• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Parts and Pieces: A True ARC Mystery

In designing my own amps with the KT120 (as output tubes) I’ve always tried to keep them operating at KT88 levels so they would be interchangeable. So I don’t have any stress info on the KT120 operating at greater than 42 watts plate dissipation.

Will hotspots develop between 42 and 60 watts sustained plate dissipation? Will it withstand the stress while still providing long life? There seems to be little stress info written about the KT120. Here’s hoping it’s a viable solution.

Another thought. Put EFB on it and do away with screen regulation all together. But that would mean reconfiguring the power supply a little to 1) deliver a low impedance unregulated screen voltage (don’t know how that could easily be done), 2) adding more cap filtering to replace equivalent level of filtering offered by the regulator, and 3) removing the bias regulator.

Plus then it wouldn’t be a “stock” amp any longer. So this is probably not a great idea. Seems to me the power supply is one of the hallmark features of this amp (the other being the differential global feedback) and taking one hallmark feature away is taking the amp too far off course.
 
I'm surprised that ARC didn't go solid state with the regulator, as much of their power supply was already solid state
 
OTOH, the only time an audio amp typically sees continuous rated output is on the test bench. Even playing at the onset of clipping with typical program/music, i.e. "band duty", average output power is much less than continuous rated power. That's likely why it works.
 
Appreciate everyone's comments:

Dazza -- You're striking at the heart of the practicality issue. Unless the amp is soooooo heads and tails over anything else in terms of sound quality, or you truly have a real need for that much power (not just a lookie lookie what I've got complex), then the practicality element starts to become an issue real quick. The D250 sounds like an amp most people should lease rather than own -- if that were possible. Like owning a nice sports car -- lease the darn thing, get it out of your emotional system, and then return it. Ownership of such things isn't always real practical.

S -- On the premium level that ARC operates at with their products, I smell the marketing department on this one -- or at least BJ's intent. He was a big believer in the sound produced by tubes over that produced by SS components. As designed, with just a couple of support ICs and a handful of transistors (ok, and all the various diodes and Zeners as well), by keeping the screen regulator primarily vacuum tube based, it could still be called an "all vacuum tube design" without anybody quipping too much about that claim. There is no doubt however that making the design of that regulator fully SS would have completely eliminated the dissipation issue I've raised. Some quick napkin calculations indicate that a single pass device could easily have been used, and it's dissipation level limited to no more than about 10 watts under any condition of power output in one or both channels together. That, and the second screen B+ winding in the power transformer, it's rectifiers and filter cap, could also have been either eliminated, or greatly reduced in size and capability, as at best, then it would only need to support the driver stage...........

99 -- Exactly. That's why I made my comment " To be fair........" in my thoughts on the subject. Still, in their D70 and D250, this issues doesn't exist -- only in the D115. ARC went to great efforts to champion their premium design, components, overall quality, and performance specifications -- and lord knows they charged for those things accordingly. My intent isn't to beat up ARC over the issue but rather, to draw attention that weakness does exist, and suggest use of the KT120 as "hopefully" a perfect solution. It also addresses my inner Band Rated complex. :)

K -- Your thought process followed mine almost to a T when this thing first arrived on my doorstep. But like you, I rather quickly came to the conclusion that no: This thing is what it is, and because of its relative rarity, I should at least finish it out (as a true D115 design, not a V modification of it) to make it operate as intended. That would at least give me the opportunity to put it through its paces, play with it and enjoy it, and retain as much of the original ARC design philosophy as possible. It deserves that much given its likely origin story, and is completely in keeping with my desire to own products -- in part -- out of respect for a company's founder based on the contributions their company made to audio history. William Johnson deserves great credit for keeping vacuum tubes alive in the high fidelity industry, at a time when everybody else (and I mean everybody) had abandoned them. So even though it will be a bit of a bastardized version of the D115 (due to its being mounted in a V70 chassis), it will still be true to its electronic beginnings, with the physical differences hidden behind the front panel.

Wojo -- An amazing story in repairing those boards, but perfectly illustrates how electronics has moved beyond troubleshooting down to the component level, and long ago became a board swap out procedure. I'm sure those boards were incredibly expensive, and the company clearly had a caliber tech in being able to determine what failed with the board. But I doubt it was ever intended to be repaired in that way -- if ever "repaired" at all!

