It's possible there is a lot of RF in the area. It's weird though, I've never had any problems with previous builds being used with no lid on. In fact, I just put together a 6-24 Biamp active crossover and it's just sitting on the shelf, no enclosure. And it works just fine.
I read your thread on that Nobsound preamp a couple of months ago and considered trying the 2J27L. But I don't think it's low plate current would be a good fit for driving a mosfet's gate directly. That and Ale Moglia's comments on it. I felt the 2P29L was a more worthwhile pursuit.
I read your thread on that Nobsound preamp a couple of months ago and considered trying the 2J27L. But I don't think it's low plate current would be a good fit for driving a mosfet's gate directly. That and Ale Moglia's comments on it. I felt the 2P29L was a more worthwhile pursuit.
Actually, it's not . . . as I mentioned in the first post of the thread I linked. "They claim that the tube is being operated at a higher voltage, supposedly 105v on the plate. When I breadboarded its cousin, the 2P29L, I was running very similar voltage (112v on the plates) and many DIY builders run 2P29Ls in the same range, sometimes even a bit under 100v. So, if the tubes are operated at normal voltages, that's another positive."
If you read through that thread you'll see a lot of very positive comments. No doubt it's possible to build something using the same tube, the similar 2P29L or something else that's better but for only $50 to $60, which includes shipping, the Nobsound E6 is worth checking out, IMO. As I mentioned, I was skeptical too, so I got mine through Amazon because they offer totally free returns.
And, while I've also been critical of the low voltage operating point used in the popular FX and its cousins, they are not "pseudo tube preamps". They do have gain so they are preamps, not buffers. And the gain is provided by the tube, not a transistor. So, despite the fact that they use sub-optimal operating points, they are real tube preamps. No, I don't own an FX but I breadboarded it and I was not impressed, even when I ran the tubes at normal voltages.
Last edited:
Looking for the simplest implementation of 2P29L I ended up with this circuit. I used a small 12+12v 12VA transformer for the filaments which gave me 16.7V. No heatsinks necessary on my 3mm aluminium top plate. I used the older V4 version of Rod Coleman regs but building fresh you would use the V9 which needs 3.5V headroom rather than 6V. This would allow for a higher bias voltage like 10V on the cathode and the rest would be altered accordingly. No rectifier tube, just 1200V Schottky diodes. Simple 230/230V transformer for the B+. About as easy to build as possible. You can also put a 1:4 step-up in front of it and drive something like a SE 2a3 stage. Very nice.
Thanks, Merlin. The other reason I built this 2P29L stage is to replace my beloved 10Y driver stage in my 2a3/6C4C 2 stage SE amp when the weather gets hot. My listening room is also my bedroom and heatsinks warm up the room. The 10Y has quite a hot heat sink since it's filament bias. When the heat gets above 25° I have difficulty sleeping, and I leave my system on 24/7 because I often play music quietly to send me to sleep. I also have a 2a3 output stage with AC heating to eliminate another heat sink. I usually use my 6C4C output stage with Rod's regs and, of course, that has another heatsink.
What do you do about the heat in Spain in Summer? I know you use tubes.
What do you do about the heat in Spain in Summer? I know you use tubes.
Hello Andy, about the heat weather more complicated like in your country fortunately I live in Girona - Catalonia north in Spain, about power amps can't use the Erno Borbely monos SS 75W RMS in it's place I use a surprisely Tripath TA2022 of course modified the input cap for Auricaps & the output filter change with a low cutoff the corner frequency also with Auricaps, also a potent R-core 120VA, it's other amp, never see the Borbelys moving the 10" woofers of my DIY 3 way loudspeakers. About DIY lately I'm setting up the 26 with Bela help, the PSU tx is a Bartolucci 125VAC resulting a little bit headroom of B+ to use regulation so opted for a cascode CCS, un fortunately can't use the hybrid mu-follower from Ale nor the great SSHV2 from Salas if do not change this power tx, my 26 uses filament bias with a mix of 2R2 Mundorf M-Resist Supreme & 3 Sic because don't like full Sics.
Last edited:
I put the 2P29L stage into my system in place of the 10Y stage and it's surprisingly good. I've said before that it is 90% of a 10Y and I still think that's about right. Not quite the magic but otherwise a similar sound. So my system is now DAC > 1:4 step-up > 2P29L > 6C4C. Both in filament bias. Very simple, very cheap, sounds wonderful.
I haven't used the 4P1L for several years. It has its good points - good treble and decent bass. I found it rather hollow sounding in the midrange. This can be improved but not completely cured by putting at least 25mA through it and even up to 35mA. Not something I want to do so I'll pass on that. The 2P29L has a balanced sound without any hollowness in the mids. The treble is quite good and overall it's quite satisfying. What it lacks is low level detail and transparency in the midrange. The 10Y does this better than any other DHT I've used. Loads of detail, wonderful vocals. In comparison the 2P29L has a relatively "thick" midrange. If vocals are vital to you then you will choose the 10Y even though it's a much more expensive and challenging build, especially in filament bias which gives you the cleanest sound. Actually, going from the 10Y to the 2P29L in my system makes me yearn to put the 10Y back. I just love voices with the 10Y and for voices I'd put the 2P29L at more like 80% of a 10Y.
