I felt that way for a long, long time, but certain tubes work great with sand and others just don't. I was a purist, but at a certain point wondered what I was missing since I now have some pretty good sounding solid state gear (sources in particular). Dogma I am finding has little place in achieving my goals so I decided to give it a shot, so far it has allowed me to achieve the wanted design goals including being subjectively pleasing to its designer, which tbh is not always the case.
Try graphite plate 801 (not cheap, but worth it).
I tried several VT-25 (NOT A)/10Y, 801, 801a tubes in my 10/801 preamp -gyrator, 20mA anode current-, and graphite plate tubes were the best of them in tonality.
Interesting feedback Andy. I haven' tried the 26 again back with the latest gyrator I use, although posted the design after some measurements over the bench long time ago.
The VT-25 is amongst my favourite DHT stages I'm currently using along with the 01a, 2P29L and 4P1L. here is my original version with BSH11BK at lower FET position which has higher transconductance than BF862 at and can do more than 20mA. It also sounds really nice. The bass is powerful with this hybrid mu-follower stage which can't be easily obtained with any IT given the anode resistance of this valve.
Using sand in the right places is perfectly fine. You will be surprised with the results. For me as anode loads (CCS or VCCS/gyrator), phono stage inputs and/or source followers are top performer roles for silicon.
Despite I don't own a graphite plate of 801, I will try Elrog's version of the 801a when they become available later this year
Also is on my list the EML20A as Kevin suggested. Great valve!
Cheers
Ale
The VT-25 is amongst my favourite DHT stages I'm currently using along with the 01a, 2P29L and 4P1L. here is my original version with BSH11BK at lower FET position which has higher transconductance than BF862 at and can do more than 20mA. It also sounds really nice. The bass is powerful with this hybrid mu-follower stage which can't be easily obtained with any IT given the anode resistance of this valve.
Using sand in the right places is perfectly fine. You will be surprised with the results. For me as anode loads (CCS or VCCS/gyrator), phono stage inputs and/or source followers are top performer roles for silicon.
Despite I don't own a graphite plate of 801, I will try Elrog's version of the 801a when they become available later this year
Also is on my list the EML20A as Kevin suggested. Great valve!
Cheers
Ale
I think we're on to something here - Kevin's comments are very interesting indeed. We're now looking at the best possible DHTs in the best possible circuits. I've built with all these DHTs myself and tried Lundahl interstages like the LL1660, plate chokes like the 126C and Ale's gyrator. I've eliminated interstages as an option unless something comes along better than the LL1660 - maybe amorphous. Ale's gyrator sounded better.
With plate chokes I'm not so sure. Maybe it depends on the DHT in question. Kevin and Ale have tried out a large number of the combinations as well. I think we've all moved away from the 26, though it may still please in some incarnation. The 10Y will always be good, and seems to survive just about any circuit. The 4P1L in all the builds I did was a little better than the 26, and the 2P29L is promising as well. Both of these can be run at 15mA and over, like the 10Y, so they can be used as drivers.
This leaves the 01A - great sound, right up there with the very best, but 3mA current so not a driver. Ale has been working away steadily to adapt this for use as a driver, and I have his latest source follower which I haven't tried yet. This then opens the door to other low current DHTs and there are a few of those that are of interest, some quite esoteric. The 01A is already a great preamp tube followed by a driver stage, but this could give it greater versatility. I do agree that 01A, 10Y, 2P29L and 4P1L are the current contenders, as Ale says. Our ears seem to agree on this. Maybe there are other options - certainly some low mu DHTs. But Ale and I use 2 stage amps with PSE 4P1L output stages, all in filament bias. So this requires the input tube to have a mu of 8 or more for the single unit speakers we use - Alpair 10s for Ale and myself.
We're determined men! We're trying to take the 2 stage filament biased DHT amp to its limits and beyond..... And add Thomas Mayer to the list - he turned us on to filament bias in the first place and paved the way for Rod's regs used in filament bias. And many other builders who have picked up the torch of this fascinating journey into the world of DHTs from the 1920s and later, which still haven't met their match as audio devices.
With plate chokes I'm not so sure. Maybe it depends on the DHT in question. Kevin and Ale have tried out a large number of the combinations as well. I think we've all moved away from the 26, though it may still please in some incarnation. The 10Y will always be good, and seems to survive just about any circuit. The 4P1L in all the builds I did was a little better than the 26, and the 2P29L is promising as well. Both of these can be run at 15mA and over, like the 10Y, so they can be used as drivers.
This leaves the 01A - great sound, right up there with the very best, but 3mA current so not a driver. Ale has been working away steadily to adapt this for use as a driver, and I have his latest source follower which I haven't tried yet. This then opens the door to other low current DHTs and there are a few of those that are of interest, some quite esoteric. The 01A is already a great preamp tube followed by a driver stage, but this could give it greater versatility. I do agree that 01A, 10Y, 2P29L and 4P1L are the current contenders, as Ale says. Our ears seem to agree on this. Maybe there are other options - certainly some low mu DHTs. But Ale and I use 2 stage amps with PSE 4P1L output stages, all in filament bias. So this requires the input tube to have a mu of 8 or more for the single unit speakers we use - Alpair 10s for Ale and myself.
