Main reason I see cloning the Mac-275 re-issue version possible now, besides the simpler two winding OT primary, is the class AB2 relief now versus the class B originally imagined. This will have low crossover distortion, even if one has to reduce the global feedback some for stability now. The Hammond 1650T may seem too pedestrian an OT, but in Circlotron mode its leakage L is halved versus normal operation (leakage L from both primary sides are now in parallel), meaning the HF bandlimit is increased by 40%.
Most of the feedback in the Mac appears to be local CFB (not thru the OT) back to the driver stage and the output stage. So only the global feedback from the secondary is in question as to stability. It could easily be reduced if necessary for stability.
The global feedback pickoff might be possible with either the newer balanced scheme or the older single ended scheme by using the secondary instead of the special tertiary feedback winding on the Mac OT (for the newer balanced feedback, the 4 Ohm tap gets grounded). If Pieter K can tell us the turns ratio of the Mac OT 8 Ohm secondary versus the feedback winding turns, we can alter the feedback resistors appropriately.
----------------------------------
I was just looking at the Citation II schematic, which is another highly regarded high feedback amplifier. And I have to eat my earlier comments about its "losing it" in the crossed feedbacks. It DOES have a differential driver stage, so crossing the feedbacks makes no difference (ie for local fdbk status) (well, phasing for neg. fdbk does have to be correct obviously) (and the 680K feedback resistors are high enough to avoid loading the momentarily undriven side of the OT, avoiding leakage L effects in the feedbacks).
Putting in a CCS for the driver tail might be beneficial for the splitting. (if 3rd harmonic nulling is not resulting from the present finite R tail) Just the UL output is sub-par now in my opinion. Unfortunately this requires a very good OT, so no cheapie OT is possible here. Might be some new equivalent OTs available for it, likely pricey.
What is most interesting here now is that the differential driver stage with the crossed local feedbacks from the OT ---IS---- implementing the "magic" Schade feedback scheme I mentioned earlier. (can either go directly to the driver cathodes like earlier, or can go crossed wise to the driver grids like here) This may well be part of the real magic in the Citation II.
-------------------------
Just for fun, I conjured up some changes to make the Cit II into a Mac-Citation:
We change the output OT to Elliptron mode, so we can use the cheap Hammond 1650T in Circlotron like mode. The Elliptron mode, using the 40% OT UL taps, gives 28% CFB for the KT88s. Now we need more grid drive for the KT88s, so we change the driver stage from 12BY7 to 12HL7 or 12GN7/12HG7 to get more gm. The individual driver plate resistors get "booted" over to the opposite KT88 plates, like the Mac, to provide sufficient B+ headroom. Then some grid biasing V fixup for 12HL7/12GN7/...s in place of the 12BY7s.
And the local driver tube "Schade" networks (12HL7 plate to grid) get modded to increase the "Mu" of the drivers for the higher grid drive needed for the KT88s (due to the 28% CFB). The crossed "magic" Schade networks ( crossed KT88 plates to the 12HL7 driver grids ) get modded slightly to allow for the reduced 3/4 signal now on the KT88 plates (the other 28% moved to the KT88 cathodes). Input stage can stay the same, but maybe use the more readily available 6JC6A, or just overkill it with another 12HL7. Oh, and the KT88 screen grids get a fixed regulated voltage. (this still provides reduced 28% UL though, due to the cathodes moving) And the B+ gets lowered to the Mac B+ since we are using a 4K P-P load effectively and we don't want to toast those 12HL7s either. And NO cheeso voltage doubler for the B+ anymore either. Well, we do need a couple floating B+s for the KT88 Elliptron setup now too, beside a lowish B+ for the front end tube.
Let the games begin.......!
Most of the feedback in the Mac appears to be local CFB (not thru the OT) back to the driver stage and the output stage. So only the global feedback from the secondary is in question as to stability. It could easily be reduced if necessary for stability.
The global feedback pickoff might be possible with either the newer balanced scheme or the older single ended scheme by using the secondary instead of the special tertiary feedback winding on the Mac OT (for the newer balanced feedback, the 4 Ohm tap gets grounded). If Pieter K can tell us the turns ratio of the Mac OT 8 Ohm secondary versus the feedback winding turns, we can alter the feedback resistors appropriately.
