Just a bit too much power for me, though if I'd nabbed a pair of those Plitron output transformers when they were still available, I might be singing a slightly different tune. I think my experiments will be confined to what I can accomplish with a pair of Edcor 100W 3.5k output transformers. That's still enough power so that the neighbors will know precisely what I'm doing, whether they like it or not... It's nice to know that my 35LR6s won't even be breaking a sweat at that power level.
Nice work but 18 watts ? why bother, proof of concept perhaps, I guess.
Still nice looking amp.
I'd like to build something a bit more substantial, something in the 200 watt per channel would be a good starting point
Still nice looking amp.
I'd like to build something a bit more substantial, something in the 200 watt per channel would be a good starting point
I think I developed this fascination with electrical power when the paper clip met the wall outlet somewhere around age 5. My personal maximum occurred a long time ago. I made a 1200 watt solid state amp back in high school. It was really simple and used 24 X 2N3773's. It didn't even have a power transformer, so it can't be discussed here. Do I need or even have a use for an amp that big? No, but it is still cool. That one went to a rock band.
I built tube and solid state amps ever since except for a ten year period when I just bought a big Carver / Phase Linear system. After I figured out how bad it sounded I started building tube amps again.
I think 50 to 100 WPC is more than enough for my needs, but when a pair of BAT's come along for a good price I couldn't say no. Eventually I will find a proper way to use them to their full capability. For now the red board is a learning tool. The red board makes a nice 75 to 100 WPC amp with 6HJ5's, so if it survives my experiments that will be its final resting place.
I built tube and solid state amps ever since except for a ten year period when I just bought a big Carver / Phase Linear system. After I figured out how bad it sounded I started building tube amps again.
I think 50 to 100 WPC is more than enough for my needs, but when a pair of BAT's come along for a good price I couldn't say no. Eventually I will find a proper way to use them to their full capability. For now the red board is a learning tool. The red board makes a nice 75 to 100 WPC amp with 6HJ5's, so if it survives my experiments that will be its final resting place.
Well, Maybe 200 wpc is asking for too much,
How about about 100 wpc then.
My McIntosh currently can deliver 30 wpc and there are times when more power would be nice,
As a quick fix, I've setup bi amping where the low freq goes to a sub woofer amp (250 wpc @ 4ohm) and separate speaker, this sort of fixed the problem a little but it's still lacking....
btw: what is a 35LR6 ?
How about about 100 wpc then.
My McIntosh currently can deliver 30 wpc and there are times when more power would be nice,
As a quick fix, I've setup bi amping where the low freq goes to a sub woofer amp (250 wpc @ 4ohm) and separate speaker, this sort of fixed the problem a little but it's still lacking....
btw: what is a 35LR6 ?
100WPC can be done quite nicely with off the shelf OPTs from several winders using a pair of 6550 or KT88 and a single stage pentode driver with local feedback. Sounds like a sweet spot to me...
A 35LR6 is a 6LR6 with a 35 volt heater. Basically a BAST; big as they come ;-) Well actually 6LF6 is the biggest of all BASTs
Michael
....
Any chance you have a amp design or a particular mfg and model in mind that uses 4 of these in a push-pull conf.?
and what kind of power would you see?
Any thoughts on using something like a 4cx250b transmitter tubes for a audio amp?
I found a set of 3 for about 100 bucks.. Could one use their existing amp to drive something like this?
The KT 120 sounds like a tube to design around as well for a few more dollars....
I don't have any experience designing tube electronics, any ideas where to get the necessary learning tools off the internet ?
Last edited:
I don't have any circuit at present for what I want to do, which would be 100-120W output using 550-600V B+. There are three possibilities - using the 35LR6 in straight pentode mode like here, conventional screen drive, or G1 +G2 combination drive. I've yet to make up my mind as to which approach, and I probably won't get really serious until I actually go and spend the money on the Edcor output transformers.
Anyway, whichever route I choose will not be too easy to duplicate, as it will also involve a custom SMPS of my own design.
Transmitter tubes like the one you mention usually need a lot of B+ (danger, Will Robinson!) to make them come alive. Sweep tubes are a better bang for the buck.
