You "rookies" are going to do some of the lifting. That's how you learn!
All the really tough stuff has been done.
TungSol's 7591 data sheet gives you the operating conditions. The condition's for the 12AT7 LTP are all over the WWW: IB = 3 mA. and 200 to 220 V. on the plate.Use Ohm's Law to calculate the value needed in a Carbon film part, in addition to the 50 KOhm wirewound, to drop 180 V. at 3 mA.
Run the ZVN0545A LTP buffers at approx. 2 mA., which stays well within the device's power dissipation limit. The source load resistor value needed is 120 KOhms/1 W. Use 220 nF./400 WVDC 716P series Orange Drops to couple the buffers to the "finals".
Now, get to work!
Figure out a PSU. Scrounge up some O/P trafos.BTW, I'm NOT a "guru". I'm just another hacker who has been around the block enough times to have a few things pounded into his head.
You "rookies" are going to do some of the lifting. That's how you learn!
Sorry man, I respect your idealism, but this is merely a build project, not a design project (read the datasheet). There's plenty of time to read books, learn and discuss the endless possibilities while waiting for parts to arrive and then while listening to the all-important project number 1.
It's a perfectly reasonable approach to getting one's feet wet and there's a lot to be learned by building/testing an amp first, and tackling the minutiae and math next. It's not the Eli approach, but then I'm not Eli.
Analogy: I'm looking for a decent enough recipe so that I don't starve while I'm studying French cuisine.
..Todd
Last edited:
sigh... Dude, El Cheapo doesn't fit the requirements. Not what I'm looking for. Again, read the datasheet.
I'm looking for beef stew, not lamb stew, tasty as it might be.
..Todd
Last edited:
I gave you a heads-up.
Now you just bein' rude to Eli, who has been more tolerant of your brashness than I. If you are simply going to be offhand and dismissive of the efforts of someone who is genuinely trying to help you then you won't find help so easy to get in the future.
Before you can be witty you'll have to convince us you've got wits. Until then it's just cheek. Go back and read my post again. You must have missed something.
w
Now you just bein' rude to Eli, who has been more tolerant of your brashness than I. If you are simply going to be offhand and dismissive of the efforts of someone who is genuinely trying to help you then you won't find help so easy to get in the future.
Before you can be witty you'll have to convince us you've got wits. Until then it's just cheek. Go back and read my post again. You must have missed something.
w
I interpreted "never happen" to mean that it's impossible to create a good sounding, easy to build amp, using EL34s and commonly available parts -- which is of course, nonsense, to say the least. If you meant something different, say so. If you want to further defend that statement, you would be wasting everyone's time (yours included).
It's ironic that you're lecturing me on off-hand, dismissive statements and questionable wit. Valid perhaps, but ironic nevertheless.
As for Eli, my reply to him was actually a copy/paste of a response he once gave me to an innocent, valid question. It was rude. But I'm confident Eli can take what he hands out. For what it's worth I have great respect for Eli.
..Todd
Last edited:
Todd,
If a paint by the numbers project is what you want, buy a ST70 "knockoff" kit and add a ganged, stereo, 50 KOhm log. (audio) taper pot. and coupling caps. between the volume control and the I/P stage grid to ground resistors.
FWIW, I favor Triode Electronics driver board for ST70s, as it retains Dyna's circuitry, while employing better active devices. A TE board populated with 2X EF86s and an ECC99 will make some hay.
Canada's own PEC makes excellent controls, whose cost is rational. PEC part # KKA5031S28 is what you want. Mount the control along with 1 or more sets of I/P RCA jacks, a source selector switch (if necessary), and short, captive, braided, unshielded interconnect cables in an Aluminum project box. Bring the signal source(s) to the box and connect the captive cables to the ST70.
Replace the wires that connect the amp's RCA jacks to the EF86 control grids with low WVDC/22 nF. Soviet surplus paper in oil (PIO) capacitors. A ST70 uses 470 KOhm I/P grid to ground resistors. That's OK with a pentode, as CMiller is not the problem it is with triodes.
If a paint by the numbers project is what you want, buy a ST70 "knockoff" kit and add a ganged, stereo, 50 KOhm log. (audio) taper pot. and coupling caps. between the volume control and the I/P stage grid to ground resistors.
FWIW, I favor Triode Electronics driver board for ST70s, as it retains Dyna's circuitry, while employing better active devices. A TE board populated with 2X EF86s and an ECC99 will make some hay.
Canada's own PEC makes excellent controls, whose cost is rational. PEC part # KKA5031S28 is what you want. Mount the control along with 1 or more sets of I/P RCA jacks, a source selector switch (if necessary), and short, captive, braided, unshielded interconnect cables in an Aluminum project box. Bring the signal source(s) to the box and connect the captive cables to the ST70.
