Rx For the Magnavox 8800 Series

Once again then, this really only leaves the new manufacture Z565 Dynaco transformers as (to my knowledge) the only truly high quality transformers that could properly work in this application on BOTH the 4Ω and 8Ω taps -- and that would take some significant design changes to accomplish at that.
Was the Z565 tested earlier? If so, could you please provide a link or the test results.

TIA,
cz
 
Dave, what all factors did you include when changing from 2200/390 to 1200/100 ohm NFB resisters in the new front end. (edit for clarification, were you optimizing for HF stability, maximum output, best measurements or just to compensate for changes) I want to convert my 8802 to using your floating front end, my output stage is already modified using different OPT w/7.7K reflected impedance, and I've already separated the channels, increased output tube bias, and capacitor bypassed just less the AC and DC balancing circuits you've included. I appreciate that you described how to incorporate the front end less the balancing resistors. Love these little underdog amps.
 
Last edited:
Looking forward to your further testing with the other OPT as they would more closely match mine @ 8k. I really enjoy how you always include a lot of measurements both pre and post modification.
 
Last edited:
A REAL Dragon Slayer -- In More Ways Than One -- Part I

As the third in a series of Magnavox redux projects, this one should have been a fairly straight forward project, unique only in the things that set the 8800 apart from the previous 8600 and 9300 projects -- that being smaller OPTs and less capable power supply (than the 9300 series), and the use of 6V6 output tubes. The clear bottom feeder of Magnavox push-pull amplifiers, the 8800 series presents the usual opportunities for performance improvement in biasing the output stage correctly, improving the phase inverter stage for a more balanced output that is maintained in spite of tube changes, and improving the NFB/HF Stability circuits for a flatter response and greater stability. But on top of the usual 4Ω output challenge, it is also challenged with only 255 vdc of B+ and an OPT little bigger than that on an 8600. The initial modification offered for the 8800 did a lot to resolve those issues, with the addition of full Bias and AC/DC Balance controls added icing to the cake, and all of these changes collectively making the unit very capable for many folks -- and indeed it is. But moving seriously forward from this point in the Magnavox world always involves replacing the OPT.

The first effort towards that end returned a mixed bag of results, with the Raphaelite OPTs used in that effort returning very good performance on the 8Ω tap (albeit at 9.5 watts maximum power), and even offering well placed UL taps as well for that mode of operation. But performance on the 4Ω tap was poor, with the resulting recommendation to think carefully before considering the purchase of these transformers as they are not inexpensive, and with the poor 4Ω performance produced, it was hard to justify their cost even from the most inexpensive vendor. These transformers were also tried in UL configuration, with the resulting power loss this produced added to the power loss due to the higher than expected primary impedance measured (more on that later). Besides the power loss, performance at 20 kHz also suffered notably in UL mode with these transformers. With no other known (or deemed worthy) candidates for the project then besides the modern manufactured Z-565 4/8 transformers, a pair were ordered to keep the project moving. But with the possibilities these superb transformers provided, it took the project down a couple of interesting rabbit holes, starting with determining the most accurate way of measuring the primary impedance of an OPT.

Some years ago, I contributed to a thread regarding what I had measured the primary impedance of a particular transformer to be. A well respected AKer also contributed his findings to the thread -- but his results were somewhat different from mine (slightly higher), stating that the difference was due to the fact that during his measurements, he terminated the secondary with a resistive load equal to the rated impedance of the winding. That approach seems logical, as that process would measure the transformer under what more represents real use conditions, and therefore generate a more accurate reflected primary impedance. As will be shown later, it became important to know with accuracy what the primary impedance was with the transformers of this project, and therefore what the correct measurement method was, so that accurate performance versus load impedance assessments could be made. In an effort to nail this down, I took two well known vintage transformer manufacturing companies who published the impedance of their transformers (Acrosound and Dynaco), and set about measuring the primary impedance of their transformers using both a loaded and unloaded secondary winding. In all cases, exactly 100.0 vac @ 60 Hz was applied to the full primary winding, with the full secondary voltage (loaded or unloaded as necessary for the test) measured to determine the turns ratio and resulting primary impedance. The results I got were as follows:

1. Acrosound TO-330: Manufacturer Specified: Early on 3300Ω, then later, 3800Ω. Measured: 4018Ω unloaded. Loaded = 4673Ω. Loaded to unloaded ratio = 116.3%

