Need Help Understanding/Optimising a Schematic

[ak_47_boy]

I'm a bit of a noob to this stuff.
The schematic of amp i'm choosing to build is not noob friendly. It has no wattages on resistors, no voltage on capacitors, unnamed tubes, hell i don't even know where inputs and outputs are. If somone would be so kind as to do a little ms paint edit on it i will try to re cad it. I need a suggestion on scematic cad too. I would like to cad in everything so it is complete and working with no fill in the dots.
This is the amp:

http://www.bonavolta.ch/hobby/en/audio/kt88_7.htm

thanks

 

[316a]

I'm a bit of a noob to this stuff.

In that case build something much simpler . A 400watt valve amp is not a good first project . The cost of the output valves , mains transformer and output transformer alone will cost an arm and a leg plus there is the safety and debugging aspect to consider . Buy a book such as Morgan Jones 'Valve Amplifiers' or buy a kit , it will save a lot of hassle

cheers

316a

 

[nickds1]

I'm a bit of a noob to this stuff.
... hell i don't even know where inputs and outputs are.

No disrespect intended but you need to get a grasp on reality. I have a degree in EE, and 25 years of experience, in a mix of tecnologies, and even though I can well understand the schematic and have built a few small valve designs, I wouldn't consider building this one - I'm working my way up through other, less risky, designs.

Apart from the physical risk to yourself and others, you will spend many 100s of dollars and end up with something that, if it doesn't catch fire first, will probably not work.

As another poster has said, get a copy of Mogan Jones' book, and try to learn about that you are trying to do. This isn't Lego, it's not kids stuff, it's dangerous and could easily kill you (or more worringly, someone else)...

 

[lndm]

This seems to be a Williamson amplifier with cathode followers and multiple output devices.

FWIW, I think that the common implementation of the Williamson is quite reasonable, ie more straightforward to build and sounds good. http://www.xs4all.nl/~ideas/amps/chapt14.html

 

[ak_47_boy]

Don't worrie. This is why i am asking about this stuff, i could have just built it but there would have been a 50% chance that i messed somthing up. What are those unlabled tubes?

edit: btw i'm reading that link right now

 

[ak_47_boy]

The scematic is mono but the powersupply can run stereo right?

 

[Eli Duttman]

For a Stereo amp, you need 2X the B+ current that a mono amp needs.

Build monoblocks or you WILL get a hernia. The "iron" for that circuit is VERY heavy.

The unlabeled O/P tubes are KT88 beam tetrodes.

I, as do other posters, think you are in WAY over your head with this project. There is a GOOD chance you will get yourself killed. Get your feet wet on a simpler, safer, project (possibly a preamp).

BTW, what sort of "Watt Hog" speakers do you have in mind for use with such a powerful circuit?

 

[Wavebourn]

This seems to be a Williamson amplifier with cathode followers and multiple output devices.

FWIW, I think that the common implementation of the Williamson is quite reasonable, ie more straightforward to build and sounds good. http://www.xs4all.nl/~ideas/amps/chapt14.html

Looks similar (bubbles with dotted lines inside), but absolutely different devices. Were sir Williamson an engineer actually?

 

[sorenj07]

for a really really simple circuit to practise on, and that should barely cost you anything (assuming that you had the money for that 400w beast), try this following schem from the acrosound catalog:

An externally hosted image should be here but it was not working when we last tested it.

very simple, i'd use the hammond 1650E as an output and probably 6SL7's instead of 12AX7's but it's the same idea. power supply could be a 274bx rectified by a 5ar4 feeding a 550uF 385V cap through a 193J. i'm not exactly sure what the load is but it's bound to be under 200ma.

you'd be doing both yourself and us a favor by starting on a small project, yourself so you don't hurt yourself, and us (namely me) so i don't worry about you hurting yourself . don't make me regret giving you that schematic

 

[Tubelab_com]

I took about 1 minute to look at the schematic, and figured out that if built as shown it would melt down rather quickly! The 2uF caps in the bias supply are in backwards. They will not live long, followed by death of the 6X4, and then all of the KT-88's due to lack of bias.

