A friend brought over a circulotron amplifier and we connected it to my Quad 63s I was immediately overhelmed with boomy one note bass. Knowing the Quads and most ESLs have high impedance in the bass I hooked up a generator to the amp and voltmeter to the speaker and found a very sharp 8dB peak at 50 Hz, the cause of the boomy one note bass. The treble was down several dB for the same reason; the non flat impedance curve.
Cone speakers have similar impedance peaks at resonance in addition to peaks and dips in the entire curve due to multiple drivers and their crossovers. This cannot be ignored unless you really like the modification to the frequency response. Very few speakers have flat impedance curves as this is hard to achieve and still get good frequency response. I know of no speaker maker who designs for high output impedance amplifiers.
One other point. The proper marriage of the ESL and OTL was the KLH 9 speaker and the Futterman amplifier. It was a good marriage for two reasons. The KLH 9 was a 16 ohm speaker that did not go below 12 ohms and the Futterman was a high damping amplifier. The Futterman, like all OTLs love a high impedance load and the KLH was. Now we have ESLs that dip to an ohm which no OTL likes.
I know of no other combination of ESL and high output impedance amplfier that is good one. |
Ralph,
I did read your white paper before entering this discussion. Lets just agree to disagree. Peter Walker designed the 57 to be used with an amplifier with a damping of 20 and even specified the series inductance. To say that these speakers will play with a damping factor of 1 is not fair to the speaker, no matter that some people like a widely altered frequency response. How can we discuss little differences in distortion when the frequency response has been so altered to make the speaker unreconizable?
Thats a nice little list of expensive speakers that represent a vanishingly small part of the market. I agree a single driver speaker like the Lowther will work just fine with your amplifier.
As an antique radio collector I have given a lot of thought to how early SE amps (the 45 in particular) got relatively flat response without feedback. In that case the driver was a single cone in an open baffle cabinet so the impedance rise was not so severe and relatively flat. These early radios sound pretty good. However that does not represent modern popular speakers. |
|
Ralhp,
At least we agree that Paul was a good guy and liked to experiment with different things.
|
Thanks guys, I appreciate the compliments. Making amplifiers has been my life's work as I find them forever fascintating in all their many forms. Though I have put into producion only three power amps and two preamps, there are lots of designs hanging around waiting for the right person to come along. |
HI BIF,
I looked up your amp on the ARC database but no schematic for the 150. Great site for those who are curious about ARC gear. From what you tell me it appears there is only one bias pot per channel. Do you have a schematic for me to look at? I need to know if the driver tube is direct coupled, I do that in the RM-200 and they did in some amps.
I was the first to computer test tube and match them to very tight specs in 1982. I discoverd the Two Point match where once we find the grid volage for a particular current we then find the transconductance. If one matches both of those numbers the tubes will track over a wide range or voltages and currents. Others, as far as I can tell, still set a grid voltage and get a current but they aren't doing what I am doing.
I also test for grid leakage, which is one aspect of what ARC used to call "Low Gas". What they wanted was tubes that didn't run away when they got hot as ARC tubes do..But tubes usually arent Gassy when new, they get gassy from running hot.
On life all I can say is that I consulted with Sylvania Engineers, spend a full day at the plant in Altoona, PA and learned a lot. I was advised that their power tubes can last 10,000 hours if run at 50% of rated dissipation. Its dissipation that kills tubes, makes them gassy and run away. ARC likes to run tubes at high idle currents such as 60-70 MA and at 500 volts thats can be 35 watts which is the max rating for 6550s. Tube life is not linear with dissipation and max rating can reduce life by a factor of 10, so 1000-2000 hours is typical for many amps, though not for my amps which run them at 15 watts in the RM-9.
So it was good for them to go to the KT120 which is rated higher and the KT 150 higher still. Im not sure I agree with those ratings but they certainly are bigger tubes and have a hight dissipation. I would say the KT150 is a bit of overkill but thats what ARC likes to do. While I keep a large stock of KT-120s on hand I have not yet bought a large number of KT150 as they are much more expensive than the 120s. However I think i can beat their price and know I can do better matching if I can get some demand for those tubes. So lets see if that will occurr..
