OTL Amp

[J M Fahey]

A friends amp is ALWAYS the one that turns out to be tricky and time consuming.

And that's *nothing* compare to a girl friend's amp or any other device she owns.
Ho do you make her understand the thing is nuked, replacement costs U$100 and repairing costs U$10.000, not including the trip to Mars to get the parts?
Or if a mother-in-law's, access to a time machine, since vital irreplaceable parts stopped being made and stocked around 1963?

[Chuck H]

That's where someone with fabrication skills comes in... J M. A standard response to several such circumstances is "Nothin' but time."

[km6xz]

Juan, that is sort of the point, there is nothing the tube could impact with it's small current drain(measured) compared to the rest of the circuit that works without triggering anything until soft clipping occurs at 800watts. The 300 B+ supply was monitored and there is only 2 volts sag between cold and warm tube, and that winding is not monitored by the PWM sense circuit. Windings on the same transformer provide the +/- 15, including the heater, which is part of the sense circuit. By removing the heater connection, and supplying it with an external current regulated supply, still shuts down, and repeats the cycling of soft-starts, just as if there was no fault and the power switch was tripped. Grrrr....
He sent it back to Moscow, to the distributor which is the only warranty station in the country, even though under warranty they sent it back unrepaired telling him to buy a new one. So it is a matter of honor to fix the darned thing.
I am generally the one who people come to for fixing things.....not just electronics but cars, air conditioning, computer programs, small business plans, relationship problems etc because the exact same diagnostic skills are needed in each, just different terms and buzz words.
Getting back to the important stuff....had a great dinner Wednesday night as an invited guest to a 800 year old Paris based gourmet club. The wine choice for each course was excellent so naturally I needed to compliment the sommelier who just happened to be a really really cute young woman who eagerly agreed to going to a jazz club tomorrow night, as a show of my appreciation;>) Did I mention that I LOVE it here...?? Photos to follow;>)

[km6xz]

Great explanation Juan, you have more patience than I do. A more direct question might be( to forget method) "what sonic characteristics are sought as the result of all this?" There are many ways to accomplish desired results if the results are specific. It will probably come to pass that the desired outcome could not even be achieved if tubes with 300ohm plate resistance were found using OTL.
What is difficult or impossible, is to design with no specific goal, only a nebulous notion of "pure tube sound". The variables are so many and varied that impact the impression of sound that the specific topography of the system is a couple orders of magnitude lower on the scale of influences compared to other variables that have nothing to do with whether tubes were involved.
The success or failure of a design is primarily determined by how specific and realistic the goals are. Is there an example of music that exhibits the tone characteristics desired? IS the difference heard at all related to being tubes or OTL or is it primarily the magic of modern digital recording? How controllable do you want the onset of the effect to be? And the most important question is whether if achieved, would it make any difference to the music as perceived by the listener? I have decades in the recording industry and have yet to hear a good song enhanced by a particular tone, or a great tone enhancing a bad song. Those who stress most over tone are people who do not play for other people, or have much to communicate, it has been my observation over the years. The people with the communications and performance skills to express what they want to an audience are not so obsessed with tone because they know they can get what is needed out of whatever they play, by how they play and how they craft the song.
If on the other hand this OTL project is simply an engineering exercise....which can be fun in its own right....starting with the basics, such as a mid 50s edition of the Radiotron Handbook would provide an excellent grounding in practical and theory. Not much has changed since the heyday of tube production and design, except names for some parameters.

[J M Fahey]

Thanks.
Can you please post here the Ampeg schematic and point out to the "multi sensor" protection circuit?
So far we have analyzed the tube influence by what it does "simply by existing"= filament consumption, +B consumption, etc.
Now we should see "what it does", "functionally".
Example and just thinking out loud, it might be part of a gain stage , within a feedback loop, it working (amplifying) or not might be detected somewhere *or* its "body heat" might be detected or affect something which is neither physically nor electrically in contact withit *or* it might produce a "turn on thump" which ......
I hope my muddy deductions point somewhere.
The point is; "Magic does not exist" , there is a reason for everything, even if not immediatly apparent.
Of course sometimes it's very difficult to find it.
I guess one way to go is to go backwards, disconnecting sensors one by one until protection is not triggered and only then start guessing "what the h*ck is going on related to the tube" ... or not, it might be a very indirect relation.

Sometimes it's on purpose: In December 2001 we had a big Crisis in Argentina, the President had to run away in helicopter, the Police killing people in the streets, supermarkets were invaded, emptied and burnt to the ground, etc.

It started as a "beat pots and pans" noisy but pacific protest and in 48 hours the Country was on fire, literally.
I had to deliver 2 tube amps, fully paid for (in Argentine Pesos) and I had to buy tube parts in US Dollars, which went from 1 Peso to almost 5.
Had to get unavailable Dollars paying 5X the price I expected or suppliers preferred not to sell.
I mounted full SS amps in them, but each one also had 2 old beaten 6L6 with the filaments connected for looks.
And I put their filaments in series and that current activated a speaker mute delayed relay.
So the amp turned on silently (no SS turn on thump), after 30 seconds warmup sound appeared , and if you pulled either 6L6 the amp never produced any sound.
Perfect illusion.
Best of all this is that I got much praised by the "Tube sound", go figure.
The point is that your Ampeg *may* somewhat sense that the tube is not there , and refuse to power on.

EDIT: video accompanied by some tasty Argentine Rock.

[km6xz]

Hi Juan
The amp is working but it sure took some head-scratching. The problem was a bizarrely mostly function switch mode controller chip. It was hard to find here but I found one with close enough specs. The original is an ICE2A165, a small 8 pin dip that is expected to handle 31 watts of dissipation.....somehow.
I am impressed with the amp's performance, great tone and a Class D with soft-knee'd clipping which should rather tube like. The whole 750 watt amp weighs about 5kgs but 3 kg of that is the heavy steel case. The preamp compressor sounds good and the tone stack sounds a lot like other older SVT's. Class D or some other class of pulse width modulated driver is going to be the standard type power output section in the near future. They sound good, are reliable, much more efficient, generate little heat and have very simple circuits besides weighing almost nothing. Looking at the spectrum analyzer when driving the PA with a complex signal or clipping sinewave really looks more tube like than a conventional push-pull Class AB SS amp, with a natural sounding soft limiting and compression. I've heard enough Class D amps that did not gracefully overload but this one sure is very civilized even with 20db of overdrive.
This all comes back the OP's interest in pure tube sound. Building a tiny reference amp, maybe a push pull 12AT7 driving a moderately hi-Z reactive load to generate tone and using it to drive a large Class D power stage that never gets close to clipping would be the only efficient, light and cheap way to get the power for stage and performance levels with almost no hassle.