Silicon instead of EZ81 - UK Vintage Radio Repair and Restoration Discussion Forum

bikerhifinut · 19th Jan 2016, 9:26 pm

Hi guys,
I have a mains transformer that was originally in an Armstrong stereo radiogram running 2 x EL84.
The issue that I have is that I can't get the 300V I want for my EL84 SE amps, nearer to 250 is what I get, using an EZ81.
So I thought if I use a couple of silicon diodes instead I should get a bit more Ht out of it?
Also I can save on a heater transformer for the EZ81, I never did like the idea of using the same 6.3V winding for the rectifier and signal valves.
So I figured from the 210V secondary which is what I remember it as, I will measure it again when i get the thing on the bench, I should get something around 300V using Silicon rectifiers (1N4007 or the UF variant)? Have I overlooked anything important? I reckon the regulation of the transformer should be pretty good and the primary is not loaded as much as it was in its original application as theres one EZ81 plus three other ECC etc types less on the 6.3V.
I bet I have gone wrong here somewhere...

The reason why I am considering this is that I want to rework my SE El84 stereo "Mini Amp", the circuit works and sounds pretty darn good but I want to tidy everything up and get it nice so making a new top plate for the case/chassis isn't an issue.

Thanks gentlemen

Andy.

G6Tanuki · 19th Jan 2016, 9:46 pm

Just pop in a couple of BY127 or equivalent in place of the old EZ81 - you may get a few volts more HT, but it'll still be within the usual +/10% these sorts of amps were designed to anyway.

paulsherwin · 19th Jan 2016, 9:50 pm

Check the voltage rating of the smoothers is adequate. The HT supply will be unloaded until the valves warm up and will rise quite a lot.

bikerhifinut · 19th Jan 2016, 10:29 pm

Thanks guys,
Voltage ratings not an issue Paul. This is a scratch (re)build, I know I can get a reasonable performance as it stands. It's just that the circuit works better with 300V rather than 250V, gets me a bit more power. As it stands the amp as measured is good for about 2.5W before the sinewave clips/distorts. 3W or better would be nice. I just figured rather than waste valuable volts through an EZ81 I could make the most of whats available. And dispense with the extra heater transformer.
Just for the record, I'm using a CCS on the cathodes of the EL84's rather than conventional resistors, set to 46mA.
Its based on the RH84 circuit link here:
http://rh-amps.blogspot.co.uk/2013_02_01_archive.html

Simple circuit, works surprisingly well and sounds nice within its limits in my fairly decent stereo system. less is more ?

regards

Andy.

Herald1360 · 19th Jan 2016, 11:29 pm

Won't running with a lower Vg2 mean that less -ve bias is needed by the grid meaning that at full drive it'll hit grid current sooner? It might be interesting to experiment here if you're after as much output as possible.

It would be interesting to compare performance of this against a 3-3.

bikerhifinut · 19th Jan 2016, 11:42 pm

It would wouldn't it?
I wasn't expecting much from it, given its simplicity but I was pleasantly surprised and given its modest power output compared to my usual 30watt valve amps iot only proved to me that unless I am rocking out at neighbour annoying volumes, you can get a very pleasurable sound from a couple of watts.
I do have a smallish living room so that helps of course.

A.

G8HQP Dave · 20th Jan 2016, 3:28 pm

Quote:

Originally Posted by bikerhifinut

Just for the record, I'm using a CCS on the cathodes of the EL84's rather than conventional resistors, set to 46mA.

A resistor will do a better job, but for some strange reason the CCS meme is persistent.

Use a CCS when you need either a high impedance or a precise current. Neither is true for cathode biasing an output valve. In fact, the CCS will precisely maintain the wrong current: average current instead of quiescent current. This ensures that the bias shift due to second-order distortion when signal is present will be worse with CCS than for a simple resistor.

bikerhifinut · 20th Jan 2016, 6:40 pm

I see what you mean.
If its on a single ended stage though, which will be working in Class A, or a push pull stage running in Class A then I thought the quiescent current was the same as the on-load current? at least at "normal" listening volumes?
I hadn't really thought about that aspect, but it does sound just fine and I quite like what I hear.
Bearing in mind CCS in this instance is a constant current SINK not source.
I'll look it up in the book again, there's a paragraph in the latest Morgan Jones Valve amplifiers book relating to the use of a regulator Constant current sink in output valves cathode circuits and the whys and why nots.

Anyway this is slightly off topic for my original question which was about the extra voltage gained by using silicon instead of a valve rectifier. But thanks for the insight.

Andy.