----------------------------------
I was just looking at the Citation II schematic, which is another highly regarded high feedback amplifier. And I have to eat my earlier comments about its "losing it" in the crossed feedbacks. It DOES have a differential driver stage, so crossing the feedbacks makes no difference (ie for local fdbk status) (well, phasing for neg. fdbk does have to be correct obviously) (and the 680K feedback resistors are high enough to avoid loading the momentarily undriven side of the OT, avoiding leakage L effects in the feedbacks).
Putting in a CCS for the driver tail might be beneficial for the splitting. (if 3rd harmonic nulling is not resulting from the present finite R tail) Just the UL output is sub-par now in my opinion. Unfortunately this requires a very good OT, so no cheapie OT is possible here. Might be some new equivalent OTs available for it, likely pricey.
What is most interesting here now is that the differential driver stage with the crossed local feedbacks from the OT ---IS---- implementing the "magic" Schade feedback scheme I mentioned earlier. (can either go directly to the driver cathodes like earlier, or can go crossed wise to the driver grids like here) This may well be part of the real magic in the Citation II.
-------------------------
Just for fun, I conjured up some changes to make the Cit II into a Mac-Citation:
We change the output OT to Elliptron mode, so we can use the cheap Hammond 1650T in Circlotron like mode. The Elliptron mode, using the 40% OT UL taps, gives 28% CFB for the KT88s. Now we need more grid drive for the KT88s, so we change the driver stage from 12BY7 to 12HL7 or 12GN7/12HG7 to get more gm. The individual driver plate resistors get "booted" over to the opposite KT88 plates, like the Mac, to provide sufficient B+ headroom. Then some grid biasing V fixup for 12HL7/12GN7/...s in place of the 12BY7s.
And the local driver tube "Schade" networks (12HL7 plate to grid) get modded to increase the "Mu" of the drivers for the higher grid drive needed for the KT88s (due to the 28% CFB). The crossed "magic" Schade networks ( crossed KT88 plates to the 12HL7 driver grids ) get modded slightly to allow for the reduced 3/4 signal now on the KT88 plates (the other 28% moved to the KT88 cathodes). Input stage can stay the same, but maybe use the more readily available 6JC6A, or just overkill it with another 12HL7. Oh, and the KT88 screen grids get a fixed regulated voltage. (this still provides reduced 28% UL though, due to the cathodes moving) And the B+ gets lowered to the Mac B+ since we are using a 4K P-P load effectively and we don't want to toast those 12HL7s either. And NO cheeso voltage doubler for the B+ anymore either. Well, we do need a couple floating B+s for the KT88 Elliptron setup now too, beside a lowish B+ for the front end tube.
Let the games begin.......!
Last edited:
"confusification"
Take two Schades and call me in the morning.
Mc-275 re-issue schematic:
http://farm5.static.flickr.com/4037/...fa12827f_b.jpg
Citation II schematic:
http://pages.prodigy.net/jimmcshane/CitIIsch.jpg
Elliptron schematic:
http://www.diyaudio.com/forums/tubes-valves/19624-elliptron-schematic.html
(Elliptron is just a Circlotron with the cathodes returned to the UL taps to lower the CFB % to around 28%)
Some afterthoughts on the Cit-Mac confusion II:
Moving the 12HL7 driver load resistors over to the opposite KT-88 plates for booting is not going to work correctly like for the Mac, because the KT-88 plate is running 72% and the cathode is running 28%. (not 50 - 50 matched, like for Mac or Circlotron)
So this would require the old fashioned type bootstraps, with two resistors in series to B+ and the resistor center point cap coupled to the same side KT88 cathode.
Already having 28% CFB in the KT88 outputs now with Elliptron, we really don't need any more UL Fdbk, so the KT88 screens could be similarly cap coupled to the KT88 cathodes, a zener across the cap, and a resistor to the screen supply.
Might be useful to move the splitter up front too, instead of the driver stage as used in the Cit II, so two 6JC6A or 12HL7 up front instead, CCS tail for them. Trying to get big increased gain for CFB drives and clean splitting all in one driver stage could be asking too much.