2. Acrosound TO-300: Manufacturer Specified: 6600Ω. Measured: 7003Ω unloaded. Loaded = 8775Ω. Loaded to unloaded ratio = 125.3%

3. Dynaco A-431: Manufacturer Specified: 4300Ω. Measured: 4328Ω unloaded. Loaded = 4856Ω. Loaded to unloaded ratio = 112.2%

4. Dynaco A-470: Manufacturer Specified: 4300Ω. Measured: 4328Ω unloaded. Loaded = 5157Ω. Loaded to unloaded ratio = 119.2%

5. Dynaco Z-565: Manufacturer Specified: Never specified. Third party (Dynakit Parts) specifies Z-565 4/8 as 8000Ω. Measured (4/8 version): 7668Ω unloaded. Loaded = 9780Ω. Loaded to unloaded ratio = 127.5%

6. RaphaeliteOP10K15AB: Manufacturer Specified: 10000Ω. Measured: 8714Ω unloaded. Loaded - 11654Ω (retested here to present unloaded test results for comparison). Loaded to unloaded ratio = 133.7%

In my experience, Acrosound transformers have always measured about 6% high compared to their published data (using the unloaded test), so other factors may have been considered when they determined their published value. The measured data for the Raphaelite transformer (as given in a prior post) was originally determined using the loaded test method, while unloaded testing indicates the transformer to measure notably below specified impedance, producing the greatest loaded to unloaded ratio of the bunch.

So the measured to specified results are different between the two vintage manufacturers (and likely others as well), but the Dynaco transformers produce very close conformity when tested without a secondary load. In either case however, the unloaded approach gives the closest conformity to the published data from both manufacturers, so that approach is shown to be most accurate. Unloaded testing is supported by theory as well, as any electrical procedure to determine the turns ratio of a given transformer must minimize any losses for greatest accuracy, and also be conducted at a frequency that eliminates any capacitive coupling effects within the core. The results suggest that when a load is applied, it causes a re-characterization of what would normally be viewed as transformer losses, turning them into (a false indication of) an increased turns ratio, and resulting higher reflected primary impedance. The effect is no less real. But intermingling transformer losses with turns ratio data clearly distorts the turns ratio data generated. This is further supported by the loaded to unloaded results produced by the various transformers used in the test. Specifically, it is apparent that physically smaller transformers show a larger discrepancy between loaded and unloaded test results, as would be expected. Smaller transformers typically have lower power ratings and higher impedance ratios, with greater DC resistance in their windings, all which increase losses. To generate the most accurate information then, only the unloaded testing approach can give consistent, accurate results with any size transformer, without the variable influences of loading. When that is understood, it can be seen that what was really going on with the Raphaelite transformer then is actually greater than normal transformer losses, rather than a higher measured primary impedance versus that specified by the manufacturer, as originally thought. But why is all of this so important to this project?

The answer to that question begins with the fact that it is specifically these two transformer companies -- and David Hafler in particular who was a principle engineer at both of them -- that developed the recommended Ultra-Linear and Tapped Screen Grid operating conditions for 6V6 tubes -- for the products from both companies -- that the entire audio community then took its queue from. Therefore, any independent testing (when required) must produce results that are consistent with the manufacturer's own processes to characterize their products and recommended operating conditions, for the successful outcome of any project. That brings us to the Z-565 transformer.

Testing indicates then that the Z-565 transformer actually has a 7.6KΩ primary -- not the 8KΩ reported by Dynakit parts. It's close, I'll grant you. But not as close as Dynaco's published specifications are historically to the measured value of their transformers. To my knowledge, Dynaco never offered the Z-565 transformer for sale as a separate piece (it was never part of their catalogs), meaning there is no official manufacturers information published on the transformer. It has always been widely assumed that the Z-565 was simply the un-potted guts of the A-410 -- the separate Dynaco offering for tapped screen grid operation of 6V6 and EL84 tubes. The A-410 -- being part of their catalog -- is specified as being an 8KΩ transformer, which is likely where the assumption came from then that the Z-565 transformer is also an 8KΩ transformer as well. However, I strongly believe they are two separate transformers.

