2003 Location: North London, UK. Posts: 6,168	Re: Amplifier circuit query There's still no signal path to the output stage! The cap that's meant to be coupling signal to the control grid of the pentode is connected to HT+ where there is no signal. Not intentionally anyway. There must be some kind of anode load resistor for the triode.

28th Jan 2006, 3:15 pm	#10
quantum Retired Dormant Member Join Date: Mar 2003 Location: Birmingham, West Midlands, UK. Posts: 785	Re: Amplifier circuit query Well, also as it stands the cathode resistor on the triode is still too low, you could increase it to 2.2k (or even higher) or, as Paul suggests, use grid bias and connect the cathode straight to chassis and use a grid bias resistor of 4.7-10 meg or so. If you do this and have a anode load on the triode plus decoupling, you shouldn't get any motorboating - if it continues check the wiring layout (for possible positive feedback) and make sure the interstage cap feeding the pentode's grid is OK, as well as the cap across the primary of the output tranny. You could also try reducing the triode anode load, down to 47k or so, and see if that helps (that reduces the gain, so may help reduce the tendency for positive feedback). If you use cathode bias on the EF91, try removing the cathode bypass cap (and, perhaps the cathode bypass cap on the ECL80's pentode as well) as this will introduce degenerative negative feedback which might also help reduce the tendency to positive feedback.