-----------------
On the leakage L improvement in Circlotron (and Elliptron), the halving of leakage L may actually increase the upper HF limit by 2X (instead of 1.414 X, I was thinking Sqrt(LC) ) since the distributed capacitance is more or less knocked out by the low drive Z from local feedbacks. We might get close to the Mac 275; 95 KHz was it?
Take two Schades and call me in the morning.
Mc-275 re-issue schematic:
http://farm5.static.flickr.com/4037/...fa12827f_b.jpg
Citation II schematic:
http://pages.prodigy.net/jimmcshane/CitIIsch.jpg
Elliptron schematic:
http://www.diyaudio.com/forums/tubes-valves/19624-elliptron-schematic.html
(Elliptron is just a Circlotron with the cathodes returned to the UL taps to lower the CFB % to around 28%)
Some afterthoughts on the Cit-Mac confusion II:
Moving the 12HL7 driver load resistors over to the opposite KT-88 plates for booting is not going to work correctly like for the Mac, because the KT-88 plate is running 72% and the cathode is running 28%. (not 50 - 50 matched, like for Mac or Circlotron)
So this would require the old fashioned type bootstraps, with two resistors in series to B+ and the resistor center point cap coupled to the same side KT88 cathode.
Already having 28% CFB in the KT88 outputs now with Elliptron, we really don't need any more UL Fdbk, so the KT88 screens could be similarly cap coupled to the KT88 cathodes, a zener across the cap, and a resistor to the screen supply.
Might be useful to move the splitter up front too, instead of the driver stage as used in the Cit II, so two 6JC6A or 12HL7 up front instead, CCS tail for them. Trying to get big increased gain for CFB drives and clean splitting all in one driver stage could be asking too much.
-----------------
On the leakage L improvement in Circlotron (and Elliptron), the halving of leakage L may actually increase the upper HF limit by 2X (instead of 1.414 X, I was thinking Sqrt(LC) ) since the distributed capacitance is more or less knocked out by the low drive Z from local feedbacks. We might get close to the Mac 275; 95 KHz was it?
Attachments
Last edited:
There is no cathode feedback in that elliptothingamabobber.
Unless those floating power supplies have some ref to each
other and/or ground, there can be none. its still just a plain
old circlotron that doesn't take full advantage of the primary.
Uh, nevermind: Input grids ref to GND, my bad...
MC275 link is broke though, at least that one's not my fault.
Unless those floating power supplies have some ref to each
other and/or ground, there can be none. its still just a plain
old circlotron that doesn't take full advantage of the primary.
Uh, nevermind: Input grids ref to GND, my bad...
MC275 link is broke though, at least that one's not my fault.
Last edited:
The 8 ohm tap has 120 turns; the feedback taps have 80 turns centertapped (40 turns/phase).
The Mac re-issue link:
http://farm5.static.flickr.com/4037/4204605299_bdfa12827f_b.jpg
(Seems I can't pull a link out of another post using just the copy edit.)
Pieter:
Thanks for the info. I figured the feedback winding would be similar to a secondary. Easy to adjust the circuit.
http://farm5.static.flickr.com/4037/4204605299_bdfa12827f_b.jpg
(Seems I can't pull a link out of another post using just the copy edit.)
Pieter:
Thanks for the info. I figured the feedback winding would be similar to a secondary. Easy to adjust the circuit.
Last edited:
About McIntosh amps
IMHO, dear friend, I were read this post from the year 2010 and maybe a litle obsolete, I don't know if some body of the coforers dye at this time. The fact is you have a lot of energy to build an amp similar to MCint. The answer is you can if you want (Sigmund Freud) and if you say not can is because you don't want....
Words but you need facts, in the middle of them SY is a partner of this Monster Forum, counsel you that read Morgan Jones book. I see this book for years and all the time I asked to myself, what you need this book? But one big fact is I got this book and my life change for ever. When I going to anywhere i bring my valve amplifiers. Is an amazing book.
I have a friend here in Menorca, that is cooker and play electric guitar, He doesn't know any electronics science but any at all. He got a diagram of a Fender valve guitar amp, he went to the metal store, he buy a piece of stainless steel, he cut it and he has buy every parts and he made a Fender clone that sound amazing.