Since Hafler was a driving force at both Acrosound and Dynaco, he was almost certainly quite involved with the development of the Acro TO-310 transformer while there, which was released before he left, and was specifically developed for UL operation of 6V6 and EL84 tubes. Between his article "Ultra-Linear Operation of 6V6 Tubes" (Radio and Television News, June, 1954) and the Acrosound catalogs, the TO-310 has published specifications indicating it is an 8KΩ transformer (by the catalogs), with screen taps placed at 24% (by the article) -- although remember that Acro transformers measure about 6% higher than their published specifications indicate. While I don't have any measured data on the tap placement of an A-410 transformer, the A-410 and TO-310 are often seen as interchangeable, and both are certainly considered as playing in the same ball park where the Z-565 lives, whose taps are located at 25.0%. But again, why is all of this so important?

With the acquisition of Z-565 transformers, one hopeful outcome of this project was (again), to operate the output tubes in UL mode. That was tried with the Raphaelite transformer, but the results were less than satisfactory as previously mention. But with new, bonafide, Dynaco transformers on hand, of the type that should be acceptable for UL operation of 6V6 tubes, it gave new hope on that front -- except, that's not what happened. The project was already taking twists and turns I had not expected, and this was yet another one to unravel.

Next time.

Dave
 
Could the differential between loaded and unloaded impedance readings also be interpreted to indicate transformer efficiency? Looks like they're around 80% efficient on average, which sounds about right.
 
Nice cliff hanger there Dave. Anxiously waiting the next post because I have it on the plan this summer to build my daughter a 6V6 push pull amp and I was going to use the clone Z-565's.....

===

As an aside, I've measured Hammond 1650H's using this same method (100V input, loaded and unloaded secondary). This particular Hammond measured very close to 6.6K (the published spec) unloaded, when secondary was configured for 8 ohms, and measured high when 8 ohm secondary was loaded with an 8 ohm real load (but I don't remember how much higher it was). So I've also resorted to the unloaded approach as being the more accurate even if Hammond was my only data point.
 
Gadget -- I think it certainly can be taken as a good indication in comparing one transformer to the another, and even a reasonable approximation as well. But because it is a low level test, it does not account for all the losses that occur at full power output. Years ago, I had an experimental "100 watt" 6550 amplifier that employed an Acrosound TO-340 OPT. In my work with that unit, I measured the primary AC voltage at full power output, with my notes showing the AC voltage there being about 8% greater than what the (unloaded) measured turns ratio of the transformer suggested should be there based on the secondary voltage being delivered into the dummy load. Or looked at the other way, about 92.5% of the actual primary voltage appeared in the load connected to the secondary, based on the (unloaded) measured turns ratio. From that, I have always used a figure of 92.5% as a transformer voltage efficiency rating when predicting RMS power output capability, or a power efficiency rating of 85.6% based on the same voltage efficiency figure. Against the data given for the TO-330 transformer above (same 100 watt transformer class from Acro as the TO-340), the correlation of power efficiency based on the TO-340 tests of years ago to the losses suggested by the impedance measurements of the TO-330 today is in fact very good (within 2%)!

Dave
 
The loaded impedance is pretty close to the unloaded impedance (as measured by winding ratio) PLUS twice the DC resistance of the primary (secondary loss is similar in magnitude to primary). I measured an A-410 as 9400 Ohms, Z565 (original Dyna and Dynakit Parts) as 8300, and an Acro TO-310 as 7850 Ohms, all loaded with 8 Ohms. UL ratio of all were about 28% (ratio of loaded impedance), efficiency 87-88%. The Acro was best at high frequencies, only 10 degrees at 20 KHz. A-410 was second, at 16 degrees, Z565s good at 20-21 degrees. The Z-565s had the best low end, at least at the 10V level tested.

<edit> why did I say minus?
 
Last edited:
Thanks for weighing in, Tom. Your knowledge is always appreciated and respected. A notable difference between your tests and mine are that I always determine the turns ratio using the full secondary winding, as that configuration must reflect the original design intent of the transformer. The 8Ω tap does not always represent exactly 8Ω based on the rated primary impedance but a close approximation, due to constraints of the coil form's construction in placing the tap at exactly 8Ω. Your approach certainly has merit however, since the 8Ω tap is the most widely used tap by far today. From that perspective then, it is helpful to know what the reflected primary impedance is based on typical use, in addition to that of the overall design intention of the transformer. For the record, the transformers I measured came from Triode Electronics, whose characteristics have shown to be for all intents and purposes identical to the original Dynaco product. Thanks also for providing the additional detail on your loaded impedance tests. May I assume that the UL screen tap percent you cited was in fact one of turns ratio rather than impedance?

