Magnavox Imperial model 1ST233

[electra225]

Dennis, thanks for filling in the blanks. I didn't understand what effect the hum balance control had on filaments. I appreciate your explanation.

Bill, those are filter caps. Those are the Nichicon PZ "skinny" caps I like so well. There are three in the original can, plus those. One is for cathode bypass. The absolutely only downside to those caps is they have to be purchased from Mouser. I despise their website. I'll get brave one of these days and order some more. I read somewhere that if you "divide" B+ in the output tubes from the rest of the amp, something good will happen, I forget now what that even was. All I know is I have a dead-silent power supply, with nary a hint of hum. Of course, I also have nary a hint of bass, but we are in the process of finding out just why.......

And, Bill, while we are at it, does your amp have the two .1uf caps at the right top corner of the chassis? The schematic for your amp shows them. Those two caps are a mystery to me. Why did Magnavox use tiny value "coupling caps" for cathode bypass duty in first audio in the bass audio amps?

Walter, I used PB BLaster on the metal of the amp. The tuner hasn't been cleaned yet. I also put a good coat on the Multiplex adapter. PB Blaster is the only stuff I have run across that keeps that white "cad scum" from building up on the metal chassis. That stuff will migrate to tube sockets and create a poor contact. I suspect it may also migrate to controls, since they seem to get dirty relatively frequently. This shows the "better than junk" tube complement, pending biasing and final test.

Another "unique" feature of the amp is that Magnavox used a common "split load" or "cathodyne" phase inverter, half of each 12AT7 amplifier tube. This type of phase inverter affords no amplification. Typically, one finds another amplifier stage after this type of phase inverter. Why didn't Magnavox use another stage of amplification in this amp?

The plan for this afternoon, while Western Red is doing his thing (on the Concert Grand!) I'll run a resistance check on the amp chassis to make sure I haven't created a short. I haven't done anything to the tuner yet besides take the eye tube socket apart. I'll isolate that so I can power up. If all is well in the amp, I'll connect speakers and power up. I'm curious to see if anything good happened with what I have already done.

[electra225]

I powered up this afternoon and listened to part of Western Red. It was on AM, so there is that. I have 15" speakers for bass and horns for treble that came out of a solid state Stereo Theater. It doesn't sound bad if the speakers are laying face down on the cabinet, way more bass. If the cones are facing up, the sound is nothing to write home about. The treble controls on the amp are really dirty. The balance control has an open spot in it. I read where many people just bypass the balance control with a 390 ohm resistor to each channel then run to ground. I may try that if I can't get this one to work right. Is the bass response better? I just can't tell at this point.

I ran some biasing measurements. I can tell you that the tubes I put in aren't even close to being matched. V-13 has 14.38 watts plate dissipation, V-14 has 11.05 watts, V-18 has 11.7 watts, V-19 has 7.66 Watts plate dissipation. Plate voltages range from 342.5 VDC to a high of 345 VDC, so that is pretty even. The difference is in how hard the tube is conducting. V-13 & V-14 are one bass channel, V-18 & V-19 are the other bass channel. According to these readings, biasing is pretty close to max, fairly hot. Most guitar amp guys want plate dissipation to be a max of 12 watts with each tube being within a couple of watts on a P-P pair of tubes. JJ says you can run their tubes to 14 watts plate dissipation, but I worry about the output transformers running with bias that hot. I ran this for about an hour. I got 202 watts from the line at `122.1 VAC line voltage. The power transformer was nice and cool after that time, laying on the bench. The wattage I reported is without the Multiplex adapter in the works.

What do I know so far? I need to finish the resistor replacement. I don't have any shorts nor did I make any booboos. I sorely need a new set of matched output tubes and also a set of matched 12AT7's. I will measure bias readings again after I get a new set of tubes in it. I need to continue with the eye tube repair. After I get the amp sorted, on to the tuner. I need to clean and make sure of the condition of the balance control and both treble controls. Then I'll match the balance and treble channels with a VTVM on a speaker in each channel. Then I'll be ready to tinker and try to find some bass response if I nothing I have done makes any bass.

[electra225]

I ran the calculations on the 12AX7, first audio tube in the treble amps. It is biased very close in crossover frequency to the bass amps. 100% of the signal starts passing at about 1500hz. This finding may explain why it sounds like the singers are "down the hall from the mic" when music is played. If these calculations are right, it appears that neither the bass amps nor the treble amps are amplifying the midrange and bass frequencies at 100%. Both the treble and the bass amps start passing the frequencies 100% at about 1500hz. I'm waiting for the resistor kit to arrive, then I'll finish the resistors, take some voltage measurements, then figure out where we are from there.