If this is fixed, and there are no other mistakes in the schematic, the design would require a matched set of 10 output tubes. They all run off the same bias supply with no individual adjustment provisions. This would reasonably well matched tubes. The tubes that are made today are nowhere near matched enough to use without a selected set. 10 matched KT88's alone could cost $300 USD at least, the OPT another $250, and the power transformer another $150. This will easilly be a $1000 per channel amp.

As stated by everybody else, this is not a first time build. It would take some experimenting to get it to work. The chances of success are not great, and the cost of the build is high. Spending a bunch of time and money on a project that never works will kill enthusiasm for a new hobby rather quickly.

I have seen many people quit this hobby for this reason. I have bought a few of their projects cheap, just for the transformers. I started building tube amps in 1965, and I still make amps that I wind up tearing down.

You will (hopefully) learn by building amps, and get ideas about how to do it better next time as you go along. Find a suitable amp for your first build. It should be a design that has been built by other people. Build it exactly as shown. This will allow for a high probability of success. Then modify the **** out of it. Learn what makes it sound better, and what makes it sound bad. Blow it up a few times (you will figure this out on your own), and learn how to fix it. Then you can tackle a more complicated project. If you built that monster amp, and it didn't work, how are you going to know if you made a mistake, or it was a bad schematic?

 

[Wavebourn]

for a really really simple circuit to practise on, and that should barely cost you anything (assuming that you had the money for that 400w beast), try this following schem from the acrosound catalog:

An externally hosted image should be here but it was not working when we last tested it.

very simple, i'd use the hammond 1650E as an output and probably 6SL7's instead of 12AX7's but it's the same idea. power supply could be a 274bx rectified by a 5ar4 feeding a 550uF 385V cap through a 193J. i'm not exactly sure what the load is but it's bound to be under 200ma.

you'd be doing both yourself and us a favor by starting on a small project, yourself so you don't hurt yourself, and us (namely me) so i don't worry about you hurting yourself . don't make me regret giving you that schematic

Good classical class A design, circa mid 1930'th.

 

[sorenj07]

yup. i figured i'd trust good old herbert over at acrosound to provide economical and solid amps. i'm looking to build this thing for a friend actually, or kind of help him build it on his own anyway i'll quit threadjacking, either he or i will make a new thread when the time comes. carry on, ak47boy.

 

[lndm]

Looks similar (bubbles with dotted lines inside), but absolutely different devices. Were sir Williamson an engineer actually?

I believe you're right. I used the term 'common implementation' to suggest it has been butchered over the years. Of course, it was no pot shot, being so popular and all. Even using commonly available OPTs I think it makes for a quite reasonable amplifier.

 

[ak_47_boy]

I would not be worried about me hurting myself. I'm very safe when working around high voltages. This definaly won't be the first time i have worked with HV. My last hobby was diy gas lasers... I have been working with this kind of stuff since i was 10. I do like to ask alot of questions because a stupit question is better than a stupit mistake. I totaly understand your worries, from personal experience i can definaly say that most people that tryed to build this project would hurt themselfs. I have never even burned myself soldering/brazing/welding this year and i do most every week. I'm safe, i like to go slow and if have doubts i will ask somone before proceding.
I am a long way from building this amp, i will have completed a headphone and linestage before i start working on this. So im at least a year away from building it.
I do however want to start watching for key parts i need, with some luck i will be able to find some high quality iron that won't break the bank. I recon i may have a hard time finding some of this stuff.

 

[sorenj07]

for the 400w opt's i'd suggest finding someone to do them as a custom job. iron like that doesn't really show up on ebay, from what i've seen, unless you plan on re-winding some industrial control transformers or something. as for chassis design, i can only forsee 4 separate chassis, one for each channel and ps.

 

[Miles Prower]

An externally hosted image should be here but it was not working when we last tested it.

This is just about the most ridiculous thing I have ever seen. Who needs 400W? Except for a high powered, plate modulated AM transmitter (these days, most hams use SSB anyway), or maybe a large PA system, I don't see any point outside of braggin' rights. The 25W/Ch VT amp I have here is plenty loud enough as it is to get the neighbours complaining if I crank it.

Even if I did need 400W, I'd use something like an 8877 -- big, high gain RF triode that would just be loping along at 400W. That would need forced air cooling and 3000Vdc power supplies. To get 10VTs tracking, you'd probably need a PIC-based bias circuit.