Dave
 
Last edited:
It's the square root of the measured impedance ratio, which would be the winding ratio if the transformer were lossless. Winding resistance has some effect here too.
 
A Real Dragon Slayer..... Part II: Ultra-Linear Operation of 6V6 Tubes

As previously noted, both Acrosound (first) and Dynaco (second) released transformers targeting push-pull UL operation with these tubes, but scant evidence exists on the complete performance picture actually produced from their use. From back in the day, there are precious few references where these transformers are shown to be used in conjunction with specifically 6V6 tubes. One is an Acrosound transformer catalog with a suggested circuit included, one is the previously mentioned Hafler article in Radio and Television magazine (which basically details the presentation of the Acrosound catalog circuit), and another is a an article by Robert M. Voss in the August 1959 edition of Audio -- High Fidelity magazine entitled "Designing a Low-Distortion 12-Watt Amplifier". Also, in an early Dynaco transformer catalog, there is an A-410 design (based specifically on EL84 tubes), with sub-note indicating that 6V6 tubes can be substituted with only a change in the output stage cathode bias resistor. Finally, in 1953, Stan White commercially released his two "Powertron" amplifiers, the smaller of which used 6V6 tubes.

The Voss article is based on the TO-310 and amounts to little more than a list of desired requirements he presents for the design of his amplifier, yet offers not a shred of evidence they were actually achieved other than the subjective comments of "sounds clean and can't hear any noise". The Hafler article (written while he was still at Acrosound and also based on the TO-310) is certainly much more complete as would be expected, with square wave depictions shown, comments provided about the specific Frequency Response, Power Response, and IM Distortion achieved, and more general comments offered about the transient performance and input sensitivity level that will be achieved based on the amount of NFB used. An Acrosound transformer catalog offers a bit more information on this amplifier, presenting a graph of IMD and Frequency and Power Response Curves produced in a circuit with only minor variations from the article circuit, while the Dynaco catalog circuit capable of use with 6V6 tubes is presented standalone style, with no supporting comments or performance results, other than a general comment in the catalog introduction that the circuit is capable of 12 watts at less than 1% IMD. The White Powertron circuit apparently uses the TO-310 transformer as well, but like Voss, offers no performance results, as his article on his amplifiers is primarily about the benefits of including negative current feedback in his designs.

The Hafler influenced designs presented in his article and both transformer catalogs use floating paraphase inverters of either 6SL7 or 12AX7 design, while the Voss design uses ST-35 type input circuity, also using a 12AX7. The Powertron circuit is not given, but is said to be very similar in design to the circuit of its bigger 20 watt brother (whose circuit is provided in White's article entitled "The White Powertron Amplifier", that appeared in the November 1953 issue of Audio Magazine), which uses the TO-300 transformer, cathode bias, and a cross-coupled phase inverter design. In all cases, the basic output stage design is identical, other than the use of one or the other manufacturer's OPTs, and choice of output tube cathode bias resistor, which was 250Ω in all known cases except for 300Ω as used in the Hafler article, as the details of his discussion there were executed in the conversion of an existing commercial product (Grommes model 100BA) to UL operation, and he elected to keep the cathode resistor value of the original design. Finally, a few AKers have converted Magnavox 6V6 amplifiers to UL operation with a transformer change, but none have used the Z-565 transformer (to my knowledge), and in any event, have only provided subjective results for analysis.

So there you have it -- all the information I could find detailing the use of 6V6 tubes in UL operation. And while Hafler does by far the best job of detailing his performance results, one topic is nearly void from all of these sources: that of Total Harmonic Distortion performance, other than general comments by Hafler in his article that the "distortion characteristics (of the TO-310 transformer) complement those of the "Ultra-Linear" circuit and permit low distortion at both high and low levels from 20 cps to over 20 kc". But no supporting data is given to put that comment into context (what was considered low distortion in 1954? And what constitutes low and high levels?). Later in the same article, he states that a "clean waveform is preserved from 20 cps to 30 kc" at an implied 10 watt level. But again, what is "clean" -- no clipping (likely)?