Yes I am indeed Roger A. Modjeski last time I checked :) |
Amen to this Tube Amp post-we need more engineers on here... HI hemigreg, Your have brought up some good points here and I hope you take my expansion on them as a compliment. Many of the readers of this thread have been pulled into impedance and power thinking by some new paradigms that I do not particularly agree with. So here goes. We do need more engineers on here, for one thing to keep the known laws of electricity correct in our discussions. Some manufacturers like to make up stories to defend their equipment's performance, or lack there of. I am always entertained when John Atkinson, a very strait shooter, measures an amplifier and finds horrible results. It is unwise to send JA and amplifier that measures poorly and then have to defend it in the Manufacturers Comments.... which are even more amusing. QUOTE :;"’6-18-2018 2:05amThis is an old topic...literally. In the times when tubes were dominant amplification devises, the output transformer was a necessity (one might say a necessary evil. The tubes are amplifying voltage, so that they work with high impedance loads. The loudspeakers normally have quite low impedance. Thus a transformer is required to match the output of tube amplifier and the loudspeaker. When first transistors have emerged, the schematic design did not evolve immediately. Thus, early transistor amplifiers were very similar to the matured at that time tube amplifiers.
I think its safe to say that when engineers first got transistors they didn't know what to do with them and the evolution of design topologies took awhile to develop into what we have now. I was learning transistors at the same time and had the same early RCA and Motorola books. I was 14 years old and it really wasn't all that difficult to get something that worked. We were all making the circuits out of those manuals which were the most advanced at the time which was 1965. However those schematics did not look like tube schematics at all. However, over time it was realized that transistors work better and amplifiers of current as opposite for tubes which work better as amplifiers of voltage. Thus, transistor amplifiers can and do work well with low impedance load such as loadspeaker directly.
While we consider that transistors are indeed current amplifiers they do not respond linearly to voltage inputs as tubes do. Pentodes act like transdconductance devices which means their current is proportional to their input votage which is a very handy thing to drive a speaker since we start with voltage (from the preamp) and want current for the speaker. Speakers are current driven devices by nature of their physics. What makes tubes need a transformer and transistors not is the fact that high current transistors come easily but high current tubes do not The most popular transistor in the world is the 2N3055, which Dynaco, NAD and many others used. For $1.50 it can do 15 amps of current. The best Horizontal Output tubes (mistakenly called Video Tubes because they were in TV sets) can only do one amp of current but they can do this at a very high voltage, thus the need for the transformer. We are really not matching impedances here, we are using the transformer to exchange voltage for current, as all transformers do. Indeed output transformers are difficult to make, more on that if people want to know, and expensive because they have to get the best workers in the factory to make them. I had one guy I trusted and he made all the RM-9 outputs with exceedingly consistant results. Now, still transformer (autoformer) can assist with the loudspearkers with various impedance...i.e. 4 Ohm vs. 6 Ohm vs. 8 Ohm providing more stable load to a SS amplifier. However, transformer alone is not a perfect transfer devise and making quality transformer for audio output is difficult and costs a lot of money. The typical issues with output transformers are: reduced damping factor and difficulties with driving complex loads, slew rate reduction, additional distortions, etc.
This is the biggest myth being promoted in this threat. An Autoformer or any transformer cannot fix a difficult load and only affects stability in a poorly designed, on the edge amplifier. A good transformer does not reduce damping factor appreciably, or slew rate or change distortion except at the frequency extremes. M6 iron is extremely linear, low hystersis, and low eddy current loss. Those fancy amophous irons don't make transformers appreciably better. What Transformers do is present to the amplifier a particular voltage and current that the output devices like. The taps on tube amps are there for that purpose. You might as well call them A, B and C and choose the one you like best. At least you have a choice. What an Audoformer does is give you that choice with ampifiers that do not have taps. The RM-10 produces an unbelievable 40 watts with one pair of EL84/6BQ5 tubes (typically 17 watts) at less than 1% distortion with an 8 ohm load on the 8 ohm tap. Many people do not need 40 watts and they can get 25 by putting their 8 ohm speaker on the 4 ohm tap resulting in twice the damping and 1/10 the distortion. Most people do this, I encourage them in the manual to try it and it works on all well designed tube amps. Try it! Sorry this is long, but there is a lot of work to be done on this topic. |
10,000 hours is typical- so we warrant the power tubes for a year on this basis, and always have.