Dave
 
Mr Gillespie

I must admire this undertaking you have brought upon Yourself. I’ve always been fascinated with the ARC product and the man behind it, Mr Johnson. Although it does boggle the mind of the time, effort and complexity that goes into Their builds. All I can say is,I wish you the best.
 
When you go from 4 tube ppp to pp, is it always as straight forward as halving the output impedance, is there no dc resistance in the opt that is considered, so turns ratio is the only thing that matters?
 
When you go from 4 tube ppp to pp, is it always as straight forward as halving the output impedance, is there no dc resistance in the opt that is considered, so turns ratio is the only thing that matters?

There is a relationship between the inductance of the primary of the output transformer, the source impedance of the tubes, and low frequency response and distortion of the amplifier. Doing what Dave suggests will lose some low frequency power delivery capability, but the ARC transformers are likely oversized to the point that it's not like to matter practically, particularly since they're now operating at half power anyway.

Another thing to consider doing this, is reduced loading on the power supply will cause filament and plate voltages to rise, and this can impact reliability and tube life. It would be prudent to conduct an engineering review on the full impact of running an amplifier this way, before proceeding to do it. IMHO you'd be better off just using a smaller amp which was designed to work with two power tubes. I have no doubts that Dave will be able to get the ARC to work acceptably this way, but I definitely wouldn't extrapolate this to all push pull parallel amplifiers, or for that matter most DIY type modifications.
 
with a regulated supply I wouldn't figure the voltages would change much. In a traditional non-regulated one, they definitely would.
 
B+ to all small signal tubes is regulated, as is output tube screen and control grid voltage. DC heater voltage to the phase inverter and 1st AF Amplifier stages is also regulated. The only power sources that are not regulated in the design are the AC heater voltage sources (for all 6550s and 6FQ7 tubes), and the plate B+ for the output stage.

When removing 4 of the output tubes, the voltage to the AC powered heaters will likely increase somewhat, so it will be something I'll have to check -- on top of checking the general AC heater voltage levels to check how appropriate the primary windings are for today's line voltage levels anyway. The schematic indicates the unit can operate properly from a line voltage range of 105 vac to 130 vac when the 120 volt tap is connected. A tap is also provided for 100 vac operation as well. But this is all based on the single primary transformer shown on the schematic. The transformer in my unit however is a dual primary type, with each (presumably) 120 volt winding also having (presumably) a 100 volt tap as well. I won't know anymore until I can get the unit actually powered up with all tubes installed.

As for the effect of removing 2 output tubes in each channel on the OPT, the effect is actually quite opposite that suggested, where the use of what then amounts to an effectively well oversized OPT extends the LF power response of the amplifier, and also increases transformer efficiency due to the lower maximum power level the transformer is dealing with. As a result, power output from such a move is typically more than half the power output that four tubes produce. If anything, the biggest concern will be to watch what effect removing two tubes has on the HF end of the power response curve. Here, the winding capacitance of a transformer designed to handle basically twice the power actually generated can potentially load down the two tubes left operating, so that the HF end of the power response curve can be reduced. Again, this will have to be checked once the unit is actually powered up, and these tests can be run.

OK, a few pics to show current status:

Below: Just as ARC had to do, the chassis has now been relieved so as to give proper clearance between the phase inverter tube, and the rear side chassis cutout for the board.
SAM_3109.JPG

Below: The clearance between the adjusting screw of the left electrolytic cap mounting ring and the rear side of the front lip of the cutout in the chassis for the circuit board is very tight -- but works. No sparks fly if they touch, but it could create a ground loop since the chassis is only grounded to the board via the board mounting screw nearest the phase inverter tube (shown above). Board positioning was critical, but has been successfully implemented.
SAM_3111.JPG

Below: A very important success point in the project: The circuit board is now fully mounted in place by all 16 board mounting screws, providing this pic, as well as the two previous pics. It's actually starting to look like something! Note too that a black terminal has now been installed to identify the Com terminal in the Right Channel's upgraded output terminals previously installed before arriving here. With this board install, notes have also been taken as to the pre-wiring that needs to be done before it is permanently installed. Some leads need to be attached to the board, while others to chassis mount components before the board's final installation. Details, details.
SAM_3110.JPG