The long standing Typical Operation data given for the 6V6 by RCA for conditions of maximum power output in Class AB1 operation have the plate and screen grid elements operating at 285 vdc into an 8000Ω load, using fixed bias operation to yield 14 (plate) watts of power output, or ~ 12 watts RMS into a real world load. Since the Design Center rating system of the day placed the maximum grid #2 (screen grid) voltage for the 6V6 at 285 volts, this scenario has the tubes producing maximum AB1 power output, since operating the plate and screen voltages at the highest allowable equal levels with fixed bias operation results in requiring the lowest possible plate load, which produces conditions of maximum power output. But as stated, this is based on fixed bias operation. Cathode bias can and most often is used, with the loss of voltage across the bias resistor made up for with a similar increase in the output of the B+ supply. But this power supply compensation is not perfect, as the bias resistor introduces a current limiting element into the equation, causing the load line to shift to a position of requiring a somewhat higher load impedance to produce maximum (although somewhat lower) un-distorted power output in this scenario. This results in the classic 10KΩ plate load used and so often seen in the classic highest quality 6V6 pentode designs that use cathode bias, and deliver about 10 Watts RMS power output; and also why Acrosound recommends use of their 10KΩ TO-270 OPT when pentode operation of their cathode biased 6V6 UL offering is desired. This is also in fact the basis of the modified 8800 design offered earlier, whose output stage retains use of the original Magnavox OPT, and whose performance (within the limitations of the OPT) is excellent.

Now all else being equal, UL operation of any output stage that otherwise operates the output tubes with equal plate and screen grid voltages in pentode mode necessarily raises the required plate load offered to the tubes over that required by straight pentode operation, to maintain a condition of maximum (although somewhat lower) un-distorted power output. That's because at conditions of maximum negative going plate voltage swing in each tube, there is also a maximum negative going AC signal applied at the screen grid, reducing the effective screen grid voltage, and therefore the amount of current the tube can pass, which then raises the required plate load for optimum performance. Also, as described above, if the design started with fixed bias, then using cathode bias will raise the loading requirements and reduce the power output delivered even further if optimum performance is to be maintained with Class AB1 operation. So if 6V6 tubes with equal plate and screen grid voltages deliver maximum power and excellent performance with pentode operation into an 8KΩ load using fixed bias, and a 10KΩ load using cathode bias -- then what happens when the tubes are operated in UL mode (using the recommended screen tap placement) with an 8KΩ load -- as both Acrosound and Dynaco recommends -- using cathode bias as their circuits show???? Or even with fixed bias???

We'll see, next time.

Dave
 
In your review of the past literature, you seemed to have missed the classic articles by Langford-Smith & Chesterman, in particular, their article on the 6V6-GT UL operation, which appeared in Radiotronics, June 1955. Here is a quote from the article wrt to the load impedance:

The load resistance of 8000 ohms plate-to-plate was selected as optimum, giving an output of 10.4 watts at 0.72% THD, even though an output of 11.2 watts was obtainable with a load resistance of 10,000 ohms. The reason for the choice is partly to make it less sensitive to increases in load resistance such as always occurs with a loudspeaker load, and partly to make the transformer simpler and with fewer primary turns.
 
Last edited:
Absolutely correct -- I did omit that excellent source, so thank-you for reminding me of it! It is without a doubt the most exhaustive technical work likely done on the subject. There are some practical issues of real world application that it omits (testing at various frequencies across the audio spectrum, testing based at target quiescent current rather than bogey tube grid bias voltage, the effects that real world OPTs produce at the extremes of the audio spectrum, etc.) that tend to distort some of its findings, but it still remains likely the most authoritative source of information on the topic. Thanks for chiming in!

Dave
 
Glad to be of some assistance, as usual, Langford-Smith was more interested in the theoretical aspect of the UL operation, but his calculations and graphs still provided a good overview of the relative performance of the UL connection vis-a-vis the bias voltage, primary load, and UL tap. I'm looking forward to your investigation into the UL operating conditions and performance with real world transformers as well as cathode bias.
 
Last edited:
Going Down A UL Rabbit Hole......... (and this was supposed to be such a quick and easy project.....)

So the Z-565 4/8 transformers have been installed, and the details of their electrical fitting finalized first for pentode operation (as I wanted to have that mode of operation to compare to), with the conversion to UL operation being easy to complete from there. Testing was conducted from a B+ supply of 303 vdc (ref: ground) with regulation better than 1.5%, at the onset of clipping, using the 8Ω tap for loading and measuring. In all cases, AC Balance was adjusted for optimum performance. The data of this exercise is presented rather broadly so as to present a bigger picture, rather than get mired down in the minutia of a specific detail.