19th Jan 2016, 9:26 pm	#1
bikerhifinut Octode Join Date: Nov 2013 Location: Penrith, Cumbria, UK Posts: 1,993	Silicon instead of EZ81 Hi guys, I have a mains transformer that was originally in an Armstrong stereo radiogram running 2 x EL84. The issue that I have is that I can't get the 300V I want for my EL84 SE amps, nearer to 250 is what I get, using an EZ81. So I thought if I use a couple of silicon diodes instead I should get a bit more Ht out of it? Also I can save on a heater transformer for the EZ81, I never did like the idea of using the same 6.3V winding for the rectifier and signal valves. So I figured from the 210V secondary which is what I remember it as, I will measure it again when i get the thing on the bench, I should get something around 300V using Silicon rectifiers (1N4007 or the UF variant)? Have I overlooked anything important? I reckon the regulation of the transformer should be pretty good and the primary is not loaded as much as it was in its original application as theres one EZ81 plus three other ECC etc types less on the 6.3V. I bet I have gone wrong here somewhere... The reason why I am considering this is that I want to rework my SE El84 stereo "Mini Amp", the circuit works and sounds pretty darn good but I want to tidy everything up and get it nice so making a new top plate for the case/chassis isn't an issue. Thanks gentlemen Andy.

19th Jan 2016, 9:46 pm	#2
G6Tanuki Dekatron Join Date: Apr 2012 Location: Wiltshire, UK. Posts: 14,005	Re: Silicon instead of EZ81 Just pop in a couple of BY127 or equivalent in place of the old EZ81 - you may get a few volts more HT, but it'll still be within the usual +/10% these sorts of amps were designed to anyway.

19th Jan 2016, 9:50 pm	#3
paulsherwin Moderator Join Date: Jun 2003 Location: Oxford, UK Posts: 27,960	Re: Silicon instead of EZ81 Check the voltage rating of the smoothers is adequate. The HT supply will be unloaded until the valves warm up and will rise quite a lot.

19th Jan 2016, 10:29 pm	#4
bikerhifinut Octode Join Date: Nov 2013 Location: Penrith, Cumbria, UK Posts: 1,993	Re: Silicon instead of EZ81 Thanks guys, Voltage ratings not an issue Paul. This is a scratch (re)build, I know I can get a reasonable performance as it stands. It's just that the circuit works better with 300V rather than 250V, gets me a bit more power. As it stands the amp as measured is good for about 2.5W before the sinewave clips/distorts. 3W or better would be nice. I just figured rather than waste valuable volts through an EZ81 I could make the most of whats available. And dispense with the extra heater transformer. Just for the record, I'm using a CCS on the cathodes of the EL84's rather than conventional resistors, set to 46mA. Its based on the RH84 circuit link here: http://rh-amps.blogspot.co.uk/2013_02_01_archive.html Simple circuit, works surprisingly well and sounds nice within its limits in my fairly decent stereo system. less is more ? regards Andy.

19th Jan 2016, 11:29 pm	#5
Herald1360 Dekatron Join Date: Feb 2007 Location: Leominster, Herefordshire, UK. Posts: 16,536	Re: Silicon instead of EZ81 Won't running with a lower Vg2 mean that less -ve bias is needed by the grid meaning that at full drive it'll hit grid current sooner? It might be interesting to experiment here if you're after as much output as possible. It would be interesting to compare performance of this against a 3-3. __________________ ....__________ ....\|____\|\|__\|__\_____ .=.\| _---\__\|__\|_---_\|. .........O..Chris....O

19th Jan 2016, 11:42 pm	#6
bikerhifinut Octode Join Date: Nov 2013 Location: Penrith, Cumbria, UK Posts: 1,993	Re: Silicon instead of EZ81 It would wouldn't it? I wasn't expecting much from it, given its simplicity but I was pleasantly surprised and given its modest power output compared to my usual 30watt valve amps iot only proved to me that unless I am rocking out at neighbour annoying volumes, you can get a very pleasurable sound from a couple of watts. I do have a smallish living room so that helps of course. A.

20th Jan 2016, 6:40 pm	#8
bikerhifinut Octode Join Date: Nov 2013 Location: Penrith, Cumbria, UK Posts: 1,993	Re: Silicon instead of EZ81 I see what you mean. If its on a single ended stage though, which will be working in Class A, or a push pull stage running in Class A then I thought the quiescent current was the same as the on-load current? at least at "normal" listening volumes? I hadn't really thought about that aspect, but it does sound just fine and I quite like what I hear. Bearing in mind CCS in this instance is a constant current SINK not source. I'll look it up in the book again, there's a paragraph in the latest Morgan Jones Valve amplifiers book relating to the use of a regulator Constant current sink in output valves cathode circuits and the whys and why nots. Anyway this is slightly off topic for my original question which was about the extra voltage gained by using silicon instead of a valve rectifier. But thanks for the insight. Andy.