George - It might be interesting to pop some of those cheap AES 6CB5s into your test circuit (though it may involve haywiring in some octal sockets) to see what can be coaxed from them.
Anyway, whichever route I choose will not be too easy to duplicate, as it will also involve a custom SMPS of my own design.
Transmitter tubes like the one you mention usually need a lot of B+ (danger, Will Robinson!) to make them come alive. Sweep tubes are a better bang for the buck.
George - It might be interesting to pop some of those cheap AES 6CB5s into your test circuit (though it may involve haywiring in some octal sockets) to see what can be coaxed from them.
Last edited:
Wrenchone:
You might want to re-examine your load line. I just ordered a pair of 100W Edcors for 550-600V B+ and I came up with 5K Zpri. With 3K5 you should only need about 480V B+ to get 100W Po. If you plan to go with only local feedback, a higher Zpri will improve your damping factor. PS - I was quoted 3-4 weeks delivery on 6/24
toddbailey:
The output power question has been discussed in a couple of other threads. OPTs will be the biggest issue going over 100W Po. I have a 300WPC project in the works using 4CX250s and the B+ will need to be 1500V.
http://www.diyaudio.com/forums/tubes-valves/162950-monstrous-1kw-amp-8.html#post2183786 The OPTs are 26 lb monsters that are 8K Zpri. The power transformer is a 55 lb. 2KVA machine tool transformer feeding a voltage doubler. It isn't meant to be a practical amp for home use or anything like that... But, they are just beam tubes and need about the same drive as a 6L6.
Cheers,
Michael
Attachments
Last edited:
One way to get past the 100 Watt easy OT limit would be to go with Crowhurst's Twin Coupled idea. A couple of EDCOR CXPP100-MS-2.5K say (or CXPP100-MS-1.7K), would give 5K Ohm Zpri (or 3.4 K Ohm) for some BASTs or KT120s. Some small output tubes like an EL84 or 6GF5 for drivers (Mac-like bootstrapped plate resistors) to get the very large g1 drives needed for the outputs. CFB is just another version of Schade. Might want to use B- and B+ so as to reduce the primary to secondary voltage stresses at 200 Watt.
OH yeah, Hammond has those 280 Watt 1650W 1.9 K Ohm OTs.
OH yeah, Hammond has those 280 Watt 1650W 1.9 K Ohm OTs.
Last edited:
Really, you mean that little wimp in between two real BAST's that are about to kick its...
The photo shows 4 tubes. The 35LR6 is on the left. Next is a 6LW6, then a 6LF6 and another 6LW6. You know which tubes are my favorites. In reality the central core of all of these tubes are the same. I have seen versions of all 4 with different configurations of heat radiating fins welded to the plates, and I have seen versions of all 4 without the fins. The 35LR6 and the 6LF6 shown here do not have the fins. I have seen 6LW6's in small bottles (same size as a 6LF6) and in fat bottles like these. The versions without heat radiating fins start to glow at about 50 to 60 watts of dissipation. The fat bottled 6LW6 with the big fins start to glow at 90 to 100 watts. The octal base is a plus too.
I have a long list of tubes that I want to "test". 6LW6, 6CD6, 6CB5, 6DQ5, 6146.....6HV5...307A....45. I skipped a few, but note the wide range of tubes. The red board has been a great learning tool, but I think I have explored about everything I can within its limitations. Maybe I will just do as it was intended and make an amp out of it. I liked the 6HJ5's for about 100WPC and I do have a big pair of 6000 ohm OPT's that have been around here for several years. Add one big Antek, stir well.....
What I need is a new board. What I don't know yet is exactly what to put on it. The red board couldn't drive grid current, so I need some mosfet followers. I need to try octal, 9 pin and 12 pin output tubes, so I need the ability to swap tube sockets. I want to experiment with screen drive and G1 = G2 drive too. I know its going to blow up a few times so easilly changeable parts is a requirement. Clip lead wiring doesn't work well with high Gm devices. I'll figure something out....
Pete designed the board, and specified the parts such that it could be assembled by builders of varying skill levels at a reasonable cost, and achieve a high probability of first time success. Unfortunately it is a FACT that building an amp of a higher power level results in an exponentially higher cost and a lower probability of first time success. The likelyhood of expensive mistakes increases dramatically too. This is true of most other hobbies too.
