Looking to be talked off a ledge or to be encouraged to jump. I tend to go a little overboard, so I want to make sure I'm only biting off only a little more than I can chew. I'm all for a bit of challenge but not in for a suffer-fest or pounding my head against EE nuances.
Background: I have very limited diy experience with a bottlehead crack and a pass aleph j with a dual mono fed power supply. I had a 45 SET that I bought but has since been sold with the idea I'd replace it with something I built. Speakers are single driver 98db / 8 ohm Omega 8" floor standers. (Does anyone have an impedance curve?)
Desirements; Not too complicated of a circuit, Single Ended, ~2-6wpc (I don't listen loud), but with good frequency response. One of the reasons I sold the 45 was it rolled off a little too much for me on both ends but especially on low frequencies. With the help of another member here, a frequency response plot confirmed what I was hearing (or rather not hearing). Also, this is as much about building a decent sounding amp as it is about learning and getting a feel for what some other tubes might sound like. My hi-fi experience is very limited. I value soundstage, presence, and relaxed sound I can sink into over an ultra precise reproduction or sparkling highs. I listen to all sorts of music, but mostly rock (60%?) with the remainder evenly split across things like singer/songwriter vocals, jazz, classical, opera, country/folk, and electronic.
Initial Thoughts: Originally I was thinking about buying another bottlehead kit, and will likely do that in the future, but I came across this page to build a seemingly straightforward EL84 based amp in a switchable UL/Triode config called the "El Mighty Cacahuate". Started to think I'd build it straight from the schematic but then I realized I have a few 6SN7's leftover from that other amp as well as one or two 5Y3's that I could use to reduce cost. One of the 5Y3's is actually a 6106 Bendix IIRC. Then I started to look at the OPT he spec'd and wasnt super keen on it due to its low Inductance and how that would roll the low frequency, like what my 45 used to do. Then I got to reading and am wondering about switching the EL84's for EL34's for a bit more power, and it seems to be preferred over it's small sibling, and then all would be right in the world with this OCD mechanical engineer with using 4 octals.
With the EL34's this has led me to looking at something like a Hammond 1630SEA or Edcor CXSE25-3.75K with an OP of around ~315-330v B+ / 70mA / -15v g1, depending on if I use the 5Y3 or stick with the 5U4. I'm also having trouble finding the triode tube curves for the EL34 to know if I've got the right values for all of this. Does this all look right / can anyone point me to one? Further, the PT spec'd on that blog will also be upgraded to one with a higher current rating for the EL34's.
All said and done, it's turning into something that will require using iron that will likely cost ~as much as it would have taken to complete originally as shown on the blog. Hence the part earlier about going a little overboard on my projects. This EL34 would likely make a good compliment to whatever kit I'd get from bottlehead down the road using either a 2A3 or 300B. Am I being rational here? Should I go for it or is there another amp to consider? Right now one with a good schematic and/or a few pictures of the completed design would really help.
Background: I have very limited diy experience with a bottlehead crack and a pass aleph j with a dual mono fed power supply. I had a 45 SET that I bought but has since been sold with the idea I'd replace it with something I built. Speakers are single driver 98db / 8 ohm Omega 8" floor standers. (Does anyone have an impedance curve?)
Desirements; Not too complicated of a circuit, Single Ended, ~2-6wpc (I don't listen loud), but with good frequency response. One of the reasons I sold the 45 was it rolled off a little too much for me on both ends but especially on low frequencies. With the help of another member here, a frequency response plot confirmed what I was hearing (or rather not hearing). Also, this is as much about building a decent sounding amp as it is about learning and getting a feel for what some other tubes might sound like. My hi-fi experience is very limited. I value soundstage, presence, and relaxed sound I can sink into over an ultra precise reproduction or sparkling highs. I listen to all sorts of music, but mostly rock (60%?) with the remainder evenly split across things like singer/songwriter vocals, jazz, classical, opera, country/folk, and electronic.
Initial Thoughts: Originally I was thinking about buying another bottlehead kit, and will likely do that in the future, but I came across this page to build a seemingly straightforward EL84 based amp in a switchable UL/Triode config called the "El Mighty Cacahuate". Started to think I'd build it straight from the schematic but then I realized I have a few 6SN7's leftover from that other amp as well as one or two 5Y3's that I could use to reduce cost. One of the 5Y3's is actually a 6106 Bendix IIRC. Then I started to look at the OPT he spec'd and wasnt super keen on it due to its low Inductance and how that would roll the low frequency, like what my 45 used to do. Then I got to reading and am wondering about switching the EL84's for EL34's for a bit more power, and it seems to be preferred over it's small sibling, and then all would be right in the world with this OCD mechanical engineer with using 4 octals.