Anyway Mcintosh is to different animal, maybe you can make a Porshe 911 or you can repair the gearbox of a ferrari, but the most probably fact is : you need many time for that. Now after many many time, i trying to get a Mc75 old fashion ones. Maybe is a good choice for you, get one, there are for 2000 dollars in US, you can rebuild and experiment with the circuit or changing some valves. I listen one MC275 many time ago, is euphonic to my taste. Maybe so transitorized sound like. But I want one in my living room. Previously i want it in my worktable for change some guts. Probably is a obsession like the yours. Or obstination more.
Best Regards and i hope this message helps you.
IMHO, dear friend, I were read this post from the year 2010 and maybe a litle obsolete, I don't know if some body of the coforers dye at this time. The fact is you have a lot of energy to build an amp similar to MCint. The answer is you can if you want (Sigmund Freud) and if you say not can is because you don't want....
Words but you need facts, in the middle of them SY is a partner of this Monster Forum, counsel you that read Morgan Jones book. I see this book for years and all the time I asked to myself, what you need this book? But one big fact is I got this book and my life change for ever. When I going to anywhere i bring my valve amplifiers. Is an amazing book.
I have a friend here in Menorca, that is cooker and play electric guitar, He doesn't know any electronics science but any at all. He got a diagram of a Fender valve guitar amp, he went to the metal store, he buy a piece of stainless steel, he cut it and he has buy every parts and he made a Fender clone that sound amazing.
Anyway Mcintosh is to different animal, maybe you can make a Porshe 911 or you can repair the gearbox of a ferrari, but the most probably fact is : you need many time for that. Now after many many time, i trying to get a Mc75 old fashion ones. Maybe is a good choice for you, get one, there are for 2000 dollars in US, you can rebuild and experiment with the circuit or changing some valves. I listen one MC275 many time ago, is euphonic to my taste. Maybe so transitorized sound like. But I want one in my living room. Previously i want it in my worktable for change some guts. Probably is a obsession like the yours. Or obstination more.
Best Regards and i hope this message helps you.
Last edited:
There are several derivations of "the McIntosh circuit". Crowhurst goes through about every conceivable one in some of his articles, but it is of academic interest because many are not feasible to build in real life.
The genius of McIntosh is that _their version_ is the most elegant imaginable from a construction and maintenance standpoint, and was relatively inexpensive to build if one was willing and able to wind the transformers to suit. They were.
They aggressively patented their products, and they never had much infringement. Those patents have all expired of course.
No one really wants to wind McIntosh multifilar transformers because they require some odd fixturing and a specially ordered high voltage insulated magnet wire. This includes McIntosh themselves who have changed the transformer and circuit substantially on the reissue 75/275.
Others have built similar or derivative amplifiers. Several of 'Captain Catchfire's' designs are Mc derivative including one of the Michaelson and Austin TVAs and the EAR 509/549. Milojub Nestorovic's post-Mcintosh units, and one or two other high end production units. Amateur designs include the Lockhart thesis project amp, and there was an article by a professor of the University of Cincinnati, on a DIY class B amplifier of essentially McIntosh layout with some changes. But none of these quite worked as well as McIntosh's own designs.
They were all double C cores with a small fishpaper gap, banded with a regular steel shipping-box-banding machine. This per Larry Fish RIP.
The MI200 was wound on 3 C-cores in an odd shape but is an anomaly.
Quad Formvar wire was specified and was specially ordered.
Mcintosh built their own winding machine and as sometimes happens did what is considered impossible because they did not know any better. Eery aspect of the design is dependent on every other:for example, it would have started buzzing badly in short order except it was tar potted in a can made for oil burner ignition transformers.
Mac power transformers were usually outswourced often from Endicott COil. They are EI designs because C cores were expensive. There is no inessential build cost in any McIntosh product.
IMHO, dear friend, I were read this post from the year 2010 and maybe a litle obsolete, I don't know if some body of the coforers dye at this time. The fact is you have a lot of energy to build an amp similar to MCint. The answer is you can if you want (Sigmund Freud) and if you say not can is because you don't want....
