• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

What makes the old McIntosh stuff so good?

Status
Not open for further replies.
Jump to Latest
#101
bear said:
Fine, fine. But since we don't have ur schematics to compare nor your results,
Click to expand...

You have it, if you want it. I posted it here several times, no secrets from you. As well as my results. One of first prototypes we measured with SY, before I tweaked it to reduce distortions on power approaching the max limit, you may ask him: he saw nothing except -80 dB of 2'nd order harmonic only on power below 40W.

Also, you can as Cerrem for his opinion as to the complexity of winding the basic McIntosh output transformer. Having not wound any, I can't say that it is more or less difficult. Complexity, it actually seems less than a standard audio transformer. I could be wrong.
Click to expand...

Transformer must be optimal. Higher inductance, lower losses in wire resistance, lower leakage inductances, lower interwinding capacitances, lower distortions due to core non-linearities. It sounds complicated, but actually is not so complex, if it is made for power 50W and below. Just layers need to be wound separately and properly soldered together. That's why typical hi-fi amps were limited by 50W output: to get cheaper transformers. For power 100W and higher it is harder to get right equilibrium between saturation on 20 Hz and ringing right above 20 KHz, that's why local feedbacks are used.
 
Last edited:
#102
Hi, what's a McIntosh K-104? As for the thread theme: I believe what makes the old McIntosh stuff so good is the combination of the cathode coupling and the incredible OPT's. They are the only Xfmrs to give a spark when you put on ohmmeter across the primary and pull the test lead off. I've never seen an EI Xfmr do that. Listen to a Quad or Audio Research with the cathode coupled to the OPT and they all sound very good,IMHO.
 
#104
ChrisA said:
It means if you look up the date each was made you will find one date comes before the other, or "pre-dates" it

The use of horses predates the use of cars for transportation. But the term would be used mainly when the dates are closer than for horses and cars as is the case here
Click to expand...

I got the clarification for term `predate`long time ago from KevinKr.

Anyway: thanks ChrisA
 
#105
Many people also claim that this amp will deliver close to 500 watts. That seems quite a stretch for 8 6lq6 bottles.
Click to expand...

Not a problem. Look through the red board thread if you want to see what sweep tubes can do. I don't play with the 6JE6C/6LQ6 because of the cost, but I have seen 525 watts come from a quad of 35LR6's. Granted they wouldn't live long if asked to make that kind of power for ever, but 8 of them could.

I am not going to debate the pros and cons of the McIntosh design, but you must admit that it was innovative and solved some of the issues with operating tubes at reduced idle currents without undue distortion. The reduced idle current leads to long tube life. Yes, we can do better today, because we have things like mosfets and CCS chips.

I have owned three Mac's all of which were obtained very cheap, fixed up and sold. All had been exposed to the Florida elements and had rusty chassis. They still fetched good money as "parts units". I thought they sounded good, but not outstanding. I could not justify keeping them since I didn't listen to them on a regular basis and needed the cash.

I still have a Scott 272 in my closet. It hasn't seen electricity in about 15 years. That one brought me back from the dark side (Carver / Phase Linear stuff). I have been building tube amps ever since.
 
#106
what is a McIntosh K 104?

bear said:
Bandersnatch, imo, you are just plain wrong. Have you measured any McIntosh style output iron in terms of bandwidth??

Have you looked at what the square wave looks like on a typical McIntosh compared to a typical tube amp?? Do you think you can achieve the same lack of ringing without output iron of wide bandwidth??

Not sure what you mean when you say "it has is a power stage that does not require it."?? What is "it", and what do you mean by "power stage"??

One does not need extra headroom on the standard power McIntosh amp's driver stage, it was needed on the the K-104 with the 810 (hope I have the number right) tubes in the output stage, there positive feedback was needed since the B+ was iirc over 1kv...

dunno who or what you are addressing with the comments on adjustable bias circuits... or matching tubes, etc...

And, I am certainly NOT saying that one can not build an even better McIntosh type output iron than the early factory units, one can today...

I am talking about the original McIntosh tube design, not the later Nestrovich designs here.

_-_-bear
Click to expand...
 
