Yes, I have some equal experiences with TVC / AVC as standalone in my system.... now I'm changing to activate my system using a DSP as crossover. My phono stage (all interstage / transformer coupled) has gain enough at the output (around 2V signal output) to drive the dsp and amplifiers.
SJS,
correct, that was a way too fast in my answer - missed the closer look at the base of the schematic. Anyway, with a double RC-network the preamp could be more quieter...
A phono amp would be next on my list of things to do. But I will need advice on what to choose; not solid state of course, possibly LCR equalisation? But I don't know enough to make an informed choice, at the moment.
As an old fashioned analogue type, I am suspicious of digital processing, especially if it can be done in the analogue domain. I still have my four way Ben Duncan designed active crossover, that I built in the early '80s, but considerably updated over the years, so that's what I would use. Besides, the idea of having a system where I made everything myself is still a dream of mine.
Mine is a tube phono stage using an LCR-RIAA, but transformer coupled. So I can better "play" with gain structure. Anyway I'm sure you'll find your way to go - just keep in mind the impedance of the LCR-RIAA network. You can find phono stages with higher LCR-RIAA - around 10KOhm (but you have take care about hum with higher impedance LCR).
This is off topic here, and I'm an old fashioned analog friend as well, but just a point to think for digital processing. I have a horn system and you can easely manage all time delay of the drivers / horns, as well as phase and filter functions. In the end you will have more plus then minus and the sound is the reference not our sensitivities. Just to mention and now back to the topic #26 preamp.
I've been listening to my 4P1L SET for quite a while - the 4P1L driver is in the same chassis. The driver has a plate choke and FT-3 cap, and OPT is a LL1682/50mA
I'm in the mood for a bit more experimenting, so I made up a separate output stage. This has SE 4P1L into a LL1620/80mA wired as 5K. I intend to try out various driver stages:
- 4P1L with plate choke/cap
- 4P1L with Ale's Gyrator
- 26 with LL1660/5mA
- 26 with Ale's gyrator
- 01A with Ale's gyrator
Listening right now to 26 with LL1660/5mA. Sounds very nice - not too much difference from the 4P1L driver, to be honest. Loudness is similar as well. So back in time to about 2010 for me, which is when I must have built it!
I should have used this separate stage design a while ago - it's really useful for experimenting. The driver stage (or line stage or whatever - I use 2 stages total) gets its B+ from the output stage chassis. So all from the one PSU. Fortunately both the 26 and 4P1L driver stages have a 17vDC filament supply to Rod's boards so they're interchangeable. The 4P1L output stage has a 27v supply to the single 4P1L. Since I have elaborate choke input filament supplies this does make it easier.
Hope to have some fun this Autumn. I'm going to have some left over 4P1L SET and PSET amps if anybody is interested - send me a message.
I'm in the mood for a bit more experimenting, so I made up a separate output stage. This has SE 4P1L into a LL1620/80mA wired as 5K. I intend to try out various driver stages:
- 4P1L with plate choke/cap
- 4P1L with Ale's Gyrator
- 26 with LL1660/5mA
- 26 with Ale's gyrator
- 01A with Ale's gyrator
Listening right now to 26 with LL1660/5mA. Sounds very nice - not too much difference from the 4P1L driver, to be honest. Loudness is similar as well. So back in time to about 2010 for me, which is when I must have built it!
I should have used this separate stage design a while ago - it's really useful for experimenting. The driver stage (or line stage or whatever - I use 2 stages total) gets its B+ from the output stage chassis. So all from the one PSU. Fortunately both the 26 and 4P1L driver stages have a 17vDC filament supply to Rod's boards so they're interchangeable. The 4P1L output stage has a 27v supply to the single 4P1L. Since I have elaborate choke input filament supplies this does make it easier.
Hope to have some fun this Autumn. I'm going to have some left over 4P1L SET and PSET amps if anybody is interested - send me a message.
Sounds like a fun and interesting work ahead. Keen to listen to your final experiments. Try 01a for sure, we fixed Tony's recently and it does sound really nice. The tone and clarity is the best so far.
Here are some interesting measurements as well:
http://www.bartola.co.uk/valves/2016/08/27/tonys-01a-preamp/
Here are some interesting measurements as well:
http://www.bartola.co.uk/valves/2016/08/27/tonys-01a-preamp/
Very interesting - the 01A is certainly going to be built. I have several chassis from my previous experiments I can make use of, including an 01A with an interstage.
I am getting a little less overall volume going from a 4P1L driver to the 26, into SE 4P!L. From mu 11 down to mu 8-ish. My volume is right on the limit, so only just enough unless I get more output from a different DAC.
My previous OPT was LL1682 and now using LL1620. I'll have to try a direct comparison. I'm getting more clarity on vocals (hear the words better) with the 26+LL1620, but a little less treble bite. Hard to explain - I always found the 4P1L superior in the treble to the 26, though with a slight edge, but I never noticed increased clarity with the 26. This may be the LL1620. As with most things it's sorting out the variables, which is why separate chassis for inout and output is useful.