Words but you need facts, in the middle of them SY is a partner of this Monster Forum, counsel you that read Morgan Jones book. I see this book for years and all the time I asked to myself, what you need this book? But one big fact is I got this book and my life change for ever. When I going to anywhere i bring my valve amplifiers. Is an amazing book.
I have a friend here in Menorca, that is cooker and play electric guitar, He doesn't know any electronics science but any at all. He got a diagram of a Fender valve guitar amp, he went to the metal store, he buy a piece of stainless steel, he cut it and he has buy every parts and he made a Fender clone that sound amazing.
Anyway Mcintosh is to different animal, maybe you can make a Porshe 911 or you can repair the gearbox of a ferrari, but the most probably fact is : you need many time for that. Now after many many time, i trying to get a Mc75 old fashion ones. Maybe is a good choice for you, get one, there are for 2000 dollars in US, you can rebuild and experiment with the circuit or changing some valves. I listen one MC275 many time ago, is euphonic to my taste. Maybe so transitorized sound like. But I want one in my living room. Previously i want it in my worktable for change some guts. Probably is a obsession like the yours. Or obstination more.
Best Regards and i hope this message helps you.
Every time I read things like this I have to roll my eyes a bit. Tube sound. I guess there is a little truth to that and there are a few reasons why that has a ring of truth.
First, there is the bit about distortion of tubes versus transistors. However, that argument falls apart when you consider that normal (if not ideal) listening levels should never induce distortion in the amp, if you size the system correctly for the task.
That means the amp should never operate in a zone where it is pushed so hard as to generate audible distortion in the first place. It doesn't matter what topology you choose, it should ever have to be overdriven to get the SPL level required for the job.
So, poof! There goes the distortion myth.
So, where is the tube sound? It turns out there is actually a tube sound, but it comes from a source that you would not expect - your speakers!
Sounds crazy (no pun intended)? Not really. You can clearly get a different sound between a tube amp and a transistor amp (running below audible distortion levels) and the difference has to do with frequency response.
In an ideal world a transistor amp and a tube amp should sound exactly the same, but speakers are not an ideal load and impedance varies with frequency when inductors are introduced to the load.
The root of the problem is the way a tube amp and a transistor amp operate into a load. A transistor amp is a Voltage Drive amplifier and a tube amp is a Current Drive amplifier - well sort of, but for this discussion we will just call it so.
What that means in a nutshell is that a voltage drive amp's power will decrease as the impedance of the speaker rises. The opposite happens when a tube amp sees a higher impedance load. its power output increases.
Now, let's consider the average speaker impedance curve over its operating frequency. The type of base cabinet (sealed versus ported) has some impact, but both tend to have a higher impedance in the bass region, which decreases in the midrange, and rises slightly in the treble region.
What does that mean to the listener? Well, a transistor amp sounds a little sterile because the bass is a bit wimpy and the treble lacks that sparkle because the power drops off as the impedance goes up.
If the impedance curve has some bumps associated with the crossover it further aggravates the problem a little.
A tube amp tends to boost the bass a little as well as the treble and just shines more than the transistor amp, which seems a bit lifeless compared to the boosted tube sound.
There are other factors at work here, but I wanted to illustrate one of the important differences that makes the two amplifier types sound the way they do.
A high end sound system with a really good speaker system (where the impedance is fairly constant) should sound about the same with a tube amp versus a transistor amp. However, most speakers are not a flat line when it comes to impedance and therefore react differently when driven by a voltage drive and current drive amplifier.
Then there is the whole subjective idea of "magic", but I refuse to go down to that level of a discussion.
The bottom line is you should correctly size your amp for the job at hand and make sure your speaker impedance is as flat as possible. Then there is the whole room acoustic thing, but that is another form.
Old thread, but I might as well add that I have built a Unity-Coupled amp with the Plitron transformers and they worked quite well. The winding resistances were quite low as well so I ended up with a very low Zout (~1 Ohm) despite the fact that I used no global negative feedback loop. It is a very nice transformer.