Ralph, What is your dissipation figure as a percentage of rated? How do you figure in high current peaks when the amp is used at full power?
I agree about the feedback 'used properly' (which many designers do not) comment. Proper application of feedback is tricky to say the least, and may not have been possible until the age of personal computing, due to the number of variables involved. Here is a nice primer on the topic:http://www.normankoren.com/Audio/FeedbackFidelity.html I have spoken with DImitry and a group of people trying to improve the tube models for SPICE. He put the parameters of the RM-10 into their SPICE program and got results that did not even come close to what an RM-10 actually does, and this is only for the midrange where the output tranformer is considered perfect. He did not attempt any feedback analysis as the output transformer is almost impossible to model where it matters. I know of no tube amp designer who uses SPICE. can you name a few? |
Since the RM-200 produces about the same power on both taps, what possible reason is there for NOT hooking up a nominal 8 ohm speaker to
the 4 ohm tap? Thanks---Eric.
This is something that needs clarification. The reason we have taps on tube amplifiers is to get the full rated power into different loads. That means when we test an amplifier it produces full power into a load of the same resistance as the tap. Indeed hooking a 8 ohm load to the 4 ohm tap results in reduced power. Usually though only a 30% reduction, not 50% due to other factors (reduced loss in the output transformer, power supply and tube saturation voltage). So a typical 100 watt amp puts out 100 watts into any matched tap and somewhere around 75 watts into a tap mismatched by one step. In this case the tube are loafing along, distortion is reduced and damping increased. But this requires that the load does not go significantly below the tap impedance. However going in the other direction where the load is lower than the tap impedance bad things happen. In that case the amplifier puts out less power, works harder and the tubes get overly hot to the point or radically shortening their life. What an RM-200 does is to go into AB2 mode in the above case The tubes stay happy and the reduced load gets extra power in the same way a transistor amp gives more power into a lower load. |
|
Ralph,
Thanks for your answer. I think most of us develop amplifiers in real space verses cyber space. SPICE, by its name, was indeed designed to help people design ICs not tube power amps.
The thing that concerns me about 6AS7s, it that they emit cathode sparks in my Tektronix 530 curver tracer at zero grid.That is consistant with your statement of cathode coating failure as it appears the sparks are bits of cathode coating. They actually look like sparks off of a 4th of July sparkler. |
BIF,
The tubes will be happiest if you use the tap that is equal to or lower than the lowest impedance of the speaker. So if your lowest is 4 ohms use the 4 ohm tap.
Although David Manley (rest his soul) believed that we match the primary of the tranformer to the tube’s impedance this is not true. Once again its all about the voltage and current that the tube is comfortable with. With that in mind, a reduced impedance will require more current from the tube. If the tube is already at max current minimum voltage (the correct place to be at full power) the reduced load impedance will demand too much current and cause a great increase in voltage drop across the tube and overheat it.
A 10% overcurrent may double or triple the voltage drop aross the tube increasing the dissipation by many times. Its a horrible situation. In fact the RM-9 Special addressed this problem by allowing the output tubes to go into AB2 mode thus saving the tubes. The speaker that prompted this was a Theil whose impedance dropped below 4 ohms in the treble region where loud trumpet music would just bake the tubes in a regular RM-9. The AB2 mode is also used in the RM-200 as i liked what it did.
For normal tube amplifiers it would be best to connect the amplifier to the tap that matches the lowest impedance of the speaker in the region where there is a lot of music.
Speaker manufacturers are not so honest about their impedance range and a curve is the only way to know. If they didn’t do so many tricks in the crossover we would not have this problem. The drivers are not the problem, the crossover is.
