I think the danger when a group of enthusiasts are asked an opened ended question such as 'what would you build if you were me', is that you will end up answers that represent the pre-conditions for each of the people that offer suggestions.
So that opens a variety of battle lines: SE vs PP, direct heating vs indirect heating, octal vs noval, P2P vs PCB, toroids vs EI, silcon vs sand-free, and so on.
I think you have to reflect on the myriad* of possibilities and direct your audience a bit. What construction style are you aiming for? What speakers do you have? What music do you like, and at what volumes?
Also, if you invest in decent power and output transformers, and power tubes, then that is 80% of your sunk cost if you rebuild it anyway, so you can consider it to be the start of a journey, rather than a short commute.
*myriad - your English 'mot du jour'. It means 'plethora'.
So that opens a variety of battle lines: SE vs PP, direct heating vs indirect heating, octal vs noval, P2P vs PCB, toroids vs EI, silcon vs sand-free, and so on.
I think you have to reflect on the myriad* of possibilities and direct your audience a bit. What construction style are you aiming for? What speakers do you have? What music do you like, and at what volumes?
Also, if you invest in decent power and output transformers, and power tubes, then that is 80% of your sunk cost if you rebuild it anyway, so you can consider it to be the start of a journey, rather than a short commute.
*myriad - your English 'mot du jour'. It means 'plethora'.
Last edited:
Tubecad
I am also a tube newbie and have found the calculators from Glass-ware audio to be excellent. The SE amp covers fixed and cathode bias, many tubes, and many transformers.
GlassWare Audio Design
I've used the SE calculator to validate my parallel KT77 SE amp output stage but there is one thing confusing me.
The schematic options all show the screen grid connected to the anode/plate. Whereas other websites such as: The Valve Wizard -Single Ended recommend it being significantly lower - 250 to 350v (B+ will be around 475)
Suggestions please.
GlassWare also have some power supply kits, one of which, the PS-11, is ideal for many tube amps.
I am also a tube newbie and have found the calculators from Glass-ware audio to be excellent. The SE amp covers fixed and cathode bias, many tubes, and many transformers.
GlassWare Audio Design
I've used the SE calculator to validate my parallel KT77 SE amp output stage but there is one thing confusing me.
The schematic options all show the screen grid connected to the anode/plate. Whereas other websites such as: The Valve Wizard -Single Ended recommend it being significantly lower - 250 to 350v (B+ will be around 475)
Suggestions please.
GlassWare also have some power supply kits, one of which, the PS-11, is ideal for many tube amps.
I am not one of the experts here, still an enthusiastic beginner, but there are two factors in play, I think.
The first is 'partition noise' due to the electron stream being interupted by the screen grid. If the screen is tied to the plate, then that effect is essentially cancelled.
The second is the fact that the the current that flows through the tube is largely controlled by the screen, but the screen itself can tolerate the dissipation of energy less than the anode. So there is a juggling act to find the right operating points at different grid potentials to balance power output and tube characteristics, for different amplification types (A, AB, AB2, etc).
And then there is ultralinear, a half way house.
The first is 'partition noise' due to the electron stream being interupted by the screen grid. If the screen is tied to the plate, then that effect is essentially cancelled.
The second is the fact that the the current that flows through the tube is largely controlled by the screen, but the screen itself can tolerate the dissipation of energy less than the anode. So there is a juggling act to find the right operating points at different grid potentials to balance power output and tube characteristics, for different amplification types (A, AB, AB2, etc).
And then there is ultralinear, a half way house.
I found a discussion on this thread.
Triode versus Pentode
With the (expected?) absence of a definitive answer.
I'll go with triode strapped pentode for now I think.
Triode versus Pentode
With the (expected?) absence of a definitive answer.
I'll go with triode strapped pentode for now I think.
This is mostly due to chassis size limitations, and in general it is quite bad practice to place PT so close to OPTs, according to Morgan Jones in "Building Tube Amplifiers". There is an incredible amount of detail in that book about iron/tube layout planning and all related factors (heat, interference/noise, etc.)
I will echo OldHector's recommendation of the Tubelab amps. Visit tubelab.com for details on the various options. Pretty-much idiot-proof as long as you don't inadvertently purchase defective components (long story). Buy one of the PCBs and all necessary components, then follow the build guide on the website and come here to ask the designer, George (and some others) for help if you get stuck.
I will contradict OldHector on one point. For me, the easiest (most comfortable/familiar) part was the chassis.
Last edited:
For the OP, have you considered purchasing an amplifier kit for your build? It could still give you the opportunity to get your feet wet with tube amps. and would give you a known, functioning circuit that should work right of the gate. You could also start to learn what changing various components can do to effect the sound of the amp.
There are kits that would fit your budget and it would save you time. Just a thought.
There are kits that would fit your budget and it would save you time. Just a thought.
I don't think so. Everyone who is familiar with the advanced winding techniques applied by the better audio transformers becomes visible in a split second that those couldn't be applied on a toroid design.
Thats the reason why nobody (expect some very small companies, very few) regards this principle of being prooven to build a high quality audio transformer.
And of course its cheap, because of its "advanced winding structure". I wonder wether a simple AC toroid would sound the same as this example of fine audio transformer build. That is cheap garbage. Just in my opinion. Try a hacked radio transformer for EL84 and it will have more quality. Cheaper tough. You can get the whole amp for $50 if you do it that way.
If I hadn't knowledge about those things, I could wonder what kind of hints being given in an audio forum of real "experts".