If you have never built a hot rod engine before, would you spend $2000 to make a 350 HP Small Block Chevy engine, or spend $10000 (parts cost) to make a 700 HP SBC engine. The lower powered engine will be a lot more tolerant of minor build errors like bearing and ring clearances, so it will work the first time. It also runs at a lower RPM range (voltage) for longevity. The big nasty race motor is best built by an experienced engine builder.
Just as with the automobile engine, there are a few experienced "hot rodders" here that can take Petes 18 watt board, swap a few parts, upgrade a few things, and dial up a bunch more power. I have been building tube amps for over 40 years, and I still blow up stuff. The red board has gone BANG twice in search of extreme power levels. I have the experience to figure out why, and fix the problem before continuing on. An inexperienced builder would have a bunch of toasted parts and nowhere to go.
There have been threads dealing with high powered tube amps on this forum since I joined. A lot of people have discussed building the "big one" but there are very few documented successful builds. Having the schematic of a working amp is a good start, but it won't make you an amp without the skill set necessary to implement it.
Attachments
I guess I take my building skills for granted. I've been a eletech for a few years back in the 70's or 80's and then as a hobby continuously. I like the sound of my McIntosh amp but @ 30wpc volume levels are limited. If I remember the math a 3db volume increase means you need 4x the power, Is this correct? Assuming it is then the next power level would be 120 and so on.
So unless I can find a healthy 100 plus watt amp on ebay, at a decent price building one over time seems like a worthy project.
I think anything larger than 200 wpc would be quite spendy and finding the rare parts would be the limiting factor.
but at this stage I have a fair bit of knowledge to take in.
So unless I can find a healthy 100 plus watt amp on ebay, at a decent price building one over time seems like a worthy project.
I think anything larger than 200 wpc would be quite spendy and finding the rare parts would be the limiting factor.
but at this stage I have a fair bit of knowledge to take in.
That would be me.😀
No, it's actually double the power for a 3dB increase.
Secrets of Amplifier and Speaker Power Requirements Revealed - eCoustics.com
Jeff
Mr Koster -
I've already been over my load line. I'm going for the 3.5k transformer so I can get a clean 100W and a little over, going with the assumption that I consider the output tubes to be bottomed out at ~50V Vp. This seems to be a pretty much universal number for specifying sweep tubes, with VG1 =0 and around 100-120V on G2. Under that assumption, 5k gets me a little under 100W, without accounting for any voltage drop in the transformer windings. I'm playing things conservative to make sure I get what I want. If this results in more power in the actual product, that's just a little more peak reserve I can keep tucked in my back pocket.
At any rate, I'm not in a big hurry on this project, as I have several (more like 6 or 7) stacked up ahead of it. If I go with screen drive and relatively low bias current, I'll have to take a really hard look at the dynamic characteristics of my SMPS to make sure that my cross regulation doesn't fall apart under big load transients. I want this to be a relatively light and compact amp (except for those output transformers), so an SMPS is the order of the day, maybe one for each channel. Splitting up the supplies will also make it possible for me to still use ZVS flybacks like in my other amps without things getting too bloated.
At any rate, this really doesn't have a whole lot to do with Pete's design here, so I will take any more speculations over to my original "Killer" thread when I have something more concrete, like a firm design concept and preliminary schematic.
I've already been over my load line. I'm going for the 3.5k transformer so I can get a clean 100W and a little over, going with the assumption that I consider the output tubes to be bottomed out at ~50V Vp. This seems to be a pretty much universal number for specifying sweep tubes, with VG1 =0 and around 100-120V on G2. Under that assumption, 5k gets me a little under 100W, without accounting for any voltage drop in the transformer windings. I'm playing things conservative to make sure I get what I want. If this results in more power in the actual product, that's just a little more peak reserve I can keep tucked in my back pocket.
At any rate, I'm not in a big hurry on this project, as I have several (more like 6 or 7) stacked up ahead of it. If I go with screen drive and relatively low bias current, I'll have to take a really hard look at the dynamic characteristics of my SMPS to make sure that my cross regulation doesn't fall apart under big load transients. I want this to be a relatively light and compact amp (except for those output transformers), so an SMPS is the order of the day, maybe one for each channel. Splitting up the supplies will also make it possible for me to still use ZVS flybacks like in my other amps without things getting too bloated.