With the EL34's this has led me to looking at something like a Hammond 1630SEA or Edcor CXSE25-3.75K with an OP of around ~315-330v B+ / 70mA / -15v g1, depending on if I use the 5Y3 or stick with the 5U4. I'm also having trouble finding the triode tube curves for the EL34 to know if I've got the right values for all of this. Does this all look right / can anyone point me to one? Further, the PT spec'd on that blog will also be upgraded to one with a higher current rating for the EL34's.
All said and done, it's turning into something that will require using iron that will likely cost ~as much as it would have taken to complete originally as shown on the blog. Hence the part earlier about going a little overboard on my projects. This EL34 would likely make a good compliment to whatever kit I'd get from bottlehead down the road using either a 2A3 or 300B. Am I being rational here? Should I go for it or is there another amp to consider? Right now one with a good schematic and/or a few pictures of the completed design would really help.
Head on over to Tubelab.com and read thru the SSE section. Just might be what you're looking for to run EL34's single ended.
Tubelab SSE | Tubelab
jeff
Tubelab SSE | Tubelab
jeff
EL34 hint:
Connect the Suppressor Grid pin 1, to the Cathode pin 8.
Most other Beam Power tubes and Pentode tubes in similar amplifiers do not connect
pin 1 to pin 8.
Because most other Beam Power and Pentode tubes already connect the Beam Formers or the Suppressor Grid to the Cathode Internally inside the tube.
Some marketing types and designers suspect that someone will "tube roll" an EL34 into an amplifier that was not originally designed for an EL34.
So they connect pin 1 to pin 8 just in case somebody does that.
Connect the Suppressor Grid pin 1, to the Cathode pin 8.
Most other Beam Power tubes and Pentode tubes in similar amplifiers do not connect
pin 1 to pin 8.
Because most other Beam Power and Pentode tubes already connect the Beam Formers or the Suppressor Grid to the Cathode Internally inside the tube.
Some marketing types and designers suspect that someone will "tube roll" an EL34 into an amplifier that was not originally designed for an EL34.
So they connect pin 1 to pin 8 just in case somebody does that.
Totally agree! I have SE and SSE and enjoying the amps since 2011. George (Tubelab) is amazing supporting his product! Documentation from the website and from the users made it easy to complete the project for me when I was building the amps.
Thanks for the suggestions everyone.
Am I that far off though? The Tubelab SSE does look appealing, but I think I want to do a little more of this myself. I have a tendancy to "shut off my brain" when stuffing a PCB and sometimes have trouble seeing how things actually hookup, i.e. it provides a lower learning experience for me. I may rely on its schematic a bit though, so thanks for passing along! It also seems like he's running all of these tubes SUPER hot. Well above 30w for an EL34 it seems. I thought the whole idea was to have it operate ~10-20% below peak plate dissipation, not above, in order to have a stable long lived tube.
On the Blueglow I notice that it's using a much high operating point, 450 to my 300 and using a 5k instead of a 3k'ish OPT. If I understand it right, the 41v dropped over the 500ohm resistor sets the plate current at 82mA; that's way over the 25w dissipation an EL34 is rated. Right?
The schematic, generally speaking, is on that website I linked. I'd be using a 6SN7 instead of the 6CG7 9 pin mini. Perhaps using a 5Y3 instead of a 5U4, and therefore also adjusting the PSU filter per what PSUD2 tells me. The big change being the first filter cap reduced to a value more compatible, something ~10uF or below as I understand it. And then swap the EL84 for EL34 and adjust cathode bias resistor for appropriate operation.
For ease of reference, here's the schematic from wtfamps website. I hope it's ok I link it here directly.
I'm also now reading on the Tubelab SSE website about allowing for some CFB. I guess that makes the Hammond OPT's a little more appealing with various secondary taps, though on his website he mentions these bigger hammonds not playing nicely with more efficient speakers. No idea what his definition of efficient are, but I'm assuming 98db would count.
Am I that far off though? The Tubelab SSE does look appealing, but I think I want to do a little more of this myself. I have a tendancy to "shut off my brain" when stuffing a PCB and sometimes have trouble seeing how things actually hookup, i.e. it provides a lower learning experience for me. I may rely on its schematic a bit though, so thanks for passing along! It also seems like he's running all of these tubes SUPER hot. Well above 30w for an EL34 it seems. I thought the whole idea was to have it operate ~10-20% below peak plate dissipation, not above, in order to have a stable long lived tube.