We're determined men! We're trying to take the 2 stage filament biased DHT amp to its limits and beyond..... And add Thomas Mayer to the list - he turned us on to filament bias in the first place and paved the way for Rod's regs used in filament bias. And many other builders who have picked up the torch of this fascinating journey into the world of DHTs from the 1920s and later, which still haven't met their match as audio devices.
Last edited:
Just a quick postscript. I switched back and fore between the 26 with gyrator and the 2P29L with 126C as plate choke. I do indeed prefer the 2P29L. It sounds more organic - the sound is full and coherent from top to bottom. The 26 is superficially brighter but the treble sounds somehow disconnected, hard to explain exactly. It's not as coherent top to bottom. I think that's what some say of amorphous cores - superficially brighter but less coherent top to bottom. This is subjective, of course, and I'm just throwing it into the ring in case others have a better explanation. Kevin has given some pointers in his post. Tony is still doubtful about solid state, but I really think Ale has a good product and it's worth persevering with as he goes through constant improvements and updates. He's really working hard on this and the results are already impressive in practice. The gyrator is the best solution for the 01A in my experience, for instance, and may well turn out to be for the other DHTs. The jury is still out on esoteric plate chokes, however, just haven't tried the really top ones..... And then there's the question of the coupling cap, for which I've always used Teflon FT-2 and FT-3. That may not be the end of the road, but it applies to both plate chokes and gyrator.
Last edited:
I too have become more accepting of sand in valve amps, starting with Rod's filament supplies and more recently Ale's gyrator, which I use with 26s. Both made improvements over the passive solutions I was using, the gyrator was a nice improvement over the anode choke I was using previously. Until I killed one of the channels anyway. I don't have any 01As but i should try replacing the 26s sometime (26-10Y-300B).
The sand I haven't come to terms with yet is replacing slow start valve rectifiers with, say, SiC. I've read here a number of times that SiC is superior, but I can't quite get over my prejudice for the slow start. Anyone want to help me?
The sand I haven't come to terms with yet is replacing slow start valve rectifiers with, say, SiC. I've read here a number of times that SiC is superior, but I can't quite get over my prejudice for the slow start. Anyone want to help me?
I recommend 01A > 10Y > 300b in preference. That's pretty hard to beat. Do you need slow start rectifiers with DHTs? I'm using AZ1 + AZ11 mesh, AZ12, 5R4GY or 80 as a rule, all directly heated. Mind, I leave my system on pretty much 24/7 since everything in my computer including TV and videos as well as music goes through my audio system. Will SIC beat a mesh plate AZ1?
Last edited:
I recently tried to reduce the HF droop on my 26 by adding some capacitance across the output. My power amp only has a 33k input impedance. Going up from a few pF to 5nF, the resonant peak at around 55k, and dip at 45kHz, have reduced, but I still have the droop above 10kHz. The OPT is a Lundahl LL2745 in Alt R, which has plenty of output for my needs.
The 01A uses Ale's gyrator, but has way too much gain, so needs a TVC or AVC* on the output, as it hums like mad with the pot on the input. (It is on the living room table at the moment being rewired internally with screened cable). How much gain would the 26 have with the gyrator instead of the LL2745?
* Hi there Andy.
The 01A uses Ale's gyrator, but has way too much gain, so needs a TVC or AVC* on the output, as it hums like mad with the pot on the input. (It is on the living room table at the moment being rewired internally with screened cable). How much gain would the 26 have with the gyrator instead of the LL2745?
* Hi there Andy.
I replaced my UF4007 + AZ1 hybrid Gaertz for Sic + AZ1 and resulted in a big improvement of SQ.
The gyrator doesn't have a protection when the output or anode is shorted. If it is, the top MOSFET (IXTP08N100D or DN2540) and likely the lower FET will die. Unless you exceed the VDSS max of the LND150 the CCS should survive. Generally by replacing the top MOSFET and the jFET should suffice to fix the issue.
Have you tracked what caused the issue in your case?
Cheers
Ale
The gain of the stage (hybrid mu-follower) or what we call gyrator load si close to mu of the valve. About 8 for both valves you mention.The 01A uses Ale's gyrator, but has way too much gain, so needs a TVC or AVC* on the output, as it hums like mad with the pot on the input. (It is on the living room table at the moment being rewired internally with screened cable). How much gain would the 26 have with the gyrator instead of the LL2745?
* Hi there Andy.
If this is too much gain, you should look into 71a (which sounds great btw) or other lower mu DHT. 45 may be difficult on filament bias, but doable if you run it at a lower operating point to avoid too much heat on the filament resistor.
Don't add gain when you don't need it. In my experience this is not a good solution. You won't get the most out of the DHT in your system.
cheers
Ale
Don't know Andy. On the one hand, the soft start helps make sure the valve is conducting so operates within its maximum voltage rating (assuming properly designed). So if max voltage with no current draw is catered for the valve should remain within safe operation. But, will instantaneously shoving a few 100V on the DHT harm it, particularly with Rod's gentle filament heating?