In my experience most speaker designers do not know much about electronics or care about what the amplifier may have to do.
|
These are examples of speakers that will not play well on low damping amplifiers due to wide impedance swings or very low impedance demanding excessive current. The density of music where dips occurr is important to how much the speaker will overheat the tubes. These are just in the order that I found them in Stereophile under floor standing speakers. I didnt have to look for long. these are the first 4 measured in order. Connected to an amplifier with a damping factor of one at eight ohms the following will happen. Where the impedance approaches 20 ohms there will be a 4-5 dB peak at those frequencies. Where the impedance dips below 3 ohms the respoinse will dip 4-5 dB making a total error in frequency response of 8-10 dB. In addition if the low impedance occurrs where there is a lot of music the amplifier will current clip and the tubes will get hot. Click on the link and open in a new tab for a good look.
The Wilson Alexia 2 is still a current-hungry design. Its impedance drops to 2.6 ohms at 84Hz (fig.1), and there is a demanding combination of 5.1 ohms and –44° electrical phase angle at 57Hz, both frequencies in regions where music can have high energy levels.
https://www.stereophile.com/images/718WAlex2fig1.jpg
Eggleston Viginti June 2018 needs to be connected to the 4 ohm tap and will work it hard in the bass where there is lots of energy. Also note how high the impedance is from 1Khz to 8 Khz where a low damping amplifier will cause a 4-6 dB rise in level making the speaker very bright on trumpet music and anything in that region. https://www.stereophile.com/images/618EggAndrafig1.jpg
B&W 702
The 702 S2's nominal impedance is specified as 8 ohms, with a minimum value of 3.1 ohms DO NOT connect this speaker to the 8 ohm tap. https://www.stereophile.com/images/518BW702fig1.jpg
Monitor Audio Silver 300, JA notes "Although the minimum magnitude is 3.6 ohms between 150 and 170Hz and there is a combination of 5.4 ohms and –39° at 97Hz, this speaker won't tax the amplifiers with which it is used". While this speaker won't tax most SS amps, it will bother a tube amp unless connected to the 4 ohm tap. (RAM) https://www.stereophile.com/images/318MS300fig1.jpg
|
BIF,
I can match your tubes, may have to sub or sell you a few. Will need to know the plate, screen voltages and cathode current which we do know. Write tubeaudiostore@gmail.com to set it up.
I would still like to see a schematic of the amp. Can you get one?
What ARC did to balance things in design is what I do also. Use the 4 ohm tap, it should sound better. |
BIF, Thanks for the info. Let me clairfy this part which many people make the same mistake.
Max power may be compromised at higher impedance, but if I am not pushing the amp, that shouldn't be a problem.
I've heard this from ESL owners where the impedance rises in the bass to very high values. They believe that OTL amplifiers are better for their speakers because they can supply the extra voltage to drive the high impedance in the bass. They indeed get more bass if the damping factor is low, but not the bass the speaker was designed to give. They get more and perhaps like more but it is one note bass. Speaker makers are doing a better and better job making speakers have flat response, which is a good thing. To to this they generally let the impedance do what it will. They could add a lot of parts to the crossover to try to flatten the impedance curve but they assume you are using an amplifier with reasonable damping of 10 and above. Ten is not a hard limit but 5 is better than 1. This is where we get in trouble with the power paradigm which had led people to believe that the speaker wants constant power. It does not. I cannot think of or find a modern speaker that wants constant power, the varying impedance and flat response insure that the designer uses a constant voltage amplifier, ie one with high damping. The reason that John Atkinson puts the response with the simulated speaker first is that he and I feel it is the first thing you are going to hear when you audition that amplifier. What the power paradigm is saying is that flat response is not as important as ills of feedback. Sorry, but I am not in that camp. Proper feedback has no ills. The Futterman amplifiers have over 60 dB of feedback and a good reputation. Thats a lot of feedback, more than most transistor amplifiers. I plan to start a thread on the Futterman design when I get done with this topic. Since everyone hears what they hear and like what they like I would rather not promote one paradigm over the other, I just want to point out what is going on based on my experience and my rules of amplifier and speaker design. I would rather see all speakers have flat impedance curves but this is not how the world is currently set up. Even the DeVore (which has been recommended) has wide impedance variations. While I agree with JA and Ralph that it is easy to drive (stays above 8 ohms) its peaks go off the chart above 20 ohms. A damping factor of 1 or 2 or 3 will modify the nicely flat frequency response several dB. http://https//www.stereophile.com/content/devore-fidelity-orangutan-o96-loudspeaker-measurements Thanks for your question, I hope this clears it up for you and others. Feel free to ask for any further clairication. |