On the Blueglow I notice that it's using a much high operating point, 450 to my 300 and using a 5k instead of a 3k'ish OPT. If I understand it right, the 41v dropped over the 500ohm resistor sets the plate current at 82mA; that's way over the 25w dissipation an EL34 is rated. Right?
The schematic, generally speaking, is on that website I linked. I'd be using a 6SN7 instead of the 6CG7 9 pin mini. Perhaps using a 5Y3 instead of a 5U4, and therefore also adjusting the PSU filter per what PSUD2 tells me. The big change being the first filter cap reduced to a value more compatible, something ~10uF or below as I understand it. And then swap the EL84 for EL34 and adjust cathode bias resistor for appropriate operation.
For ease of reference, here's the schematic from wtfamps website. I hope it's ok I link it here directly.
I'm also now reading on the Tubelab SSE website about allowing for some CFB. I guess that makes the Hammond OPT's a little more appealing with various secondary taps, though on his website he mentions these bigger hammonds not playing nicely with more efficient speakers. No idea what his definition of efficient are, but I'm assuming 98db would count.
This is definitely a WTF amp. No decoupling of driver and output stages, either. This is a first attempt by a 14 yr. old. You don't need a 6H choke after a 33uF cap.
Last edited:
So is the blueglow amp that was recommended here not decoupled either? Unless I'm missing something, they look similarly laid out, with the blueglow actually have some feedback from the output back to the driver. Additionally, that schematic also shows a 10H choke after its first filter cap.
Why is that one any better?
Not arguing for one or the other, just trying to understand.
Why is that one any better?
Not arguing for one or the other, just trying to understand.
The Blueglow amp driver stage is decoupled by the added 4.7K and 33uF cap ahead of the 39K driver plate load resistor. And the PS has a 10uf cap ahead of 10H choke, as opposed to 33uf. But the Blueglow amp isn't built for precision, it's built to play swap games. It's not a newbee amp. You need a scope and sig. gen. to get the most out of that amp while you have fun pullin' and pluggin' and switchin'.
WTF schematic:
The 5U4 is a direct heated rectifier (no cathode). It starts rectifying before the slow warmup EL84 draws any current.
The same thing for a slow warming EL34.
The B+ may reach close to 1.414 x the 300V (420V), minus less than 40V rectifier drop (only drops 44V at 225mA, far less with no load current).
And with 300V x 1.414, and about 40V drop in the 5U4, When the power supply is warm and stabilized, it will put out lots more than the 315V that is written on the schematic.
I bet that first cap is 3.3uF, not 33uF. That makes the B+ act about 1/2 way between a cap input filter, and a choke input filter.
Change the 33uF to 3.3uF, and see if the power supply comes closer to 315V. If it is still high, try 2uf, then 1uF for the first cap . . . you will get there.
The 5U4 is a direct heated rectifier (no cathode). It starts rectifying before the slow warmup EL84 draws any current.
The same thing for a slow warming EL34.
The B+ may reach close to 1.414 x the 300V (420V), minus less than 40V rectifier drop (only drops 44V at 225mA, far less with no load current).
And with 300V x 1.414, and about 40V drop in the 5U4, When the power supply is warm and stabilized, it will put out lots more than the 315V that is written on the schematic.
I bet that first cap is 3.3uF, not 33uF. That makes the B+ act about 1/2 way between a cap input filter, and a choke input filter.
Change the 33uF to 3.3uF, and see if the power supply comes closer to 315V. If it is still high, try 2uf, then 1uF for the first cap . . . you will get there.
Last edited:
Put on your wing-suit, here's a push. If you have done the Bottlehead thing and learned from their orientation in the manual you know something about layout and you know something of the dangers and safety practices. (The only advice I'd make sure to listen to is about safety practices and electronics know-how.) Otherwise , just build the circuit you're interested in. It won't be perfect, that's a given, so you might as well like it as long as you can and then make it better. Everybody has their favorite circuit to recommend but that's only because they've already built it or a different one they're sure must be better.
If there's no chance of little curious fingers poking at it I'd recommend building the basic circuit on a piece of plywood or something so you're not locked into a final form by a chassis while you're still mapping unexplored terrain.