I seriously doubt that anyone ever built that thing. Surely they would have noticed those backwards bias supply filter capacitors if they had.

FUGGEDABOUDIT

 

[ray_moth]

For someone who has been used to SS, a 400W amp might not seem a big deal. However, when you're trying to get that much from tubes, it is a big deal.

Anything much over 100W of audio is a challenge with tubes; many experienced people would consider a 400w tube audio amp infeasible, expensive and unsafe. OP transformers that can handle this much power are rare and to get one that performs well is going to be a challenge, to say the least.

And look at the voltages involved, remembering that the B+ supply also has to provide a lot of current to produce all that power with an efficiency of about 50% from UL mode KT88s. That makes the B+ a potential source of very scary problems.

If you want an amp that sounds very loud, you could build one at much less cost that produces less power and put the money you save towards buying efficient speakers.

 

[Tubelab_com]

I don't see any point outside of braggin' rights.

I see that as reason enough, but I have never been practical anyway. There was a thread about a 1KW guitar amp (4 X 813) a while back. I thought about it for a minute, and that little bit of common sense that remains in the back of my head talked myself out of it.

Even if I did need 400W, I'd use something like an 8877

I was thinking a push pull pair of 813's, or 2CX250's. (I have both). The lack of a suitable OPT keeps me from breadboarding something (and that little sanity voice in my head).

If you want an amp that sounds very loud, you could build one at much less cost that produces less power and put the money you save towards buying efficient speakers.

That is the most sensible thought here. The difference between 100 watts and 400 watts is 6 db. The difference between 100 watts and 400 watts (per channel) is probably over $1000. The probability of success for 100 watts per channel has got to at least double that of a 400 WPC amp. That $1000 could be spent on 6 db better speakers and the result would be better sound.

If you just need enough sound to blow the house down, consider a bi - amped (or tri - amped) system. Use a big honkin SS amp with a kilowatt per channel if you want for the bass (easily made or purchased used) and a tube amp for the mid and treble. The tube amp will be operating for the frequencies above say 400 Hz. It would only need to be 50 WPC or so. You could be building a 50 WPC tube amp today. There are plenty of good circuits around.

I could build a 100 WPC tube amplifier today. I built one back in the 1970's and it never worked right. It kept eating 6550's. Even then 6550's weren't cheap. The whole thing went into the trash. It was an expensive failure that made me forget about tubes for a few years. I now realize that the problem was runaway due to grid current, but I couldn't figure it out then.

A 400 WPC amplifier would take some careful design, and experimentation to achieve success. There are only a few people who could pull that off today. Any design with 2 banks of 5 tubes is GOING TO BE UNSTABLE. As Miles put it, some type of active bias stabilization will be required. Tubes made today have higher and more variable grid current due to the lower quality vacuum. The schematic shown may work after some experimentation (likely to burn up some tubes and other parts). The resulting amp will be prone to bias drift as the tubes age.

If you are intent on building a 400 WPC tube amp in a year or so. Spend that time wisely. Study the designs of the big tube guitar amps of the past, read up on their spactacular failure modes. The early Ampeg SVT's used to catch on FIRE! Look at the current designs (Sunn Coloseum, etc). Hang out at a store where amps like this are sold (Sam Ash, The Guitar Center, etc.) Talk to users, and the stores service tech, to find out which ones are reliable. If high power amps made by companies with big design budgets caught on fire, what do your chances of success look like?

Radio transmitters of the past used big audio amplifiers as part of the modulator circuit. Explore this route. It may be possible to find a ham radio operator who remembers how to build an AM transmitter (the Globe King comes to mind).

 

[SY]

The Globe King! 811s in the modulator, V70Ds in the final. I loved mine.

 

[Tubelab_com]

The Globe King! 811s in the modulator, V70Ds in the final. I loved mine.

I had a model 500. 500 watts of DC input to the final RF stage. The output tube was a 4-250. Like many I put in a 4-400. I could get about 300 watts out of it on 75 meters, and about 250 watts on 10 meters.

The modulator came with 2 5514 tubes driven by a 6L6. The tubes were dead, so I replaced them with 811A's (also a common mod) this required a 6 volt filament transformer. The modulator put out about 250 audio watts.

Some one who had watched the Smokey and the Bandit movie too many times paid me a lot of money for mine, so I sold it in the mid 70's.