Thanks @Sheldon - good pointers in that link.
I think I will try some of the LiFePO4 batteries and use a diode to drop the voltage the 0.5V or so that would keep the valves happy and the battery too. That should give long play time between charges, very low impedance and very low noise. A set of 4xD cells seems to give around 20hrs (alkaline types) listening time, maybe even a bit longer.
I've learned a lot implementing this, every day is a school day!
I think I will try some of the LiFePO4 batteries and use a diode to drop the voltage the 0.5V or so that would keep the valves happy and the battery too. That should give long play time between charges, very low impedance and very low noise. A set of 4xD cells seems to give around 20hrs (alkaline types) listening time, maybe even a bit longer.
I've learned a lot implementing this, every day is a school day!
With oxide coated filaments or cathodes, the filament voltage has a big influence on the lifetime of the tube. For the best lifetime, don't exceed the nominal voltage - the sub-surface barium will be burned up at a wastefully high rate, and the supply of it will be exhausted much faster. But you can run slightly low (-5%) for temporary events, like low battery.
Thoriated tungsten filaments like the 10Y and 01A can use from 5 or 10% lower filament voltage, if the microphony is improved, but gm will often be lower, and so the sound is not guaranteed to be the same.
The Philips Technical Review had somewhat to say about lifetime and heating voltages, as applied to oxide cathodes - see attached. The +/- 5% is a concession to the reality of mains line-voltage tolerances (which are more like ±10% nowadays) - it is not as desirable (lifetime-wise) as the 1% stabilized voltage method they mention - which certainly gives longer life.
Thoriated tungsten filaments like the 10Y and 01A can use from 5 or 10% lower filament voltage, if the microphony is improved, but gm will often be lower, and so the sound is not guaranteed to be the same.
The Philips Technical Review had somewhat to say about lifetime and heating voltages, as applied to oxide cathodes - see attached. The +/- 5% is a concession to the reality of mains line-voltage tolerances (which are more like ±10% nowadays) - it is not as desirable (lifetime-wise) as the 1% stabilized voltage method they mention - which certainly gives longer life.
Attachments
@Rod Coleman - thanks for this info. I have the batteries hooked up now vi a standard 1n4007 and the drop is just a tad high in circuit, dropping something more like 0.7V. I think I will try a schottky or two which should hit it just about right. I have a little voltmeter setup ready to roll on a momentary switch, so just a push of the switch lets you know the voltage.
Thank you Rod for the insight, much appreciated.
-fran....little advice from an old longtime diyer. Please don't waste your time now. Use 2xD cell batteries in series per tube with voltage dropping resistors and run/use the line stage in your system for 2/3 weeks. If you like the SQ come back and tell us what improvement you want to make. We work together to reach your goal. I am afraid in the present state gain will be too much and it will upset the gain structure of your system.
Regards
-fran....little advice from an old longtime diyer. Please don't waste your time now. Use 2xD cell batteries in series per tube with voltage dropping resistors and run/use the line stage in your system for 2/3 weeks. If you like the SQ come back and tell us what improvement you want to make. We work together to reach your goal. I am afraid in the present state gain will be too much and it will upset the gain structure of your system.
Regards
@minhaj - thank you for your wise words. At this point, using batteries, the system performs well enough, with low enough noise and high enough performance to allow for a decent comparison with other output stages.
In short I am impressed with the performance for something from such a "cheap" source. The advice here from you and all has been invaluable and has allowed me to do a fair comparison with a solid state output stage. At this point, I think the performance is excellent and I am very happy with this experiment. The gain is OK - I am working from a current output DAC so I have some scope to change the output resistor within reason and the high mu is easier to deal with. I am pretty impressed with it. I think if I was to stick with it long term I would want something other than batteries (or would at least want a system that switched between using batteries and recharging them) but for now it works well, good enough to allow comparisons.
I have some RCA 3a5 coming soon that I will check out for performance, but sitting the board on a piece of foam sorts out microphonics. I'll see how the new ones perform.