At any rate, this really doesn't have a whole lot to do with Pete's design here, so I will take any more speculations over to my original "Killer" thread when I have something more concrete, like a firm design concept and preliminary schematic.
Sorry, I had just come to a different conclusion but I think my goals are a little different and I'm not looking for a lot of headroom above 100W. I also want to use a machine tool transformer for the B+ without having to lose a lot of voltage...
I will look into your other thread.
Hey I just read the other thread LOL. You were thinking 5k at that time...
Cheers!
Last edited:
Double the power is 3db, which is a perceptable sound level increase. Double the apparent volume level is 6 to 10 db depending on who you believe. This requires 4 to 10 times more amplifier power. A 10 db increase in speaker efficiency is the same as boosting the amplifier power by a factor of 10. A 20 db increase in speaker efficiency is equal to 100 times more amplifier power!
Many speaker systems are not linear above a certain power level. They will reach a point where the cone is already moving as far as it can, so more power just makes more distortion.
In my lab I use a pair of Yamaha studio monitors. They are rated for 70 watts and are only 88db efficient. A 2 WPC amp provides for enough volume for quiet listening at full volume. A 10 WPC amp is reasonably loud and provokes my wife to yell "turn it down" from another room. A 30 WPC amp is quite loud. A bigger amp doesn't get much louder. Last night I had the 250 WPC red board connected to them. It is not much louder than the 30 WPC amp, but the sound is much cleaner and "effortless" since it is not possible to drive that amp into clipping without setting the speakers on fire!
A friend has Lowther based horns that are 106 db efficient. The little 2 WPC amp is very loud through them even at half volume.
If you are not in a hurry, keep a lookout for the expensive parts. A multi gigawatt tube amp has always been well out of my price range. Three years ago there was a discussion about high powered tube amps. Someone mentioned the 400 watt OPT's being sold as surplus for $134 each. I grabbed a pair. So did several other members. The thread went on for over a year and nobody built their big amp. The OPT is the most expensive and hardest to get piece of the puzzle. It determines what amp you build. Find that piece first.
http://www.diyaudio.com/forums/tube...807s-push-pull.html?highlight=plitron+surplus
I have the transformers and plenty of suitable tubes. I suppose I could put something together, but I really don't need 500 watts right now, so I choose to use these parts for experiments to learn more about the best way to make a high powered tube amp.
Mike -
I was thinking 5k at the time but had already realized that given the 50V proviso I've mentioned, I was only going to get 98W or so, not counting various and sundry voltage drops. The next step down from 5k in Edcor-land is 3.5k. No LOL, just compromise, as in just about any engineering enterprise. I'd much rather over-design a bit than under-design.
I was thinking 5k at the time but had already realized that given the 50V proviso I've mentioned, I was only going to get 98W or so, not counting various and sundry voltage drops. The next step down from 5k in Edcor-land is 3.5k. No LOL, just compromise, as in just about any engineering enterprise. I'd much rather over-design a bit than under-design.
"The next step down from 5k in Edcor-land is 3.5k."
Huh...?
CXPP100-MS-3.3K
CXPP100-MS-4.2K
CXPP100-MS-5K
Huh...?
CXPP100-MS-3.3K
CXPP100-MS-4.2K
CXPP100-MS-5K
Sorry, I didn't notice the 4k choice when I was looking at the list at the Edcor site, and it looks like I mis-remembered the 3.5k value. That gives me more choices when I actually buy the transformers - more power, or less distortion. I've pretty much nixed the 5ks for right now, as I don't want to go much above 550V in my SMPS design because of the way I'm partitioning my B+ windings. 4k is kinda convenient. At any rate, I'm working on a JFET crossover, a new JFET RIAA amp, a 6JB5/6JC5 P-P amp with SMPS, a JFET/Mosfet amp with diamond output stage, and a 6LR8 /JFET hybrid P-P (another SMPS, too), so it'll be a while before I get to the 100W design. Maybe I'll change my mind again by the time I get to start the project. I figure things get set in stone once I order the transformers, not until then.