1. Cathode Biased UL Operation w/Hafler Recommended 250Ω Bias Resistor: Power Response: 7.5 watts RMS, 20 Hz - 20 kHz (maximum un-distorted power output).
THD at 7.5 watts: @20 Hz = 2.15%, @1 kHz = 0.78%, @20 KHZ = 1.35%.
IMD (60 Hz & 7 kHz mixed 4:1) @ 7.1 watts equivalent RMS power output = 2.0%.

COMMENT: The IMD figure is double what Hafler suggests is possible, but this design only uses 12.0 db of NFB (4 X reduction, to keep the design in the same realm as the original Magnavox design), whereas Hafler's article design uses a FB factor of nearly double this value, suggesting that his claim of less than 1% IMD is possible. While the Hafler design uses 340 vdc B+, his claim of 11 watts of equivalent rms power output can only be considered as that which is available at the output tube plates, rather than that actually delivered into the load.

COMMENT: Using Langford-Smith's recommended 220Ω Bias Resistor, power response was basically unchanged:
THD at 7.5 watts: @20 Hz = 1.60%, @1 kHz = 0.65%, @20 KHZ = 1.25%.
IMD @ 7.1 watts equiv. = 1.3%.

COMMENT: With no other change than reducing the bias resistor to 220Ω, all distortions fell, with IM getting close to Hafler's claim while still using (basically) 12 db NFB.

2. Fixed Bias UL Operation:
w/Bias adjusted for maximum power output:
Power Response: 8.2 watts RMS, 20 Hz - 20 kHz (maximum un-distorted power output).
THD at 8.2 watts: @20 Hz = 1.10%, @1 kHz = 0.33%, @20 KHZ = 0.80%.
IMD at 8.5 watts equiv. = 1.1%.

w/Bias adjusted for lowest distortion: Power Response: 7.5 watts RMS, 20 Hz - 20 kHz (maximum un-distorted power output).
THD at 7.5 watts: @20Hz = 1.05%, @1kHz = .15%, @20 KHZ = 1.20%.
IMD at 7.5 watts equiv. = .52%.

COMMENT: Maximum power output resulted from a quiescent bias of 38 mA cathode current per tube. Minimum distortion resulted from a quiescent current of 46 mA per tube.

COMMENT: As with my original work with EFB and the SCA-35 amplifier, converting the output stage to fixed bias operation reduced all distortions, even below that obtained with Langford-Smith's 220Ω cathode resistor.

But then, there was also this little problem to chase down:
SAM_2457.JPG

Troubling as well was the fact that whether cathode or fixed bias, the maximum low distortion power output was 7.5 watts, which goes against what theory would dictate. I chose to tackle the uneven clipping issue first, with the thought that both issues were likely related.

Next time.

Dave
 
Last edited:
Continuing On:

The clipping occurs whether operating with cathode or fixed bias; the bottom half of a sine wave (1 kHz in this case) always clips early when the UL connection is made. This shot was taken as the top of the sine wave approaches the onset of clipping. However, a quick check with the pentode output stage connection shows this behavior does not occur in that mode, with clipping extremely well balanced in presentation when driven to clipping levels.

Back in UL mode, the NFB loop was opened so that a study could be made without that corrective element in place. The clipping remained. Further study revealed that the cause of the uneven clipping was due to one of the very circuit elements initially used to enhance the performance of the unit in the first place: convert the original Paraphase Inverter design into a Floating Paraphase design. It was found that when you bring:

1. A UL output stage,

2. Near full power clipping, and

3. A traditional Floating Paraphase inverter design that directly drives the output stage

..... All together in the same design, then early clipping on the bottom half of the waveform will result. You take out any of the three elements listed, and the clipping evens out on the tops and bottoms of the waveform. The problem was ultimately traced to the fact that a Floating Paraphase inverter does not present an equal drive impedance to the grids of the output tubes. More specifically, the impedance between the control grid of the lower output tube and ground is much lower than it is for that element in the upper tube -- this because of the NFB employed around the inverter section of the driver stage that turns the inverter into a floating type design. In pentode mode, that is not a problem. But when the UL connection is made, the lower impedance appearing at the control grid of the bottom output tube makes the screen grid more sensitive to the UL FB signal applied there, preventing that tube from clipping. In the upper tube however, the screen grid is not as sensitive to the UL signal, causing that tube to clip early relative to the bottom tube, and producing the display above.