[electra225]

I have the resistors, so I'll finish those up in the amp, take my resistance measurements, then if all is well, take voltage measurements.

[electra225]

I have one more resistor to replace in the amp, R88, an 8200 ohm, 1-watt power resistor. I have all the others done. There are only five resistors in the amp that were not changed. I ran the resistances again, all is well, all within 2% of spec. That is pretty tight. I got interrupted, so my next step is checking voltages. IF all is well, we go into the tuner.....

[William]

Boy, you are really making progress and before you know it you will have one great sounding Imperial. Good thoughts go your way that the rest of this adventure will go well.

Bill

[William]

Boy, you are really making progress and before you know it you will have one great sounding Imperial. Good thoughts go your way that the rest of this adventure will go well. Keep up the good work, Greg.

Bill

[electra225]

I may have run up against a stump that I can't move......

I'm not going to make prognostications for every 81-series amp Magnavox built, but this one, the one I'm looking at will not benefit in bass response if I had a semi load of matched 6V6's. Magnavox designed this amplifier to produce wimpy bass. Perhaps they had complaints so made changes between when mine was built in the middle of 1960 and when Bill's was built, about a year and a half later. The mono bi-amps didn't have weak bass, but they also ran 6L6's. The bean counters were well at work in my Imperial. Magnavox bought and used the cheapest garbage components they could get by with. The tubes they used built by Sylvania are still running in the chassis. The RCA and GE sourced ones are long gone. I am beginning to wonder if the power transformer might hold the key here. They changed it, made it beefier in Bill's amp. Why couldn't they have done that in mine and then used 6L6's? Bean counters......

Let me see if I can explain this where it makes sense. Every component between the "crossover transformers" and the grid of a first audio tube, in both the treble and the bass amps, is essentially a low-pass filter or a high-pass filter. The treble controls on the amp are not volume controls, but they are part of an adjustable high-pass filter. Varying that control bleeds more or less of the midrange frequencies to ground. There is a component shown in the drawing in dashed lines, meaning it's optional, in the input to the bass amp grid. It is a 100K resistor. Resistance passes bass, capacitance passes treble. I need to study that a bit more. I also want to compare the bias of a 93-series amp and maybe an AMP150 to see how they are biased. This amplifier PASSES bass frequencies, it just doesn't amplify them a lot. Since Magnavox has such a fetish on treble, why did they bother with 15" cone woofers at all? Why not just hang another horn on the end of the cabinet and be done with it. I'm wondering now if I need to figure out how to change the frequencies in the low-pass filter for the bass amps. Then change the bias on the first audio amps.

The differences between Bill's amp and mine are few. The p ower transformer and sealing the bass speaker enclosures are the two largest changes. I'll continue with my restoration of the "basics" project. That has to be right in any case. There sure doesn't seem to be much information on the Internet about this subject. I can't find anything where someone has dug this deeply into one of these amps.

[electra225]

I have checked every voltage on every element in the amp chassis. They are right on, down to 1.38% low, up to 1.62% high. That is pretty close. I don't expect a lot of bass response, using speakers that are just laying on top of the bench. I think the speakers I'm using were out of an Astro-Sonic stereo that had closed speaker enclosures, a bass reflex type setup, so they won't sound that good on the bench. I do get a lot of deflection in the speaker cones, for whatever that is worth. Now on to the tuner. And I still need to replace the 8200 ohm, 1-watt power resistor yet. I don't have one. It has drifted up to over 12K ohms. I don't think it makes that much difference, since my voltages are okay, but I'm doing the rest to 5%. I can say I cured the "singers down the hall from the mic" illness. The audio sounds nice that way. I may have blown one of the horns I'm using for test. It cuts in and out.....

I may have gotten a little carried away with volume on my Norma Jean CD.

[TC Chris]

You'll get lots of cone excursion with bare speakers because the front and back waves aren't separated. Speaker baffles prevent front & back from canceling each other and let the speaker move air in the room.

Chris Campbell

[electra225]

I'm thinking about replacing, temporarily, the speakers in the Imperial cabinet with the pair I am using on the bench. Just because, I have no particular reason why. I connected the speakers in the Imperial cabinet with my Marantz 2245. They sounded okay, but that Marantz would make anything sound good. The test was inconclusive. The AK guys think replacing the output transformers is the bomb. I don't know if I set much store by that. Looks to me like if the transformer is matched to the output tubes and to the speaker voice coil why is one not as good as another? The speakers in the Imperial give me the impression they are lazy. Does that make any sense?