@almarg I've seen that stated before, but I'm wondering how an amp can be designed with 60 db of feedback. Wouldn't its open loop gain have to be enormous to support a closed loop gain that is reasonable? The open loop gain is enormous and obtained entirely in the first tube. That tube drives a split load phase inverter which provides EQUAL drives to the output tubes. Many think the top output tube is a follower and not equally driven but it indeed is equally driven. What is so clever about the circuit is that it is able to employ that high level of feedback without oscillation. This because of the very wide bandwidth of the circuit up to many hundered KHz. Conventional tube amps are limited in feedback by the output transformer. While the original circuit is quite simple, published in AES Journal, it takes some time to fully appreciate what Futterman created. |
In case you are thinking I made this stuff up, this is really all about history; take a look at this Google search, in this case one for a Fisher A-55:
https://www.google.com/search?client=ubuntu&channel=fs&q=fisher+A-55&ie=utf-8&oe=utf... One of the first hits is an image from YouTube, showing the damping control on the amplifier, which is a dual-gang pot. If you look closely, it is labeled "constant voltage" with the control all the way down, "constant power" with the control at noon, and "constant current" with the control all the way up. In this case, voltage and current feedback are balanced against each other in the noon position. I dont think you are making it up Ralph. Do look more closely at the video. The pot is not dual but single with a switch attached, thus the click into the high damping condition which cuts out all the Zmatic circuitry resistor losses. It is indeed a combination of voltage and current feedback as seen in this schematic http://audiophool.com/Schem_A/Fisher_100_schem.gifOne can clearly see the the switch that grounds out the current sense resistors and returns the amp to conventional voltage feedback. The damping spec on this amp is 26, quite good for an amplifier of this vintage. It is interesing to me that although the Z matic control was of interest they made the click to entirely remove it and its losses. |
@bifwynne
Roger has one of the best tube matching services in the country.
I suspect that one reason you have issues with the taps is that the ARC amp really needs a few more db of feedback; 15 db is slightly on the low side to get right. Thanks for the compliment. I did start the whole computer matching program and feel the only correct way to do it is at fixed plate current which I can enter in the test program. Whatever feedback they use is not of importance, its the resulting output impedance that matters. For instance, a triode output tube has inherently low output impedance and provides a damping factor typically of 4-5 without feedback. This is childs play to figure out. Its just the output transformer primary impedance divided by the plate resistance. For a 300B tube that is about 3500/700 ohms= 5. For a pentode output tube its much higher for a 6550 about 3000/27,000 = 0.11 ouch, less than one. Now we need a lot of feedback to get that down. Even triode connected it is only 0.22. Thus not a good triode comparred to the 300B. This is why SET ampifiers can be made without feedback. These tubes were created to do this job. The 45, 2A3, PX4, PX25 all have similar characteristics as does the 6EM7 which I use because it is very close to a 2A3. There are three classes of output tubes. Low impedance triodes as mentioned above, Pentodes (EL34, EL84) and Beam Power tubes (6550, 6L6, and all KT series). The pentodes and beam tubes all have very high plate resistance, so high that we don’t even try to match it. When I was a young lad the idea of measuring feedback was a total mystery until I found out how easy it is to measure. Heres how you do it. Simply connect the proper load to the amplifier, set the output to a very low voltage, remove the feedback and see how many dB the output rises. ARC looks at another way, which I appreciate and use in my thinking. If you think of an amplifier as a power supply then you can easily measure its regulation, as power supply specs do. It can be noted either in percent or dB. This is measured by noting the output voltage with the load connected and then with no load. We want to see the smallest change, as does the speaker. This is a great way to demonstrate what is going on with your speaker. If you pull the load on an amplifier with a damping factor of 1 the output will go up exactly 6dB. It will do about the same on your speaker when it goes above 20 ohms! |
I agree that output impedance is important, but if you wind up using too little feedback to get there, it can be detrimental to the sound thru the process of bifurcation. What is bifurcation in feedback?