This is perfectly normal behavior for pentode tubes: A pentode always produce more gain to a signal applied to its control grid when the impedance of the circuit powering the screen grid is made very low -- this because the low impedance prevents what would otherwise be a NFB signal from being injected into the grid were the impedance to remain high. This is the purpose of the screen bypass cap used in many small signal tube pentode stages. The scenario here is exactly the same, except that the roles are reversed. In this case, it is a signal being injected into the screen grid, with the impedance present at the control grid affecting how sensitive the screen grid is to a signal applied to that element.

What was determined then, is that optimum UL operation requires the drive impedance presented to the control grids to be equal, and for maximum effectiveness, as low as possible. As a quick check to see if this theory was correct then, please refer to the schematic provided for the modified amplifier when retaining use of the stock OPT. By attaching a 10 uF 450 volt cap between the bottom of the AC Balance control (where the 470K/3.3M resistors attach) and ground, the circuit immediately reverts to that of a standard Paraphase inverter design, with the AC Balance control still performing its duty quite nicely. This now creates nearly identical, albeit rather high impedance drive circuits to the output tubes, that could still be adjusted for a balanced drive. With the AC Balance control properly readjusted, that simple change resulted in this display at clipping:
SAM_2456.JPG

Even clipping! So for UL operation, a Floating Paraphase inverter that directly drives the output stage is a problematic way to go. But what about the Split Load (Cathodyne) inverter? It too presents unequal drive impedance from its outputs. Since Dynaco used this type of inverter in all of it's commercial designs, and had it directly drive the output stage, shouldn't it create the same problem as well? Hafler/Laurent's solution to this problem was to always use the lowest practical plate and cathode load resistor values possible for the inverter stage, so that the effective impedance presented to both output tube control grids is low enough to prevent such behavior. The result is even clipping. But there's more.

From the data above, note that the low distortion quiescent current for UL operation came in at 46 mA per tube. But when the inverter was modified back to standard Paraphase operation, the optimum low distortion operating point was found to be 38 mA per tube, or almost exactly where the optimum pentode operating point was found to be (37 mA per tube). So the unequal drive impedance of the Floating Paraphase inverter not only upsets the dynamic balance of the output stage (at all power levels), but also upsets the optimum operating point as well. A couple of quick IM distortion tests were then run to see if they would supply further confirmation as to what was playing out. With the inverter modified to standard Paraphase operation, and the AC Balance control set for optimum, the following data was gathered:

1. With cathode bias, 8.2 watts equiv. power output was produced at 0.75% IMD (using 250Ω cathode resistor).

2. With fixed bias, 9.25 watts equiv. power output was produced at 0.50% IMD (38 mA quiescent current per tube).

So more power at less distortion with lower quiescent current was developed with both forms of bias when using the lowly Paraphase inverter circuit, which further validate the loss of performance when a UL output stage is directly teamed up with a Floating Paraphase design. This surely cannot bode well for the UL 6V6/6BQ5 circuits that both Acrosound and Dynaco show in their respective transformer catalogs, nor the design in Hafler's article, as all of these are of the type of design described here. No doubt however that the act of replacing an original pentode based OPT in a suitable commercial product with a premium UL transformer (as Hafler does in his article) would still produce significant performance gains.

But permanently reverting back to a standard Paraphase design in this scenario is not without its problems either: The adjustment of the AC balance control is quite sensitive in that mode, with even slight changes from optimum causing distortion to rise quite rapidly. Not everybody has a distortion test set handy to set such a control accurately. The topology of the ST-35 would seem to be the simplest and most effective approach then to directly driving small power tubes in UL from a phase inverter stage, without the problems noted previously. As I'm sure most will agree, such elegant solutions are a hallmark of Dynaco products.

For this project however, changing the phase inverter circuit to accommodate UL operation might be the last straw for some folks, as it would no longer represent anything resembling a Magnavox amplifier, but simply an ST-35 built into a Magnavox chassis, using 6V6 tubes. Fair criticism. But also the best solution for those committed to quality UL operation in the Magnavox chassis. Doubling back to pentode operation then, the following data was gathered:

1. Cathode Biased Pentode Operation with 250Ω Bias Resistor:
THD @20 Hz = 3.5% at 9.68 watts. @1 kHz = 1.1% at 10.125 watts. @20 KHZ = 1.65% at 8.20 watts.
IMD @ 10.34 watts equiv. power output = 2.0%

