I have used 6P14P-EV and 6P15P-EV in the amp so far. The 6P15P-EV sounds the best. I have a Decware SE84CS clone that runs the 6P15P-EV tubes at a bit higher voltage and more current than this amp does. This amp has run this way without issues for many years. The 6P14P-EV are a ruggedized EL84 and an EL84M equivalent.
Plate voltage is only one part of the equation. Dissipation is the more important measurement, which is based on both plate voltage and current.
The 6BQ5 / EL84 has a dissipation limit of 12w. According to his schematic, it is only dissipating 9.79w, so not even close to being a problem.
The hiss is puzzling. When I breadboarded a basic preamp a while back I tried numerous tubes, including some that I heated with a battery, and there was no hiss or noise whatsoever. So I would suspect there is something else that's causing the hiss.
Try to wiring 6,3V AC temporary with "flying" wires from resistors to EL84s filament pins.
The red/yellow wires glued it to metal plate, in may be inducing some hum, and the whole top plate like a transformer core.
The "gain" from #30 cathode to the output of transformer about 3, so if yo inject 10mV hum to the cathode, it products 30mV hum on output.
The first stage PSRR about 85dB, so the source of hum probably in the filament.
Last edited:
FlaCharlie: Other than switching from DC to battery filaments, nothing else was changed. With DC and a capacitor across the cathode resistor there was a very faint hum at the speaker. I thought there wasn’t initially, but I was listening with the room’s door open and the ambient sound in the rest of the house drowned it out somewhat. There is still a very faint hum with the cap, right at the speaker, with one channel having a bit more than the other, but no hiss. With the battery there is no hum at all but hiss that is obvious at the speaker, and just audible at the listening position.
To be fair, the La Scalas I am testing with have much higher than notmal sensitivity. I will try with a set of Heresy speakers soon, but I doubt it will be audible at the listening position with them as the speakers.
euro21: I suspect the filament transformer may have had one of its primary winding disconnected. I will try using a new transformer to see if that fixes the majority of the previous problem.
I will also try to run flying wires to the EL84 heaters to eliminate that as a possibility. I double checked yesterday, and the plates that the 30 sockets are mounted to are definitely floating with respect to ground.
Although the EL84 heater wire runs close by, there is no electrical connection between the chassis and the socket. I had tried to ground the socket plate as well with no change in hum.
Today, I tried using the repaired transformer (without the capacitor on the cathode of the 30 tubes), and there was an improvement back to the lower level hum that was present prior to moving things around and accidentally and unknowingly breaking one of the primary terminals. The louder buzzy hum was almost certainly due to this damage to the filament power transformer. The buzzy hum with the repaired transformer returned to being obviously present with the music off, but much quieter than with the broken transformer.
Rod Coleman recommends building the raw filament power supplies with a power transformer per channel, and I have been using one transformer per channel so far. Once I established that I had repaired the broken power transformer, I added a second identical transformer so that each channel of the filament regulator had its own completely separate raw supply. As with the test with the single power transformer, I tested with the two transformers not attached to the raw supply board and sitting about 6 inches under the raw supply board on the counter below the raised chassis top plate. This resulted in more hum than when using a single transformer. I think that this may be because there is more transformer magnetic field to couple with the circuit.
I believe that this amp isn't going to ever be completely quiet without the cathode resistor bypass capacitor because of it being built on a magnetic steel plate. If I had built on an aluminum plate or box, I think it would have resulted in a much quieter amp.
Now that i have reduced the buzzy hum down to its prior quieter level, I will put back the capacitors on the cathode resistor of the 30 tubes and check again for residual hum and buzz.
This is my favourite DIY amp so far. I spent some time listening to it with the capacitor in it last night. It really makes musicians sound like they are in the listening room with you. Vocals and instruments, especially acoustic instruments, sound very real. I compared it briefly to three other amps on the rack. a Decware SE84CS clone, a single ended KT120 amp based on the Alex Gendrano KT88 circuit, and a DIY version of the FI 421A amp using a 5998 as the power tube. It is most similar to the SE84CS clone. The KT120 and the 5998 amp have obviously better bass, with more quantity and quality. This amp has a more detailed and revealing midrange and treble, and sharper transients than the other amps, although the SE84CS is closest.