There's a range of sorts- 8-15 db or so is where this sort of thing can occur. Here's a nice article by Nelson Pass:
https://www.passlabs.com/press/audio-distortion-and-feedback I read the Nelson's article, Figure 11 doesnt make sense to me as there is no reduction in the lower harmonics with added feedback. I wrote Nelson to ask about it. I would not make any generalization about the proper amount of feedback. I built a Futterman to see what was going on. Sometimes that's the best way to understand something. In his circuit 40-60 dB of feedback is easily applied. Unlike most amplifiers the open loop bandwidth is so wide that the feedback is not a problem. I don't think we can dispute that the Futterman amplifiers are well loved for their sound when used in their power range. The OTL I did for Counterpoint is a Futterman style also. |
Also, Anthony echoed that even if the 4 ohm taps presents an impedance to the output tubes off the primary coil of the output tranny that is higher than the rated 3000 ohms output impedance of the tubes, no harm will occur.
Lets get clear about matching plate impedance to tranformer impedance. We don't do that in amplifier design. The primary impedance reqired is determined by the voltage and current of the output tubes. For instance a EL84 amp with a 385V B+ uses a 8,000 ohm transformer while my EL84 RM-10 has a 13,500 ohm primary because it runs at higher voltage and lower current. While the plate impedance will provide some damping this is not a large effect on the damping when the loop feedback is applied. I have doubts about the plate impedance being 3,000 ohms in pentode. It should be much higher and I will measure it at my first opportunity. In triode it looks like 1 to 2 Kohms. |
BIF, I thought you said before that if the primary impedance presented to the output tubes off the output tranny is less than the plate impedance, it can cause distortion and shorten tube life. Ergo, use the 4 ohm tap. You called it light loading. What I have been saying is that the primary impedance chosen has nothing to do with the plate impedance of the Pentode. Ideal plate resistance of a pentode is very high as demonstrated by the flatness of the plate curves above the knee. If the plate impedance of a KT150 were 3000 ohms it would produce something close to a 45 degree line on the curves. I just think the 3000 ohms is a mistake thought is it also shown on the KT-120 data sheet, which is curious it would be the same for both tubes.. Modern spec sheets are not done with the care of the good old days and there are few out there to check them. RCA, GE, Tungsol and Sylvania all had identical plate curves or at least very close and such a mistake would have been found right away. The printed specs for most tubes are IDENTICAL from one maker to the other. There is no other maker for the KT 120 or 150 to monitor the situation. Thus is the current state of tubes. Here is a much better data sheet, made not by the manufacturer but someone else.. http://www.tubeampdoctor.com/images/File/data%20sheet%20KT150%20Tung-Sol.pdfOn the first page are the pentode curves and the plate resistance is the slope of the line above the knee. The top curve has a slope of about 20 ma/200 volts = 10K ohms, the lower curves, where the amplifier typically operates is flatter making the impedance even higher. Also note on the last page the the the load impedance is given as 3,000 ohms. So that is likely the source of the mistake in the Tungsol data sheet, which was made by the Russians I would think. Perhaps something got lost in the translation. The simplest way to explain light loading is to look at what heavy loading does. If the load is two heavy (4 ohms on the 8 ohm tap) the tube will have excessive current and voltage across it (ie it wont get to the knee but be to the right of it). That product of voltage and current does two things. It reduces the output of the amplifier and it heats the tube excessively. |
The KT-150 spec sheet you showed us says that "Tube impedance is [f]rom 10.0KOhm to 12.5KOhm.