Cathode Biased Pentode Operation with 220Ω Bias Resistor:
THD @20 Hz = 2.8% at 9.9 watts. @1 kHz = 0.55% at 10.45 watts. @20 KHZ = 1.10% at 8.82 watts.
IMD @ 10.60 watts equiv. power output = 1.10%

2. Fixed Bias Operation:
THD @20 Hz = 3.0% at 10.60 watts. @1 kHz = 0.175% at 11.88 watts. @20 KHZ = 0.61% at 11.28 watts.
IMD @ 11.47 watts equiv. power output = 0.58%

COMMENT: From a purely distortion standpoint, the big advantage to UL operation is THD produced on the low end at 20 Hz -- although, when pentode power output is reduced to equal that of its UL counter parts, the difference diminishes. And, at 25 Hz, pentode distortion is just under 1.0% at 11.52 watts, so distortion drops quickly as frequency rises above 20 Hz. With a power response of 10 watts RMS from 20 Hz to 20 kHz, and all distortion under 1% from 25 Hz to 20 kHz, pentode performance is hardly shabby, with many preferring the lower damping that connection provides that endeared them to the Magnavox sound in the first place. And, maintaining pentode operation of course helps to preserve some of the original Magnavox design as well.

Of course, there are other aspects of the amplifier that are not discussed here, yet deserve a quick mention:

1. Frequency Response: +/- 0 db from 20 Hz to 20 kHz, within in -0.5 db to 40 kHz.
2. Stability: Will tolerate any value of capacitive only output loading without bursting into oscillation.
3. 10 kHz square wave presentation:
SAM_2448.JPG

4. And the performance as presented at the 4Ω tap:
SAM_2449.JPG
It is performance like this that (in part) make these transformers far and away the best product available in their power and impedance class, whether for pentode or UL service. They are not inexpensive, but unlike the Raphaelite transformers, they are absolutely worth the cost of admission for the performance they provide. The performance figures quoted throughout this post are identical whether viewed at the 4Ω or 8Ω tap.

With that then, the amplifier has been all buttoned up (for now) in pentode mode, operating with fixed bias from a small bias supply installed earlier. As such, for a quality 6V6 amplifier of conventional design -- other than operating with fixed bias and only 12 db of NFB -- its performance is quite impressive, allowing it to rise considerably from it's basement dwelling location of Magnavox designs. A schematic will be presented in due course. For now then, those who are absolutely committed to operating these transformers in UL, the best recommendation would be to simply use the ST-35 inverter and FB/stability circuits, and operate the output stage with fixed bias. The results will be excellent.

Finally, for those interested, the following information was gathered with a 16Ω load presented to the 8Ω tap:

THD @20 Hz = 7.9 watts at 5.9%. @25 Hz = 8.41 watts at 2.5%. @1 kHz = 8.41 watts at .32%. @20 KHZ = 9.0 watts at 0.93%.
IMD @ 8.50 watts equiv. power output = 1.0%.

A 10 kHz square wave presentation remains largely unchanged with a 16Ω load, as do response and stability characteristics:
SAM_2459.JPG

All in all, not too shabby. A few final pics are presented:

Top side: The chassis controls now sport knobs.
SAM_2447.JPG

Bottom Side: You can see the small bias supply added in the upper right corner of the chassis, powered by an appropriate transformer stolen from a discarded wall wart. This of course necessitated changing out all the Bias and DC Balance Controls for them to work properly with a fixed bias design:
SAM_2436.JPG

And finally, a close up of the bias supply:
SAM_2437.JPG

For now, that brings this Magnavox saga to an end. Time to enjoy some more listening!

Dave
 
Last edited:
Good question -- but I think the assumption is a pretty safe one. The math approach was primarily aimed at ensuring total primary winding consumption was not significantly different than the application the transformer was originally designed for (and it's not). But as to the HV secondary winding's capability, I think the answer to that is best supported by by the fact that HV regulation at the output of the rectifiers -- even with the output stages operating with fixed bias -- is still on the order of ~ 3.5% from quiescent conditions, to full power output in both channels -- which is actually quite remarkable. The quiescent B+ at the output of the rectifier is also a product of 1.25 times the AC voltage applied to each side of the full wave circuit as well. If the HV winding were not up to the task at hand, the resistance of a less capable winding would not allow the performance noted to be produced. Still, it's a point that needed more clarifying, so thanks for chiming in!

Dave
 
Back
Top Bottom