#107
Last edited:
#108
Since the link to the second set of pix is likely to go away in the future:
 

Attachments

  • K-107 MI-200.jpg
    K-107 MI-200.jpg
    56.9 KB · Views: 367
#113
the mcintosh outputstage is superior for good reasons:
lower halfs appear as low impedance voltagesources connected in serial,
this is parallelled by high impedance currentsources,
in other words primary voltages are split in half and currents doubled the best possible way.
The resulting appearing lower primaryimpedanz and the possible thigth coupling is the key to achieving the wide BW and on top of it without any trace crossover distortion despite lean class B operation.
Now, how would you improve on feeding a transformer from a low voltage source parallelled with a unity coupled currentsource?
Seems clear to me that the mcintosh output stage arrangement is superior to any other
known pp-output stage arrangement possible.
 
#114
If you don't like gNFB you don't have to apply it in the unity circuit. 😀 The circuit itself has plenty of local feedback and if you accept a bit higher output impedance there is no need for gNFB.
Here is a unity cathode follower without feedback. The output impedance is measured to 0.9 ohm and distortion is 0.6% 1dB below full power (40w).

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


As in the Mac the OPT acts as a voltage doubler so I need ~200Vp at the grids of the KT88 for full tilt. The OPT (4x312 ohm primary or 5k p-to-p) has an extra winding for the screen grids that tracks the cathode voltage. At high frequencies the 2uF cap shorts the screen to the cathode. The OPT is not bifilar wound so I need this cap to not run into HF phase problems.
It is all driven from a 1CT:1+1 interstage transformer feed by a pair of 6BL6.
 
#115
"Seems clear to me that the mcintosh output stage arrangement is superior to any other known pp-output stage arrangement possible. "

The Mac unity coupled design is clearly top notch, but the Circlotron uses the same scheme and does it even better with only half as much wire needed on the OT. I cannot understand all the DIYaudio threads on trying to exactly duplicate the Mac OTs, when it is so much simpler to wind a Circlotron OT with better performance and no danger of bifilar voltage breakdown either. A split bobbin Circlotron would be the ultimate, with matched P-P coupling to the secondary (which surprisingly the Mac still lacks if you look carefully).

---------------
"Here is a unity cathode follower without feedback. "

Nice circuit. The HF coupling caps for the screen winding here might be helpful in your earlier Symmetrical Feedback design to improve the stability problem (for the driver screens). But it would of course require a full unity type primary.
 
Last edited:
#116
not sure I understand the purpose of the screen windings?

If at HF the screen tracks the cathode, why not all the time?

And if it is a different screen drive (driven by the signal) then isn't that a sort of UL operations, but with the cap having some sort of EQ if the signals on the cathode and the screen winding do not track 1:1??

Trying to get a handle on your scheme...

😀

_-_-bear
 
#118
I use standard OTs in my Pyramids, with resistive parallel feedback by voltage from anodes to grids. The result is almost the same, except I don't run into loss inductance problems on HF since feedback is purely resistive and short, and I don't need huge linear voltage from driver. Though, I need relatively big swing of driving current.
 
#119
The Circlotron's doubled up power supplies are a nuisance, but still easy to obtain high performance (low common mode capacitance) using split bobbin isolated power xfmrs. You can order split bobbin 120:230 V xfmrs off the shelf from Mouser or Allied or ... Then use a Lambda PF500-360 module (Ebay) or equivalent PFC module on each to get regulated, floating 360V.

The HF caps for the screen winding are just to overcome transformer leakage inductance at HF. The screens do track otherwise. I've heard that Mac used the same HF caps in their later tube amplifiers as well for coupling plate to cathode windings. The separate screen winding allows one to use lower DC voltage on the screens than the plates. Increases the selection of useable tubes and lowers distortion as well.
 
Last edited:
#120
Ya know, I've owned many pieces of Mac over the years, apair of 50W-2's MC240, MC30's, MC-275, several preamps and one tuner and I'm still looking for a piece that actually sounds REALLY good. The solid state gear is crap. Seriously, it doesn't take much to leave MAC in the dust as far as sound goes. I've seen much less expensive gear stomp it. From a collectible standpoint it has value from a sound standpoint... bah! Fmor a standpoint of really old gear the old Marantz stuff is where it's at and the prices reflect that.

Mark
 
Status
Not open for further replies.
Top Bottom