• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

50W monoblock "Engineers Amp"

Couple questions Pete if you don't mind.
1. how much are the metal chassis from landmark if you don't mind sharing? and if I tell them I want two of them for the 50w engineers amp, they will know what I'm talking about?

2. you mentioned your new audio analyzer, curious what you are using? I've been using an HP 8903B and your software for a long time :)

Thanks
Mark
 
I guess people have been buying them speculatively in large batches. Kind of a shame, if you ask me.

This happened with the last two amps that I designed. I modified a Tubelab SPP to use the 6HB6 which was on both $1 lists and over 10,000 were in stock. Before I could get boards back, poof, someone in Hong Kong bought them ALL! Stan got a single order for 7,000 tubes.

Ditto the 13GB5, only this time it was someone in the US, allegedly he has 40,000 tubes and has been working on his amp design for a couple of years. Stan has a standing order for every 6GB5 or 13GB5 he gets.

This practice has become quite common, someone here mentions a good use for a cheap tube, and Stan gets an order for "all of them." ......I know how to get 200 watts from a pair of 3A3's....go get em!

It looks like going forward we must design amps for the same old crummy tubes that the guitar amp guys use, because they are in current production, or create universal designs that can use a dozen different tubes.....or just quit designing them. Don't give the guys with all the tubes anything to copy, let them eat them.

My 100 WPC 13GB5 amp has been destroyed, the Eagle and LT Spice files will never be posted. I have not decided whether or not it will be reborn with different tubes. I will probably build myself a 500 WPC version since I have all the parts, but may not post the design. Why give the guys with all the tubes a head start. I traded all the 13GB5's that I had for tubes to design something totally different. It's not an amp, and uses tubes so common that they can't all be bought!
 
Member
Joined 2014
Paid Member
On the bright side, at least you DID have access to these tubes cheap. Over here the choices have always been more limited unless we imported them. Better to have loved and lost. Sucky for those of you making a bit of pocket money helping out the DIY community tho.
 
Oh, no. Not again.
Not much left for useful TV Sweeps now.

Let's not ever mention the super secret 6BN6 or 6BV11 super tubes. The net worth of countries are closely related to stocks of those. And I would have added the other secret 6LE8 super tube too, but someone seems to have already bought most of those! At least 6AL5 and 6H6 are still around.

Maybe new schematics should show junk/fake tube types on them. Then when someone buys the PC board, they get a late revision notice as to what tube actually plugs in. (6JH5 as typo for 6HJ5) Or else develop designs that use banks of 20 paralleled garbage tubes for real.
 
Last edited:
Or else develop designs that use banks of 20 paralleled garbage tubes for real.

I tried that. about 15 years ago I got about 100,000 mostly used and abused tubes for the cost of a few days labor. (Mistake, it cost more time and money to store and sort them than all the tubes that I actually used were worth).

In that lot were large numbers of some useful tubes (and about 2,000 6AL5's). I had about 1000 6AQ5's. After sorting out the tubes with obvious problems (mostly severely corroded pins) I had a few hundred usable tubes. I made several of the usual amps with two and four tubes per channel. Then one day I got the idea to wire a big bunch together. I don't remember how many, but there were maybe 8 or 10 pairs. For obvious reasons I used cathode bias with a separate resistor for each tube. I could never keep the thing from getting unbalanced and red plating one or two tubes. It would idle for days, or play music at reasonable levels without issue, but crank it or worse, plug my guitar into it and it would go unstable. Multiple tube swap outs never made it work right. I chalked it up to possible gassy tubes and tossed it.

When the 13GB5 first hit the dollar menu, I got a box full (just traded them). I did the usual, two and four tubes per channel amps with good success. Then, I got the idea that this might be the way to build the "big one" that I have been dreaming about for years. I already have the iron, so could I make 500 reliable watts per channel with $12 worth of tubes?........Nope.

This time I used fixed bias with a pot for each tube, and a dozen tubes. B+ was 625 volts, Idle current about 15 mA, load impedance 1250 ohms. Same thing happened. I could leave the amp running a sine wave at 100 watts for hours and all was stable, no bias creep. Hammer it for a few minutes and the currents would be different by up to 10 mA, with red plate showing on some tubes, slowly returning to "normal" as I let it settle at 100 watts.