The 112A is quite an easy build in filament bias. It also has this slight thickening to the mids compared to the 10Y but it's a slightly better sound than the 2P29L. I may build a stage using that for my "no heat" summer system if the weather gets really hot.
The 26 is frustrating - it does have its own magic and is very good on vocals. It's a more challenging build in filament bias, and requires heatsinks which get hot. Cheaper than the 10Y but overall not as good if you're already going to make a 1 amp filament supply. The sound has a kind of warmth that I don't really want - I like a leaner, more transparent sound which is why I use a 2a3 rather than the warmer sounding 300b. This is all subjective of course, and I should point out that my listening is classical, opera, jazz and singer-songwriters. So the timbre of acoustic instruments is my highest priority, together with transparency and low level detail. The 10Y does all of this. I suppose I should mention the 46/47/49 which are all good and worth exploring. I lived for a while with a 46 stage which was very dynamic and quite involving, but the gain was a little less and in any case the 10Y had better low level detail. The 46 in filament bias had big heatsinks and cathode resistors which got hotter even than the 10Y.
The 112A is quite an easy build in filament bias. It also has this slight thickening to the mids compared to the 10Y but it's a slightly better sound than the 2P29L. I may build a stage using that for my "no heat" summer system if the weather gets really hot.
The 26 is frustrating - it does have its own magic and is very good on vocals. It's a more challenging build in filament bias, and requires heatsinks which get hot. Cheaper than the 10Y but overall not as good if you're already going to make a 1 amp filament supply. The sound has a kind of warmth that I don't really want - I like a leaner, more transparent sound which is why I use a 2a3 rather than the warmer sounding 300b. This is all subjective of course, and I should point out that my listening is classical, opera, jazz and singer-songwriters. So the timbre of acoustic instruments is my highest priority, together with transparency and low level detail. The 10Y does all of this. I suppose I should mention the 46/47/49 which are all good and worth exploring. I lived for a while with a 46 stage which was very dynamic and quite involving, but the gain was a little less and in any case the 10Y had better low level detail. The 46 in filament bias had big heatsinks and cathode resistors which got hotter even than the 10Y.
The 6SN7 is a good sounding valve, amongst the best of the indirectly heated tubes. The best version of it, and I have tried several, is 2x CV1135 aka CV6. Note the anode and grid pins are reversed from the more common 7193/2C22 and it's better sounding. With 6SN7 I like the angled anode versions.
I should also mention the European radio valves like REN904 and TDD4. I have quite a collection of them, and they are very good too.
I think there's only so far you can go with indirectly heated tubes before good DHTs take over. Small signal tubes like 26, 46/47/49, 10Y, 112A, 01A are on another level. I'd prefer a 2P29L to a 6J5 type and then you just go up from there.
I should also mention the European radio valves like REN904 and TDD4. I have quite a collection of them, and they are very good too.
I think there's only so far you can go with indirectly heated tubes before good DHTs take over. Small signal tubes like 26, 46/47/49, 10Y, 112A, 01A are on another level. I'd prefer a 2P29L to a 6J5 type and then you just go up from there.
As you have an input transformer you can use fixed bias. No need for filament bias. You can use the old back-bias trick. From the plate supply, ground the second capacitor and put a resistor in the negative leg between the first and second capacitor of the pi filter. All current supplied by that PS transformer secondary will go trough the resistor and will create a negative voltage that you can filter and use for biasing.
Maybe try battery grid bias with a 9v battery? I used this with the 26 in my Nuance iSET that I built last year. Super simple.
I've never done any A-B comparisons with other methods but it sounds good to me. I did come across an analysis somewhere that indicated that it resulted in slightly lower measurable distortion vs a resistor. I'm also using it on my current project which uses the Hammond SUT arrangement with a different DHT.
Various combinations of batteries can be used for different tubes. Battery life is essentially their shelf life if no grid current is drawn. Probably 9 or 10 years for a Lithium or 6 or 7 for an Alkaline.
Similar to the instigator of this thread, I too am very much a cut and paste builder with only basic electronics knowledge, so please excuse my 'noddy' question.
I've been building a 2P29L preamp based on the Bartola schematic, attached below for reference. I have a 50K stepped shunt attenuator I would like to use as a volume control on the input, if possible - are there any implications to the the functioning of the preamp with doing that?
Thanks for your help.
I've been building a 2P29L preamp based on the Bartola schematic, attached below for reference. I have a 50K stepped shunt attenuator I would like to use as a volume control on the input, if possible - are there any implications to the the functioning of the preamp with doing that?
Thanks for your help.