I am on the fence about whether to counter-sink the new holes for the board mounting screws as the original holes were, as the new holes run into the old ones, and I'm afraid the counter-sink might rip up the elongated holes as they currently exist to actually make it look less acceptable than it does now. Also, any cuts made to the chassis cut below the anodizing effect, causing an appearance which stands in sharp contrast to the anodized look of the original chassis stamping anywhere that contrast could be seen. As is with the screws used, the appearance is that the chassis is one that can accommodate two different size boards, with the holes for the larger size board simply being used in this case. Actually, that's all quite true now, making me tempted to leave well enough alone, or possibly find some appropriate anodized fender type washers to cover the original holes. That would require longer mounting screws, as the ones used are perfectly suited length wise as is, but probably not long enough if washers are installed. Such washers might also actually draw more attention to the unused holes as well, so again, might be best to leave well enough alone. Not being counter sunk, they do interfere with the cage setting flush on the top of the chassis. But with the low profile board mounting screws used, the cage could be mounted on 1/8th inch (or shorter) spacers, and call it a day. You'd never really know unless you knew to look for it, and it would allow more cooling air to enter the cage as well. More stuff to ponder.

Dave
 
Nice to learn that the AR is made working. Looking forward to measurement when it's up & running.
 
I really enjoy reading this thread especially since I just purchased a mint pair of Classic 120's and retubed & biased them. I have wanted a pair since they came out in the '90s but simply couldn't afford the $7G entry fee. The D 115 are amazing amps and probably the first amps that I ever heard that made me think that there was something special about tubes. I still have never heard a SS amp that gives a 3 dimensional sound stage and the jump (immediacy?) of tubes. Great thread!
 
Dave, there's a product called "Aluminum Black" that is available in gun stores, that should be able to "re-anodize" those bare spots from relieving the aluminum, back to black. It's been a while since I saw someone use that product- but I remember it working pretty well in at least some cases.

EDIT: Ah, I found it: https://www.amazon.com/BW-Casey-Aluminum-Black-Touch-Up/dp/B002JCW2CG

Regards,
Gordon.
I forgot about that stuff! My father was a gunsmith and used it. He was relay good at bluing and other older styles of metal protection. One was a very old prossess called browning. Acually heating the metal and quenching it in water and such. Even using urine to speed up the prossess of rusting. Once done it looked very nice, and just as shiny as bluing. The aluminum black should work very well for touch ups.
 
In designing my own amps with the KT120 (as output tubes) I’ve always tried to keep them operating at KT88 levels so they would be interchangeable. So I don’t have any stress info on the KT120 operating at greater than 42 watts plate dissipation.

Will hotspots develop between 42 and 60 watts sustained plate dissipation? Will it withstand the stress while still providing long life? There seems to be little stress info written about the KT120. Here’s hoping it’s a viable solution.

Another thought. Put EFB on it and do away with screen regulation all together. But that would mean reconfiguring the power supply a little to 1) deliver a low impedance unregulated screen voltage (don’t know how that could easily be done), 2) adding more cap filtering to replace equivalent level of filtering offered by the regulator, and 3) removing the bias regulator.

Plus then it wouldn’t be a “stock” amp any longer. So this is probably not a great idea. Seems to me the power supply is one of the hallmark features of this amp (the other being the differential global feedback) and taking one hallmark feature away is taking the amp too far off course.

I appreciate the wanting to remain stock theme.
This unit never existing stock sort of negates that thought:idea:

Middle of December I had to open my balcony door because after 5 minutes the room was unbearably hot. 20 KT88's probably the reason why, 2 big silent fans blowing heat out...

Just from the thought of having to buy 20 kt88s to retube periodically ( not to mention powering them or heat dissipation (the room).
Clearly those could be crossed off my wish list.

I like Dave's idea (leasing).
What's that they say Monkeys like shiny things:banana:

I had one of these for a bit and thought I was crazyatma-sphere-s-30-otl-amplifier-2-676x451.jpg
 
Back
Top Bottom