[William]

Speakers need to be mounted, kind of what Chris said, to sound right. Please don't change out the speakers in the Imperial until you have finished this project and tested it as Magnavox built it. Hear how it sounds, like it just came from the factory new. If bass is weak, then plug the holes in the bottom of the cabinet and listen again. That way you will know what changes that Magnavox made between yours and mine made the difference.

Bill

[electra225]

I've already plugged the holes in the speaker enclosures. I'm not done yet. Let's see how this plays out going forward.....

[TC Chris]

The AK guys think replacing the output transformers is the bomb. I don't know if I set much store by that. Looks to me like if the transformer is matched to the output tubes and to the speaker voice coil why is one not as good as another?

Transformers have frequency limitations. Remember IF transformers that are designed to pass the IF and reject other frequencies (OK, as part of n L-C network). You will note that component hi-fi amps typically have larger OPTs, and not just for higher power. Design & size affect LF response. It would be interesting to hear how much difference new OPTs made in amps where they were replaced.

Chris Campbell

[electra225]

They claim they get 8 watts out of an 86-series single-ended 6BQ5 amp. Remember, though, these guys gut consoles then get silly with the amps. They even figure the current draw on the pilot light to use in the output of the amp. I can get 25-watt, P-P, 8000 ohm primary output transformers, Hammond, from AES for about $70. Why would Magnavox use output transformers that are not suited for the stereo? If I were going to change transformers, I'd change the power transformer. The main difference between my IMperial and Bill's IMperial is the power transformer. His has approximately 10% higher B+ throughout than mine does. His output transformers are the same as mine, as far as I know.

[electra225]

The issue I'm experiencing this morning is FM wonkiness. I admit to using my "longwire" AM antenna for an FM antenna, so there is that. I also don't have the eye tube in its socket, since the eye tube is dead. It's in the AVC line, so that may be making a difference. I can't get the FM tuner on station where it stays. If I touch the chassis, it works better, so I'm thinking the antenna may be the issue. I've never done a detailed FM alignment in my life. I may be getting ahead of myself....

[electra225]

The following observation may not be well-received. I have been working with the speakers in my Imperial. I have a set of $50, 6-inch woofer bookshelf speakers from Parts Express that sound better than do the 15's in the Imperial cabinet. Connected to the 45-watt Marantz receiver I have in my laboratory, it seems they take a lot of power to make the speaker cones move. If I put a lot of power thru them, the output seems garbled to me. I can buy a pair of 98-decibel sensitivity speakers on the 'Net for less than $100 apiece. They have 3" voice coils and 60-ounce magnets. Something in the back of my mind tells me I need a pair of high-efficiency speakers before I make this thing sound like it should. I am well aware of the "matched components" rule in console stereos. If the tubes, resistors, and capacitors run their course and have to be replaced, why are speakers exempt from this wear and tear of old age?

[electra225]

I just saved $33! I found I may not need an eye tube. The combination of the 470K resistor in the eye tube socket and a 1.5 MEG ohm resistor being drifted caused the eye tube to be dead. Now I get good deflection and the eye is brighter. 100%? No, but not dead. I can certainly get more life out of the eye tube. Do I have more bass? I still can't tell. I have the speakers connected via a huge kluge so am having trouble keeping all the speakers working. I'm not going to remove all the speaker wiring from the cabinet so I can set it up on the bench. I have all the audio done as far as components are concerned. All I need now is a set of output tubes, then recheck bias, then set the treble channels. Is the tuner reception better? I need to get some high-value resistors for the AVC line yet. I will say that the FM tuner stays on channel better. I don't know if the eye tube being wonky had something to do with that, but it is in the circuit for the AVC, so it may have.

[William]

Progress is being made, and the end looks near as far as having all the components replaced. As far as the speakers go, please keep in mind that mine sounds great with the original speakers. I know, I keep harping on this, but it's true. Now I'm wondering if your woofers and my woofers have the same numbers on them.

Bill

[electra225]

This thing still sounds really terrible on the bench, running open speakers. It's got lots of volume, just no bass. I'm trying to decide whether I go further, or just give up, put the guts back into the cabinet, then use it for a table so the cat can see out the window. I've done all my meager ability will let me do. I have had my hands on every component in this thing, nothing has made a dime's worth of difference. I don't know whether matched output tubes will make any difference or not. Wrong if I do, still wrong if I don't....

I still believe there is too much high-frequency signal getting to the bass speakers. I have no idea how to fix that. Nothing I have tried has worked, nothing I have read about and tried has worked. This thing is determined to be a 300-pound clock radio.....