Replace the wires that connect the amp's RCA jacks to the EF86 control grids with low WVDC/22 nF. Soviet surplus paper in oil (PIO) capacitors. A ST70 uses 470 KOhm I/P grid to ground resistors. That's OK with a pentode, as CMiller is not the problem it is with triodes.
You give up too easy. You want to build an EL34 amp. You don't want a kit, but you do want a proven design. I don't know whether you want to buy each component seperately and wire it up using point to point wiring, or if you are open to using a PC board for the driver and power supply components. You can decide. I offer a few suggestions:
You want a proven circuit that uses common components. How much more proven can you get than the Dynaco ST70 design. Now I'm not saying to buy a Dynaco kit, I am saying to copy the design, or preferably one of the newer designs that have solved some of the problems with the original Dynaco.
The Dynaco schematics can be found on the web. Use some of the newer Dynaco clone transformers, or choose your own. I would ditch the original Dynaco driver circuit and use one of the newer designs. Have a look at the DynaMutt design by the forum member "GEEK". The schematics of this and several other Dynaco compatible circuits are on his web site:
Classic Valve Design - Dynaco Clone and Original Design Boards
If you want to dig just a bit deeper:
You may notice that the DynaMutt schematic bears a strong similarity to a schematic that you have been drawing in another thread, so another obvious solution is to just stuff EL34's into that design, and get a 5K to 6.6K set of OPT's. To get the gain right use the 5751 (or 12AT7) for the first tube and the 6CG7 for the second tube, as in the DynaMutt noval schematic, or if you like big tubes copy the octal version. The whole collection of solid state parts including R14, R15, R16, and C5 can all be eliminated and replaced with the 10M45 circuit from your schematic.
I asked a similar question a while back trying to find a roughly 35 wpc amp and my thread was entitled "Is the ST-70 as good as it gets?" The answers I got there are pretty similar to what you're getting here. A few people recommended the El Cheapo Grande and some more recommended several variations of the ST-70. I guess when all is said and done, the ST-70 was popular for a reason.
My issue with the "design" of the ST-70 was pretty much 100% cosmetic; I just didn't like the way it looked. I ended up buying a Chinese EL34 based amp because I just couldn't come anywhere near the price they were offering if I built it myself. I wanted a 35 wpc tube amp, I didn't necessarily have to build it from the ground up myself.
I think if I do build a P-P project from the ground up it will probably be either the KT88 thread that I've seen you doing so much schematic work on or the 6L6 AB2 amp that Tubelab has referenced. I'd like to have something in the 60+ wpc range, but that will be down the road.
My issue with the "design" of the ST-70 was pretty much 100% cosmetic; I just didn't like the way it looked. I ended up buying a Chinese EL34 based amp because I just couldn't come anywhere near the price they were offering if I built it myself. I wanted a 35 wpc tube amp, I didn't necessarily have to build it from the ground up myself.
I think if I do build a P-P project from the ground up it will probably be either the KT88 thread that I've seen you doing so much schematic work on or the 6L6 AB2 amp that Tubelab has referenced. I'd like to have something in the 60+ wpc range, but that will be down the road.
Hi TubeLab.
You've got it.
In my first message in this thread I presented 2 schematics, one of which was a Dynaco modified Mullard, or what I assume is an updated design that, as you say, has solved some of the original ST-70 problems. The tubes are different (no more 7199), that's all I can tell. The 2nd schematic was a Leak design using 12AX7's in the first 2 sections.
My question from the beginning was simply: "Are these designs okay?" and "which one of the two makes more technical sense, or might sound better, etc? (and why)
What is your opinion of these two designs?
I am gravitating towards the Dynaco Mullard, mainly because of the commonly available iron.
I'm hoping BoyWonder will help me out with updating the power supply. He's getting very good at it. I'm not having much success with PSUD2. It's old and doesn't run well on my recent model computer, which makes it very difficult to learn how to use.
I'm sure the DynaMutt and Triode (and all the other) ST-70 kits are great but I'm not looking for a kit. I'd rather build it from scratch. P2P, PCB, or whatever makes sense for the project, neither are a problem. I've been building (SS) amps from scratch for many years, but tube amps are new to me.
And I will probably build the KT88 Mullard in the other thread, assuming it ever sees the light of day.
The big downside (for me) of tube amps is that they cost a fortune in parts, so I can't embark on nearly as many projects. Building them all to see what they sound like is simply not an option with tube amps. One must choose judiciously.
..Todd
Last edited:
The part in question serves 2 purposes. As the name suggests, "El Cheapo" is an economical project and the B- rail is "taller" (well over 100 V.) than really necessary, but the 6K27VF power trafo remains a good deal. So, 10 KOhms at 6 mA. drops 60 "excess" Volts. The 2nd purpose is thermal insulation. Carbon is a poor conductor of heat. Therefore, that Carbon film part thermally isolates the "sand" from the hot tube. A smaller resistance would be better, given internal I2R heating, but Volts have to be dropped.