Last edited:
Found it here on Spice3.txt and it runs fine (change ^ to **)
http://www.diyaudio.com/forums/tubes-valves/243950-vacuum-tube-spice-models-3.html
http://www.diyaudio.com/forums/tubes-valves/243950-vacuum-tube-spice-models-3.html
After using my #26 pre for a while, there IS a little hum and microphonics, so I am looking at ways of reducing both. My valve bases are already mounted on flexible mounts similar to Thomas Mayer's system, so I am considering making phosphor bronze mesh tubes to surround each valve, with a sorbothane collar on the inside pressing against the valve near the top, hopefully killing two birds with one stone. But how much coverage by the mesh would be needed to provide effective screening? Mesh is easily available from #90 (36% open area) to #16 (66% open area). Less means the valves will be more visible, which is good, but would that reduce the screening effect? Of course they will be grounded.
Has anyone played with this technique?
Has anyone played with this technique?
These 26 valves are pretty old guys - they vary internally as regards microphonics. I don't have particularly sensitive speakers, so I'm not bothered, but if I had speakers over 100db I'd be going through several examples to see if any of this is internal to the valve itself. I have quite a few 26s. Most are fine because they were bought from dealers who tested them, but one or two bad examples crept in from ebay where the microphonics were really audible.
I have few dozens of tubes, and use -for some years ago daily- #26 preamp.
Microphonics is the smaller problem: use well damped heavy board for sockets with appropriate vibration isolation.
EACH early DHT is sensitive for radiated hum.
In my #26 stock the 50 Hz FFT component -measured tubes in row in the same position in my preamp- changing even 30 dB!!!
If you don't have any (computer) spectrum analyser, try to listening with shorted preamp input.
first try to wrap aluminium foil around the tubes and tie to ground it.
Try to put metal pencil holder over the tube and tie to ground it.
Microphonics is the smaller problem: use well damped heavy board for sockets with appropriate vibration isolation.
EACH early DHT is sensitive for radiated hum.
In my #26 stock the 50 Hz FFT component -measured tubes in row in the same position in my preamp- changing even 30 dB!!!
If you don't have any (computer) spectrum analyser, try to listening with shorted preamp input.
first try to wrap aluminium foil around the tubes and tie to ground it.
Try to put metal pencil holder over the tube and tie to ground it.
I've bought some mesh, and I am waiting for the sorbothane sheet to arrive, so I can make screening and damping tubes. But in the meantime I covered the shelf with copper foil, made a steel plate to fit above the pre-amp, and earthed them both to the chassis. Now I have to get really close to the speakers to hear any hum, so hopefully the bronze mesh tubes will be the icing on the cake. The mesh and sorbothane have cost about £14 in total; the copper foil and steel cost nothing!
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
I'm now running it through the Sowter TVCs, with the TKD pot wound up to max. I've ordered a pair of Audio Note non-magnetic 100k tantalum resistors to replace the pot, and hard wired a lead to the TVC to take the output from the pre. Sounds a bit better than the TKD, and it has got rid of the residual hum. Also waiting for a couple of AZ31s to try.
But most importantly, I decided to do something about the lack of glow from the 26s, especially inside their gilded cages. So, a pair of pink LEDs later (from a set of Christmas lights), I now have a tasteful pink glow. My wife says there should be tinsel!
But most importantly, I decided to do something about the lack of glow from the 26s, especially inside their gilded cages. So, a pair of pink LEDs later (from a set of Christmas lights), I now have a tasteful pink glow. My wife says there should be tinsel!
An externally hosted image should be here but it was not working when we last tested it.
There seems to be a preference for enamel type wirewound resistors in the filament bias supplies vs. alu clads. Why is it that the latter type doesn't sound as good? One should think that a sufficiently burned in (yes I do believe that some resistors needs a little burn in) good quality non-inductive type alu should sound just as good. Does anyone know what type Thomas Mayer is using is his excellent designs?
I did a lot of tests on filament resistors in filament bias. Worst was thick film and the big white ceramics. Alu clad weren't really much better.
The good ones were the Dale surplus stock, of which Ale and I bought a good number from the USA, the vitreous enamel ones which were mostly surplus stock, and the current production Welwyn W21-W24 series.
Thomas may be using the alu clad resistors because he needs higher wattage, and I imagine he uses current stock items in preference to surplus stock that comes up intermittently on ebay. Just guessing here.
I would certainly go for vitreous enamel on the basis of my own listening tests.
The good ones were the Dale surplus stock, of which Ale and I bought a good number from the USA, the vitreous enamel ones which were mostly surplus stock, and the current production Welwyn W21-W24 series.
Thomas may be using the alu clad resistors because he needs higher wattage, and I imagine he uses current stock items in preference to surplus stock that comes up intermittently on ebay. Just guessing here.
I would certainly go for vitreous enamel on the basis of my own listening tests.
I have recently acquired about £2000 worth (?) of cabling, power, IC, digital and speaker, which together have added clarity and definition to the sound. Amusingly a £10 AZ31 rectifier has done just as much! Luckily I paid cash for the valve, but did a bit of work (no cash) for the cables. As I am retired that was no hardship, and I learned a bit too.