And is even higher at low currents where we are often listening. It really helps for someone get a scope or some kind of monitoring device to see the voltage the speaker is using. Some listeners are at very low levels and some much higher. Anyone who is going through tubes is up at the high endor has a terrible mismatch.This may not be obvious, but once one goes over the knee at 0 grid volts the extra current comes at the expense of large voltage drops across the tube and the amplifier is surely clipping at this point. . There is no optimal load for a power tube. The load belongs to the circuit in which it is used and the load that is expected. That is why an RM-9 can do equally well with the KT 88 family of tubes (beam tubes) or EL34 (true pentodes). Since these tubes have very different curves they will certainly sound different. One must realize that installing different tubes, either the same type with slightly different specs, or another type entirely will mostly change the feedback of the amp, thus the damping and distortion. In most cases the largest difference will be in the bass and overall frequency response due to the higher damping (more feedback) with the beam tubes. |
@toshko1
Hello, I found the thread and read it very carefully. Presently I am reverse engineering this type of block scheme of power amp .
Thanks for reading the thread carefully, however there is not enough information to calculate the transformer design.
Here is what I have found:
Damping factor for MC1200 and 2000 in 8 Ohms is about 40 which translates to 0.2 Ohm output resistance. In my experience with transformers, this translates to about 100 windings of 1.2 mm wire over a large cross section of E+I laminated steel. A little more for double C-core.
2 kW into 8 ohms is 126.5V and 15.8A
What is a large core? The double C results should not be any different than the EI. One needs to know the core area and lowest frequency to calculate the turns. The output impedance is due to many things, not just transformer winding resistance. Feedback makes it low.
MAC output devices are actually step up transformers. Why are they called autoformers is not clear to me.
An auto transformer has just one tapped winding where currents are shared. I wrote an explanation of how it is wound and how it differs from a conventional output transformer. Everything above the 2 ohm tap is step up in voltage. This allowed them to use low voltage transistors.
In the power supply 100V filtering capacitors are used, thus the rail voltage must be less than that voltage, lets say 90V or even less than +/- 80V for longevity. (typical commercial capacitors begin at 1200 hours at rated voltage and temperature and there are some rated at 13000 hours, but some people don’t turn off their gear which makes for a mere 77 weeks in the latter case... lowering the voltage and temperature multiplies the life multiple times)
126.5/80=1.575 is the transmission ratio of the transformer.
100/1.575
2000W and 80V translate to 3.2 Ohms load and 25 Amps current.
Note: output transformers are designed based on allot more complicated calculations.
This is very simple and vague calculation.
It is not unthinkable for the primary winding to be sub 2 ohms and the operating voltage of the supply rails to be under 50V with large currents.
The complication in the reverse engineering comes from the separate winding of the NFB loop, which loads both primary and secondary windings additionally and thus reducing their impedances. The NFB winding does not load anything. It is there for stability under reactive loads.
From the photo of the autoformer in the McIntosh manual, the wires seem to be in the 0.8-1.2mm range. 1 mm wire can withstand 30 Amperes, which leads to primary resistance of about 2.2 Ohms. Can I see this photo please? In addition, aluminum or nickel wires are not unthinkable in regards of 22-30 meters of wire having 2-3 Ohm DCR. I presume it is wound with copper, why anything else?
Why?
Everyone who have tried to lay down the design requirements of a powerful amplifier have come across the design limitations of high voltages.
Reducing the voltage allows for high-end audio grade components such as capacitors and transistors. At 140V there are no capacitors and a limited choice of transistors.
So the autoformer enables McIntosh to design a no compromise SS amp capable of large current due to the high transistor count and powerful power supply at the cost of some manageable complication.
P.S. Regarding the distortion, it is in octaves and in multiples of the base frequency, thus the 7th harmonic of 100 Hz is 14,800 Hz and the 7th harmonic of 1000 Hz is 148,000 Hz... That is why the H-Fi standard measures only to 3rd harmonic only up to 7 kHz.
|
@toshkol Please spell out an abbreviation the first time you use it. What is Le to you? Leakage inductance? Prove you wrong about what? Thats not what I am here for. :) |