I played musical tubes, and "red" follows the tube. I believe that in order to make a multi tube amp play nice near it's maximum power level would require Gm matching in the range of currents where max power is reached, or some microcontroller based active bias system.

I now have a curve tracer, so I will explore this a bit further before committing to the final design of the "big one." Since "matched tubes" will not stay matched forever, I am also refining the concept that I used with the 100 WPC 13GB5 amp.
 
Sadly seems every time the DIY crowd find a good tube someone then buys them all up.

ya my design has evolved quite a bit. started the layout a year ago and keep updating for supply situtation.

so now i've ripped up traces again to support stuffing board with all octal , all noval, mixed octal and noval, solid state hybrid , 100% solid state, jfet, n-ch front end, p-ch front end , n-mos , germanium and so on.

lots of time spent creating hybrid passive footprints too.
i have reels of over-run smt parts so now everything can be smt or through hole.


socket hardware is ofically critical. need better sockets.
I've sampled everything out there. best looks like cmc bakelite gold plated machined brass.

but now im looking at canon/amhenol inserts.

found too a much superior candiate for tube sockets but if i mention, they will disappear too. they could be used for semis too, excellent insert removal life.

-bruce
 
Couple questions Pete if you don't mind.
1. how much are the metal chassis from landmark if you don't mind sharing? and if I tell them I want two of them for the 50w engineers amp, they will know what I'm talking about?

2. you mentioned your new audio analyzer, curious what you are using? I've been using an HP 8903B and your software for a long time :)

Thanks
Mark

Mark -

Not sure what Landfall is charging for the chassis. Send them an email and they'll give you a price, I know they have quoted it already and know what it is.

I bought a Prism Sound dScope III. It was that or an Audio Precision APx515... the dScope can do a few things that the AP cannot, and is considerably cheaper (at least in the US, due to the devaluation of the UKP). The fact that AP sales never even returned my messages didn't help either.

Pete
 
This happened with the last two amps that I designed. I modified a Tubelab SPP to use the 6HB6 which was on both $1 lists and over 10,000 were in stock. Before I could get boards back, poof, someone in Hong Kong bought them ALL! Stan got a single order for 7,000 tubes.

It looks like going forward we must design amps for the same old crummy tubes that the guitar amp guys use, because they are in current production, or create universal designs that can use a dozen different tubes.....or just quit designing them. Don't give the guys with all the tubes anything to copy, let them eat them.

Same thing happened with a headphone amp I published that used the 19J6. Thousands disappeared.

I guess before I publish any more designs that use oddball tubes, I will buy a bunch of them myself. Then pass them on at cost (more or less) in limited quantities, 2 or 4 at a time.

Or, as you say, use the plain ol' EL34/KT88/6L6/6550 stuff.

Pete
 
Well, if all the cheap TV Sweep tubes disappear, I guess I'll be using a small pentode tube + Mosfet Darlington or Sziklai combo instead. (triode emulated via V Fdbk to screen grid)

----

"I now have a curve tracer, so I will explore this a bit further before committing to the final design of the "big one."

The uTracer3+? I see that one does up to 400V B+ now.
The uTracer, a miniature Tube Tester / Curve Tracer.
Would be nice if they would extend the grid V sweep range further, for grid 2 testing.

I extended the Tek 576 curve tracer here recently to do up to 168V grid sweeps, so grid 2 is covered now. (up to 14 x 12V steps) And a simple mod to get 15 traces instead of the standard 10 traces. (Tek went to some trouble to limit it to just 10 traces! But one can just pull one IC pin out, and add a resistor to fix that. )

..
 
No, but plenty of them are spread all over the Eastern European countries. Speaking of GU-50, it is indestructible tube, also still in production in China (FU-50).

they must be still plenty judging by their cheap price, but the 6H23 has risen in price since you first introduced them to me.....:eek:
i see the 6p3c rebranded as a trigon 6L6gc and other brands....
 
Any chance that this mono-block version will see a second version which supports the same tubes as the original DCPP? That would open up a lot of other options.

Actually in retrospect I think it would be a good idea to remove the output tube sockets, and just put big round holes in the PCB where the output tubes go. That way you could wire in whatever tube and socket that you want, octal, magnoval, compactron...

What do folks think about that idea? Good, or dumb?

Pete