Don't believe whats written here. You may get completely lost in the wilderness, ending with just a fraction of whats possible in audio sound.
Last edited:
I did mention in an earlier post that once we combine with ideas to an open ended question 'an affordable amplifier to build', the battle lines would be drawn!
I think you are being somewhat harsh here. From my point of view, without the affordability of Toroidy products, I would never have been able to cut my teeth on different PP and SE builds.
And I don't think you should malign the performance until you've heard them. They probably will not compete with Plitron, Lundahl or Tango, but they have exceptional frequency performance down to low frequencies, and I haven't heard anyone complain who has made a purchase, and many have here.
There are some very safe recommendations one could make for a first tube amplifier build.
It's hard to mess up a push-pull EL84 (6BQ5) 'ultralinear' amp.
The Dynaco ST35 is probably the best known version on the USA side.
There are good examples in some Grundig consoles from the 1960s.
There was a Leak amplifier (British) that was very popular.
Start with a pair of suitable output transformers, such as the Dynaco Z-565 or the equivalent Edcor types (8k or 10k plate-plate primary with 'ultralinear' screen taps, 4 or 8 ohm secondary, about 25 watts rated).
Then get a suitable power transformer. 250-0-250 at 200mA would be about right if you use solid state rectifiers to convert AC to DC. Or use 275-0-275 200mA with a 5V winding for a suitable rectifier tube (5U4GA or GZ34).
Those are the guts of the design. Lots of options for the rest. The easiest driver topology would use the classic concertina or cathodyne phase splitter. That's what almost all of the classic PP EL84 amps used.
Tubelab_com has a PCB for exactly this type of circuit, the Simple Push Pull (SPP) pcb.
Simple Push-Pull (SPP) Board | Tubelab
There are others to choose from too.
--
It's hard to mess up a push-pull EL84 (6BQ5) 'ultralinear' amp.
The Dynaco ST35 is probably the best known version on the USA side.
There are good examples in some Grundig consoles from the 1960s.
There was a Leak amplifier (British) that was very popular.
Start with a pair of suitable output transformers, such as the Dynaco Z-565 or the equivalent Edcor types (8k or 10k plate-plate primary with 'ultralinear' screen taps, 4 or 8 ohm secondary, about 25 watts rated).
Then get a suitable power transformer. 250-0-250 at 200mA would be about right if you use solid state rectifiers to convert AC to DC. Or use 275-0-275 200mA with a 5V winding for a suitable rectifier tube (5U4GA or GZ34).
Those are the guts of the design. Lots of options for the rest. The easiest driver topology would use the classic concertina or cathodyne phase splitter. That's what almost all of the classic PP EL84 amps used.
Tubelab_com has a PCB for exactly this type of circuit, the Simple Push Pull (SPP) pcb.
Simple Push-Pull (SPP) Board | Tubelab
There are others to choose from too.
--
Last edited:
A lot has happened in the mean time in this thread.
The Z565 is mentioned a lot. Does anyone know where it is available in europe? It could save me shipping and customs costs.
I've narrowed all options down to either the El Cheapo or the Baby Huey. At least for now I'm very bad at paint by number, so I would like to understand stuff and I even might change 'm a bit. I think I will be using mostly point to point for audio and probably some pcb-like option for the CCS's.
I have considered buying a kit. But I've taken a brave-pil and decided not to. I've built a JTM45 from a kit that I changed and a JCM800 that I build from scratch. My feet are wet.
I would love to learn about output transformers. If anyone has a good website or book, please lead me to it. If anyone could point me affordable non-toroids available in europe, that would be nice.
Learning and planning is more than half of the fun for me.
The Z565 is mentioned a lot. Does anyone know where it is available in europe? It could save me shipping and customs costs.
I've narrowed all options down to either the El Cheapo or the Baby Huey. At least for now
I'm very bad at paint by number, so I would like to understand stuff and I even might change 'm a bit. I think I will be using mostly point to point for audio and probably some pcb-like option for the CCS's.I have considered buying a kit. But I've taken a brave-pil and decided not to. I've built a JTM45 from a kit that I changed and a JCM800 that I build from scratch. My feet are wet.
I would love to learn about output transformers. If anyone has a good website or book, please lead me to it. If anyone could point me affordable non-toroids available in europe, that would be nice.
Learning and planning is more than half of the fun for me.
Attachments
Last edited:
Don Audio sells a selection of Edcor transformers, shipping from Germany.
Lundahl & Edcor Transformer Distribution
--
Lundahl & Edcor Transformer Distribution
--
Start with a proven design.
Build it to spec.
Measure it.
Listen.
Do not modify it, until you have it working well, and have listened to it on a variety of music, and on more than one set of loudspeakers if you have 2 or more sets.
Then modify it (get some advice on the proposed modification(s) before changing anything).
If you change more than 1 thing at a time, then you will never know which 1 thing made it better (or worse).
Just my opinion.
Build it to spec.
Measure it.
Listen.
Do not modify it, until you have it working well, and have listened to it on a variety of music, and on more than one set of loudspeakers if you have 2 or more sets.
Then modify it (get some advice on the proposed modification(s) before changing anything).
If you change more than 1 thing at a time, then you will never know which 1 thing made it better (or worse).
Just my opinion.
Menno van der Veen's designs test and listen very well. He has published much of his thinking. This was "new" about 30 years ago but is no longer very controversial.
Home
About
Publications
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.