The main problem with the Plitron transformers is that the impedance ratio is not optimal. The transformers are rated for 70W but you have to use really high B+ to get that power output with an 8 Ohm speaker. Driving a 4 Ohm speaker would be the way to go if you wanted full power from the transformers, but I don't have 4 Ohm speakers.
If you design an amp for 8 Ohm speakers and 450-500V B+, the load line will cross well under the knee of the pentode curves because of the turns ratio of the transformer and the high screen voltage if you connect screens as McIntosh did. That bothered me, so I included a floating screen regulator to drop the screen voltage down a bit (and bring the Vg = 0 knee down as well). After loading the tubes like that, the amp delivers ~40W into 8 Ohms when driving the tubes to saturation with a B+ of 450V. To get more power than that you would have to raise B+ or go with 4 Ohm speakers. It would have been nice if van der Veen had designed these transformers with a 4-8-16 tapped secondary. Unfortunately, he didn't.
Anyway, description of my amp is here for any who are interested.
The main problem with the Plitron transformers is that the impedance ratio is not optimal. The transformers are rated for 70W but you have to use really high B+ to get that power output with an 8 Ohm speaker. Driving a 4 Ohm speaker would be the way to go if you wanted full power from the transformers, but I don't have 4 Ohm speakers.
If you design an amp for 8 Ohm speakers and 450-500V B+, the load line will cross well under the knee of the pentode curves because of the turns ratio of the transformer and the high screen voltage if you connect screens as McIntosh did. That bothered me, so I included a floating screen regulator to drop the screen voltage down a bit (and bring the Vg = 0 knee down as well). After loading the tubes like that, the amp delivers ~40W into 8 Ohms when driving the tubes to saturation with a B+ of 450V. To get more power than that you would have to raise B+ or go with 4 Ohm speakers. It would have been nice if van der Veen had designed these transformers with a 4-8-16 tapped secondary. Unfortunately, he didn't.
Anyway, description of my amp is here for any who are interested.
Must keep the zombie thread alive...
My dad's building an MC-275 clone. He picked up the power and output transformers that were custom wound by Dennis "Doc" Hoyer over at:
AUDIO TRANSFORMERS
185 N. 85th STREET
WAUWATOSA, WI 53226-4601
Hoyer is highly recommended by Roger Russel (Roger Russell, McIntosh Lab., Scripto and Norma Pencils and leads) who was the director of Acoustic Research at McIntosh.
My dad picked up the trasnformers about 10 years ago along with a pair of 100W Marantz power supply and output transformers. The MC275 transformers have been gathering dust for the last 10 years so I suggested he get going on an amp for them. I'm looking for a reproduction chassis for his Christmas present; otherwise he should have all the parts he needs.
My dad's building an MC-275 clone. He picked up the power and output transformers that were custom wound by Dennis "Doc" Hoyer over at:
AUDIO TRANSFORMERS
185 N. 85th STREET
WAUWATOSA, WI 53226-4601
Hoyer is highly recommended by Roger Russel (Roger Russell, McIntosh Lab., Scripto and Norma Pencils and leads) who was the director of Acoustic Research at McIntosh.
My dad picked up the trasnformers about 10 years ago along with a pair of 100W Marantz power supply and output transformers. The MC275 transformers have been gathering dust for the last 10 years so I suggested he get going on an amp for them. I'm looking for a reproduction chassis for his Christmas present; otherwise he should have all the parts he needs.
Good luck. I don't know that there is enough of a market for people who build McIntosh clones for someone to turn a profit on reproduction chassis. I could be wrong, though.
That's a pretty valid point. There are a lot of barriers to entry for cloning one so odds are pretty slim that there's a bustling market for reproduction parts but I'll post back when I find one. This thing's going to be hefty too with all that iron on a single chassis.
They pop up on eBay every now and then:
McIntosh MC275 MC 275 New Reproduction Chrome Chassis | eBay
Good luck, and keep us updated!
P.S.: Getting the emblems/logos is the hard part. Been (not very actively) looking for some for my MC75's for quite some time now...
McIntosh MC275 MC 275 New Reproduction Chrome Chassis | eBay
Good luck, and keep us updated!
P.S.: Getting the emblems/logos is the hard part. Been (not very actively) looking for some for my MC75's for quite some time now...
Attachments
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.