Rod Coleman recommends building the raw filament power supplies with a power transformer per channel, and I have been using one transformer per channel so far. Once I established that I had repaired the broken power transformer, I added a second identical transformer so that each channel of the filament regulator had its own completely separate raw supply. As with the test with the single power transformer, I tested with the two transformers not attached to the raw supply board and sitting about 6 inches under the raw supply board on the counter below the raised chassis top plate. This resulted in more hum than when using a single transformer. I think that this may be because there is more transformer magnetic field to couple with the circuit.
I believe that this amp isn't going to ever be completely quiet without the cathode resistor bypass capacitor because of it being built on a magnetic steel plate. If I had built on an aluminum plate or box, I think it would have resulted in a much quieter amp.
Now that i have reduced the buzzy hum down to its prior quieter level, I will put back the capacitors on the cathode resistor of the 30 tubes and check again for residual hum and buzz.
This is my favourite DIY amp so far. I spent some time listening to it with the capacitor in it last night. It really makes musicians sound like they are in the listening room with you. Vocals and instruments, especially acoustic instruments, sound very real. I compared it briefly to three other amps on the rack. a Decware SE84CS clone, a single ended KT120 amp based on the Alex Gendrano KT88 circuit, and a DIY version of the FI 421A amp using a 5998 as the power tube. It is most similar to the SE84CS clone. The KT120 and the 5998 amp have obviously better bass, with more quantity and quality. This amp has a more detailed and revealing midrange and treble, and sharper transients than the other amps, although the SE84CS is closest.
With the cathode resistor bypass capacitor added there was a tiny bit oh low hum left over. Not buzzy, just the slightest hum with my ear at the speakers. The metal shields that I had tried prior without success were still next to the amp, so I put them on the 30s again. With the shields on, the hum was eliminated, and the amp is silent with the inputs shorted.
Congratulations! I am personally very happy that you had been extremely patient and finally dealt with the hum issue.
Next to try (may be someday) :
1. Filament bias
2. 3A5 drive 6v6
3. Thoriated tungsten output like 10Y/vt25 (good for 1.5 watt), I use this as my second
amp however 1.5 watt not enough for my speakers
4. Pseudo choke input PS like 2uf film cap 10h choke 220uf cap resistor 100uf cap resistor 47uf
cap resistor 3/5uf film cap last rc should be split for each channel.
This is bliss don't you agree? Please never say never, there always is room to improve.
Regards
Next to try (may be someday) :
1. Filament bias
2. 3A5 drive 6v6
3. Thoriated tungsten output like 10Y/vt25 (good for 1.5 watt), I use this as my second
amp however 1.5 watt not enough for my speakers
4. Pseudo choke input PS like 2uf film cap 10h choke 220uf cap resistor 100uf cap resistor 47uf
cap resistor 3/5uf film cap last rc should be split for each channel.
This is bliss don't you agree? Please never say never, there always is room to improve.
Regards
Thanks! I am very pleased with the outcome. The amp sounds great. I’ll build a separate amp to try out the different approaches to plate load and biasing.
For thoriated tungsten, a 01A driving an EL84, could also be very nice. The 01A has an amplification factor of 8, the 30’s is 9.3. The 30 is enough to drive the EL84 into audible clipping at full volume on program material. At clipping it is loud in the room. The 01A should be able to do close to the same and impart it’s character to the EL84.
With a 6V6 in my amp there isn’t enough gain for the 6V6 to run out of steam. It should be fine with a 6V6 when operating as treble amp though. My favourite output tube in this amp is the 6P15P-EV.
The amp stands up very well as a full range amp with its current OPTs. When I start using it as a treble amp, I will probably try the AliExpress OPTs in it again.
The pseudo choke input approach is something that I actually try to do on occasion. I often include a choke in my power supplies, so where appropriate, I try choke input. I haven’t come across a PT/choke combo that didn’t mechanically vibrate. This vibration is the loudest thing in the room when the music isn’t playing. I didn’t ty it in this amp, and it’s probably going to cause problems with the steel chassis.
If you increase the bias voltage (decrease the anode current) the #30 will be even more unusable in this situation.