BTW, a factor associated with the 10 KOhm part is lengthening the tail of the splitter/driver a tiny amount.
This is my quote from your original post:-
My interest here, is to find a 25-50W amplifier project: commonly available and preferably inexpensive parts; easy to build; stable; sounds great. Something anyone could build as a first tube project
25W is a SUBSTANTIAL valve hi-fi amplifier. You're gonna be lookin' at ultralinear operation. Mullard 5-20 gets 20W from 2 EL34s in ultralinear. Not everybody likes ultra tho' (not triode). You may be one of these people, so it may not 'sound great'.
Parts are available, but I don't think you could describe the situation as 'commonly'.
They will not be inexpensive at this power level by any stretch of the imagination being 10 to 1000 times the cost of those for an equivalent ss amp.
As builds go, it will not be easy. The HT will be high and consequently more dangerous. Or you can have parallelled OP tubes and more complex driver circuits. Building a hi-fi ss amp of this power is trivial. Building a hi-fi tube amp of this power is non-trivial.
It may be stable, it may sound great, but if you asked for a bit less power you'd have got a lot more suggestions. The history of commercial valve amplifiers is littered with failed designs in this power range, executed by engineers who worked with valves every day, and who had ready access to the most exotic iron. They went extinct, remember? I guess maybe you don't.
Most people would recommend something a little less ambitious as a first tube project.
w
My interest here, is to find a 25-50W amplifier project: commonly available and preferably inexpensive parts; easy to build; stable; sounds great. Something anyone could build as a first tube project
25W is a SUBSTANTIAL valve hi-fi amplifier. You're gonna be lookin' at ultralinear operation. Mullard 5-20 gets 20W from 2 EL34s in ultralinear. Not everybody likes ultra tho' (not triode). You may be one of these people, so it may not 'sound great'.
Parts are available, but I don't think you could describe the situation as 'commonly'.
They will not be inexpensive at this power level by any stretch of the imagination being 10 to 1000 times the cost of those for an equivalent ss amp.
As builds go, it will not be easy. The HT will be high and consequently more dangerous. Or you can have parallelled OP tubes and more complex driver circuits. Building a hi-fi ss amp of this power is trivial. Building a hi-fi tube amp of this power is non-trivial.
It may be stable, it may sound great, but if you asked for a bit less power you'd have got a lot more suggestions. The history of commercial valve amplifiers is littered with failed designs in this power range, executed by engineers who worked with valves every day, and who had ready access to the most exotic iron. They went extinct, remember? I guess maybe you don't.
Most people would recommend something a little less ambitious as a first tube project.
w
No Problem. Keep in mind I'm still a bit green at this but I don't mind the practice. I'm surprised to hear that PSUD is struggling on your shiny new computer, although I'm still running XP. It runs fine on both my laptop and desktop.
Ask lots of PSUD questions here in the forum, they'll get answered. If you want to play with PSUD, start by modeling an existing supply, and start with a current load instead of a resistive load. It's really quite easy once you get your feet wet. For simplicity, start with just a CLC section and one B+, and use the default transformer R if you don't have the unit in hand, it'll get close, and can be dialed in over time.
As a side note, since Eli hasn't asked the question yet.....how sensitive are your speakers?
(Not shiny new, just new-ish. I'm a computer systems eng., so I tend to have up-to-date computers by habit.) The PSUD2 help files are incompatible with >=Vista. I use W7. That's not insurmountable, just annoying. The help files aren't very detailed anyway. The diagrams are the size of postage stamps on today's high res monitors. That's a real pain in the butt. I have to connect two sets of reading glasses in parallel. I'm tempted to send an email to Duncan requesting/encouraging the release of the PSUD2 source code, so I (we/whomever) can update the user interface.
That's not really a concern. If my current selection turns out to be a problem [doubtful], I would simply design/build another set of speakers to suit the application. Additional fun; another few months of design therapy.
..Todd
Last edited:
A bit longer, though not as much as you suspect. I didn't start until the newfangled VIC-20's and TRS-80 arrived, and [eventually] the Apple II in our recording studio. Been at it ever since. Now I deal with [boring] massive medical failover clusters in data centers, which is why I am back at my audio roots seeking therapeutic fulfillment.
..Todd
Have you seen this design?
www.plitron.com/pdf/702Kobayashi1817.pdf
It claims to be a "Marantz 8B"
Since its a Millard topology and seems well documented, it might be of interest.
It Uses EL34 tubes and output transformers made in Canada, and is in the 35 Watt class.
Whats not to like.
www.plitron.com/pdf/702Kobayashi1817.pdf
It claims to be a "Marantz 8B"
Since its a Millard topology and seems well documented, it might be of interest.
It Uses EL34 tubes and output transformers made in Canada, and is in the 35 Watt class.
Whats not to like.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.