I believe in both schematics the load resistance is too low. This not only reduces stage gain, but, more importantly, increases distortion. I understand that there isn't enough available B+ for higher value load resistor, but plate choke would solve this problem.
DC link capacitor at power supply filter input makes a lot of sense. These capacitors have very low ESR and inductance, making them very effective in suppressing power line and rectifier noise. But in last filter capacitor position I prefer paper in oil with foil electrodes. Metallization causes a little bit of edginess. Metallized capacitors (and resistors) have "flicker distortion". As voltage across metallized surface increases, more metal particles are recruited to conductive layer. Solid metal films do not have this type of distortion.
That's how I would do it too, but I would use fixed bias in both stages and avoid bridge rectifier for filament supply.
What' your impression of the Lundahl interstage transformer?
Yes, I understand that plate chokes are technically a better solution. I'm just not convinced about the tonality of acoustic instrument which is my main consideration. That's why I don't use active loads - they add an artificial sound. Plate chokes are better than active loads in terms of tonality. Not sure they are as pure as resistors, maybe they can be.
I'm going to look more closely at the PSU caps. I suspect that the DC Link caps may need a small bypass. I've never believed in bypasses but I think these DC Link caps shift the frequency response down a little away from the high treble.
I can try the LL1660 in 2:4.5 at some point. I have some. This points to the 2P29L which has a slightly higher mu than the 112A.
According to my humble experience, the best I’ve used in a P/S were ASC oil can caps. And I also used a dual mode choke like in the following recommendation from the emission labs:
My amp is a SRPP-2A3 with tube rectification
My amp is a SRPP-2A3 with tube rectification
I have not built that circuit. But whenever I'm finding a two stage DHT circuit in my stash I've been posting it here.
I am following this thread with interest. As a mechanical engineer who can hold a soldering iron, I am not in a position to design my own, but nonetheless very interested in your input on the merits of several design choices.
I recently came into posession of a pair of 811A monoblocks designed by Acquablue. They are a three stage affair, with the 811A in cathode feedback. I am thinking of changing these to two stage to simplify things and because I use a preamp stage (Zenmod's Iron Pumpkin). There is some info and schematics here but I do have slightly different drivers in, first EF86 then KT66 to 811A. The transmitter tubes are running at 780V on an 11k OPT Lundahl LL9202.
A DHT driver would be nice, but should also have enough swing to drive the 811A. First improvement to make is dc filament regulators for the 811A, I have a pair of Rod's almost ready.
I recently came into posession of a pair of 811A monoblocks designed by Acquablue. They are a three stage affair, with the 811A in cathode feedback. I am thinking of changing these to two stage to simplify things and because I use a preamp stage (Zenmod's Iron Pumpkin). There is some info and schematics here but I do have slightly different drivers in, first EF86 then KT66 to 811A. The transmitter tubes are running at 780V on an 11k OPT Lundahl LL9202.
A DHT driver would be nice, but should also have enough swing to drive the 811A. First improvement to make is dc filament regulators for the 811A, I have a pair of Rod's almost ready.
In the schematic, the common mode choke will not reduce ripple - because ripple is differential mode.
Hi
Can you advise what values, or ratio of values, you would expect as appropriate for C1 and C2.
Cheers
Grant
I am using 4.7 uF for the first cap (obbligato) and 20 or 22 uF for the second (ASC), I don’t remember right now and the cap is painted.
So would you expect that this could be addressed by the following stages, for each channel, or would this need to be addressed prior to this?
Cheers
Grant
Yes. With parallel windings, the allowable current will be 2×, and inductance 1/4.
CM choke will block common signal in + and - lines, which is not a concern in tube amplifier power supply. Of course, filtering can be done by adding chokes and capacitors, but why wasting money on expensive Lundahl choke if it doesn't do anything?
Whether resistor is a superior load for voltage amplifier stage is an open question. At least it has to be a very good resistor. Either non-inductive wire wound, or bulk metal foil.
Non-inductive wire wound may have pretty high shunt capacitance, which may affect high frequencies. Bulk metal foil is close to ideal, but they are expensive, especially in wattage necessary for the purpose.
The common types, including metal film, carbon, and metal oxide have obvious flaws.
IMO, a typical resistor is not sound-neutral.
Non-inductive wire wound may have pretty high shunt capacitance, which may affect high frequencies. Bulk metal foil is close to ideal, but they are expensive, especially in wattage necessary for the purpose.
The common types, including metal film, carbon, and metal oxide have obvious flaws.
IMO, a typical resistor is not sound-neutral.
.... just a general question.... for those schematics in this thread that use a step up at the input of the amp... am I correct in assuming that the preamp or source before the all DHT is of very low impedance, like 100hms or less?
Thanks
Fm
Thanks
Fm
A typical CD player has output impedance of 100-150 Ohms. The output op-amp has much lower output impedance; 100-150 Ohms is from RF-blocking series resistor. This resistor can be replaced with RF coil, about 20 turns of 32 AWG magnet wire wound on 0.5 W carbon comp resistor - for even lower output impedance. Output impedance of less than DCR of input transformer primary doesn't result in any improvement.
I use ultrapath, or parafeed with fixed battery bias - so there is still a cap in the AC path. Find the space for those PIO caps in the amp is my challenge. But agree that PIO caps are great sounding.
I haven't tried battery bias recently and I can't remember why I stopped using it and preferred filament bias, but that was the case. I've been using filament bias ever since, so no cathode bypass caps. It does mean putting together a couple of Rod Coleman regs (from Lyrima) but the kits are cheap enough and they're very good, especially the new V9. I use Hammond chokes in the filament supply as well. At the moment I'm using a 112A input and liking it a lot - very modest filament requirements, no big transformers and heatsinks like 26, 10Y and 46. It's a step up from the 2P29L.