Thanks again all for your help, it is much appreciated.
fran
In short I am impressed with the performance for something from such a "cheap" source. The advice here from you and all has been invaluable and has allowed me to do a fair comparison with a solid state output stage. At this point, I think the performance is excellent and I am very happy with this experiment. The gain is OK - I am working from a current output DAC so I have some scope to change the output resistor within reason and the high mu is easier to deal with. I am pretty impressed with it. I think if I was to stick with it long term I would want something other than batteries (or would at least want a system that switched between using batteries and recharging them) but for now it works well, good enough to allow comparisons.
I have some RCA 3a5 coming soon that I will check out for performance, but sitting the board on a piece of foam sorts out microphonics. I'll see how the new ones perform.
Thanks again all for your help, it is much appreciated.
fran
So I have been listening quite a bit to this over the past few days, and I also ordered a pair of RCA 3a5 valves. These arrived today and they are certainly less microphonic than the ones that came with the kit. This is good news as far as I can see.
@minhaj - you asked me what my goal is with this. I think I want to stay with this as an IV output stage on this particular build. The sound has something extra that is very attractive.
I would like to get rid of the batteries if I could - although they last a long time, they still will need recharging. So if you have a good value, worthwhile recommendation for a replacement that would be great. I find the original PCB (circuit attached here again) useful and would rather not start on a whole new build if possible. I am also open to any improvements you might suggest - although the stage is now silent, if there is something extra to be wrung from it using a regulated B+ I am happy to look at that too.
Fran
Attachments
1. Shifting OP to more linear region
2. Filament bias
3. Omitting input cap (if no dc from source)
4. Parafeed config
5. Choke/CCS/Gyrator loaded plates
6. Choke/pseudo choke input PS
7. Coleman filament PS
Pls check this link for better op and filament bias
Filament Bias: a practical example with 3A5 DHT – Bartola(R) Valves
Regards
2. Filament bias
3. Omitting input cap (if no dc from source)
4. Parafeed config
5. Choke/CCS/Gyrator loaded plates
6. Choke/pseudo choke input PS
7. Coleman filament PS
Pls check this link for better op and filament bias
Filament Bias: a practical example with 3A5 DHT – Bartola(R) Valves
Regards
Fran I will use a simple cascode CCS with a couple of DN2540 in place of resistor anode load and change the OP as minhaj suggested for 105V 10mA so you have a better idea what's the potential of the tube, if you like it you will have time to improve the rest. I use with a lot of satisfaction Rod Coleman filaments regs. & Ale Moglia hybrid mu-follower.
Last edited:
@minhaj, @merlin el mago - thanks guys for your input. I will have a set of Coleman supplies coming soon and will adjust the operating points... I think it will be worth it based on what I'm hearing.
I was thinking of ordering a few of these boards, although the circuit could be breadboarded fairly easily
https://www.diyaudio.com/forums/tub...overs-free-gerbers-thread-10.html#post6559289
fran
I was thinking of ordering a few of these boards, although the circuit could be breadboarded fairly easily
https://www.diyaudio.com/forums/tub...overs-free-gerbers-thread-10.html#post6559289
fran
Good stuff.I thought I would give a little update on this project.
Following the recommendations above, I contacted Rod Coleman who kindly supplied a pair of filament supplies. I got these built and installed and also implemented filament bias as per the Bartola site linked above. There is still some work that can be done - like removing the input cap. The stage is very quiet indeed, and I think seems a little more dynamic than it was before. I will build a CCS for the anode, but I'm waiting in parts for that, so it will be a little bit before I get to it.
So a success thanks to your advice so far - thank you for the help.
Fran
Yes, there is a improvement in removing that capacitor - seems like just a tad more immediacy. The promise I heard in this back when just the kit from Aliexpress is really starting to shine through now, I do think this is a keeper alright and I'm glad I went for the improvements you recommended @minhaj and others. I'll draw up a schematic as it stands and post it soon.
And, the other night I bought a pair of 01A to try at some point in the future.
And, the other night I bought a pair of 01A to try at some point in the future.