Even at 5-6V RMS capable able to accomplish this grid swing.
The "best scenario" is fix biased 10Y SE (at this B+, only A1), but in this case the A1 border at 3V RMS input.
Even at 5-6V RMS capable able to accomplish this grid swing.
The "best scenario" is fix biased 10Y SE (at this B+, only A1), but in this case the A1 border at 3V RMS input.
Attachments
Good to hear that you've gotten rid of the hum, although it's a shame you have to hide the 30s with a shield.
So, did you go back to using a single filament transformer or are you still using a pair?
If you want to eliminate the electrolytic cathode bypass, try using fixed bias with batteries on the grid. You will need to add DC blocking caps to the input. But, IMO, replacing a large value lytic with a small value film cap is a good trade off - especially if the tube is directly heated. I use no electrolytics in my entire amp.
I'm heating both input tubes with a single Meanwell. Using battery grid bias, cheap Hammond plate chokes and tube regulated plate supply on the input tubes and I'm getting no hum or buzz.
Of course, my bench speakers aren't nearly as efficient as yours. And I never use steel chassis, unless I'm modding an existing commercial amp, although I've never had any issues. It's just harder to drill holes in using a hand drill, so I use either aluminum or wood.
I tried the 01A in my amp. It sounds nice and of course the glow is cool. But it seemed to emphasize some of the high mids a bit too much on some recordings. I also liked the 864 a lot but I went back to the 26. YMMV, of course, due to all the differences in the rest of the system.
I would suggest you try all of them and compare with the 30. Since their characteristics are so similar the circuit changes are minimal.
I'm surprised you're not suggesting that he try the 10/VT-25 in place of the 30.
I have a pair which I plan to use in a stand-alone preamp. I keep hearing that they are the ultimate DHT preamp tube. If so, I don't want to limit their use to one particular amp.
euro21: Thanks for the simulation. If I use a thoriated tungsten tube, it will be an 01A in the driver position for a IDHT output tube. The 10Y and like look great, but they are expensive, and I don’t have any handy. Someday maybe.
An amp with a 01A in the first position is appealing. I would love to try using a 01A with a 5998. Either one 5998 for both channels or one for each channel in parallel. I would need a tube in between to drive the 5998 sufficiently for a three stage amp.
This leads me to my next question. If I can’t build a two stage amp when using the 01A, wouldn’t it make more sense to build a 01A preamp, and combine it with whatever two stage I choose? I have only built one tube preamp, which uses two 6CG7s and has a cathode follower. I use this as a preamp for the solid state amps I have that don’t have a volume control. It doesn’t have much of a sound signature. To what degree do the qualities of an 01A preamp come through when used with a regular IDHT or a solid state amp?
Last night I tried listening with shield on and off. The slight hum without the shield isn’t audible at the listening position. The shields make it quiet at the speaker, but something sounds off with them on. I did most of my listening last night with the shields off.
I am using a single filament transformer, as two made the hum worse. The transformer is nicely over specified for the application, and cheap. The Hammond 229B16 is rated at 8 volts and 700 mA per winding and the tubes use 60mA per tube. 120 mA from an available 1400 mA.
Canon is that capacitors are bad, and electrolytics are especially bad in tube amps. I generally try to avoid electrolytics as well and have a few amps that only use them on cathode resistor bypass because of the physical size required. That said, this amp sounds fantastic. It has been a bit of a revelation actually. It has a holographic quality that is unlike anything else I have built. I am using Elna Silmic II capacitors on the 30s.
I added a small film cap at the input as I had removed it during trouble shooting, but it changed something for the worse. It lost something with the input caps so I removed them. Imagination? Maybe...
Last edited:
The 01a capable to swing 40Vpp at 2V RMS input (active load, 300V B+) with -about- 0.4% THD.
All my everyday speakers have a horn for the highs. Generally, a hot mid or treble is problematic with these speakers. With the thirty, there does seem to be a mild treble emphasis as compared to my other amps, but it isn’t grating, shrieky or annoying. I feel like this amp can be run louder without it hurting my ears. jhstewart9 is going to give the amp a looking over in the next few days. He will be able to properly measure and I’ll post the results.