I also recommend you pick your build parts with some margin ie. power/voltage ratings for resistors, capacitors, diodes and transistors etc. so you have room to explore without having to rebuild the basic structure every time you want to try something new .. . . . and you will want to.
If you post updates the more knowledgeable ones here will keep you out of the weeds and the rest of us will learn something.
Thanks
El34 Triode Curves
EL34 Triode Curves
PS. Oh yeah, one other thing. Another transformer manufacturer for your list is Transcendar
The 45 itself wasnt so much the issue but the frequency response of the amp itself was a problem. I had a chance to listen to a "good" set of 45 monoblocks that someone here built, and for sure the 45 does some wonderful things. Suffice it to say, however, the aleph j also does enough good that I was willing to sell the 45. Who knows, maybe I'll end back at 45's down the road, but right now it's about learning, trying something new, and having fun 
Appreciate the encouragement and triode curve links also, Hearinspace! I'm not able to get to the ones on the protonbasic site, even after registering. I suppose it's probably too much to ask or assume that these curves live somewhere digitally where it can interpolate grid voltages and interactively tell me intersect points for a load line I draw?
Overall, I've decided I am going to do something, but just need to narrow down now, though I feel like spending a bit more upfront on nicer OPT's will allow some extra flexibility should I decide to scrap the EL34 circuit and go for a 2A3 or whatever. Between the Blueglow/wtf there's probably enough to go on.
Is there a HT voltage winding range and/or OPT impedance that's most commonly used that would allow use in another project down the road? Or is this too forward thinking?
Appreciate the encouragement and triode curve links also, Hearinspace! I'm not able to get to the ones on the protonbasic site, even after registering. I suppose it's probably too much to ask or assume that these curves live somewhere digitally where it can interpolate grid voltages and interactively tell me intersect points for a load line I draw?
Overall, I've decided I am going to do something, but just need to narrow down now, though I feel like spending a bit more upfront on nicer OPT's will allow some extra flexibility should I decide to scrap the EL34 circuit and go for a 2A3 or whatever. Between the Blueglow/wtf there's probably enough to go on.
Is there a HT voltage winding range and/or OPT impedance that's most commonly used that would allow use in another project down the road? Or is this too forward thinking?
If you are interested in a kit, since you built the Crack amp take another look at what Bottlehead has to offer.. They are fun to build, almost moron proof assembly instructions, very helpful forum and they sound great. Since you are near them, after the pandemic is over you can visit them with your speakers and audition their selection of amps.
Last edited:
The load line is a choice between power and linearity. Lower number will in general mean more power, higher number in general means less distortion - in general and within limits.
I use so called high efficiency speakers and therefore no need to quest for power so tend toward higher load impedances, though that's not laser engraved in iridium.
If you're going to do a lot of experimenting I think a 5K primary is not a bad value to have on hand (though you'll probably end up having a few different values in your tool kit). Get a decent output transformer so if you hit on a good circuit it can sound good, but don't spend on high end transformers yet. There's too much ground to cover and even if you are both technically proficient and like to listen carefully it will still take time to cover enough ground for an orientation towards what works for your ears . . . . and even that changes as you educate yourself. You could end up getting specialized in something you wouldn't begin to imagine right now.
My take on power transformers is to get something that you can go either up or down with. Capacitor input filter takes it up. Choke input takes it down. Recommended use varies with rectifier and filter type. I use the general guidelines from Hammond here, whether with their transformers or not. Note the current limitation with certain configs. Some current margin in the spec. is nice to have.
It's your lucky day!
Triode Load Line Calculator
I use so called high efficiency speakers and therefore no need to quest for power so tend toward higher load impedances, though that's not laser engraved in iridium.
If you're going to do a lot of experimenting I think a 5K primary is not a bad value to have on hand (though you'll probably end up having a few different values in your tool kit). Get a decent output transformer so if you hit on a good circuit it can sound good, but don't spend on high end transformers yet. There's too much ground to cover and even if you are both technically proficient and like to listen carefully it will still take time to cover enough ground for an orientation towards what works for your ears . . . . and even that changes as you educate yourself. You could end up getting specialized in something you wouldn't begin to imagine right now.
My take on power transformers is to get something that you can go either up or down with. Capacitor input filter takes it up. Choke input takes it down. Recommended use varies with rectifier and filter type. I use the general guidelines from Hammond here, whether with their transformers or not. Note the current limitation with certain configs. Some current margin in the spec. is nice to have.
It's your lucky day!
Triode Load Line Calculator
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.