I received this radio about 10 years ago from someone for free. They were getting rid of it. It's large, and not cheaply made (HEAVY). The wood is thick. It's quite beautiful, but also quite beat up. I'm the only person I know who sees the potential beauty of this radio.

I received the radio without its power supply and without its '50 tubes. Probably some shmuck pulled the tubes and sold them on eBay or something, since they're quite expensive. (Personally I'd like to see people hang from a tree who rip tubes from radios just to make a buck.....well, not literally, but they make me very angry.) A long time ago I built a power supply with a transformer that was capable of 250 volts. The radio worked alright using type '71 tubes in the output.

Well, just a few days ago I rebuilt my home made power supply so that it'd give about 310 volts, which is right for a model 90 set that uses '45 tubes in the output (the supply will later be built in a nice metal chassis, but is on an oak board for now). A simple omission of one of the resistors on the field coil circuit makes the rest of the radio work at proper voltages. The set is now equipped with type '45 tubes in the output. Someday, when I have money, they'll even be globes, like the rest of the tubes. Type G-27 and G-80 tubes are a lot cheaper than type G-45 tubes.

This radio is impressive! It picks up stations as well as any superheterodyne! It is VERY sensitive. It isn't the best as far as selectivity is concerned, but it's almost as good as a superheterodyne. I'd say that it's about the same as a Philco 60, which is a superheterodyne, but lacks a tuned final IF coil, which is why it isn't as selective.

The Majestic also has superb tonal quality and power. I'm amazed with this radio. The speaker is huge and heavy. The actual cone is only about 10 inches in diameter, but it's very deep. It's very strong and suspended well. It was originally suspended with rubberized cloth. It hardened. I replaced it with suede, since this is what I had at hand at the time. Suede works very well. The field coil is massive. The amplifier had a nasty harshness which was coming from one of the two 45 tubes. It'd get worse when one was pulled, and it'd go away when the other was pulled (1st one re-inserted). The offending '45 was thankfully the one that's out of phase with the 1st audio, so I was able to put my negative feedback circuit between it and the 1st audio (which works surprizingly well with transformer circuits, even though I originally designed it for capacitor coupled amplifiers). Now the audio is all that could be desired. It has rich, deep bass, and fine treble. This radio, amazingly, is probably the best sounding radio I own. It is surprising. Majestic advertised it as having colorful tone, and it certainly does (although it's also been helped along quite a bit by me).

I have a few complaints about the radio, however. The radio lacks AVC, and also has an antenna coil fine tuning knob. Both the antenna fine tuning and the volume control must be constantly adjusted as I go from station to station (the antenna control can be left alone and then tweaked later, but the volume control must be manipulated when going from strong stations to weak stations, and vice versa). This makes the radio a little harder to use than a superheterodyne. The AVC problem can be taken care of easily enough, though I like the lack of AVC for local stations, because the music has more dynamics. Things become more annoying when listening to distant stations that drift in and out. The antenna trimmer is also necessary if excellent sensitivity is to be had all over the dial, since TRF receivers are notorious for lack of perfect synchronization (who can expect it when you have 4 linked mechanical tuning devices?).

My final complaint is that the type of detector that's used distorts at low volumes. The music sounds sort-of raspy and fuzzy when the volume is turned low. I've read that this type of circuit always does this, though. Amazingly none of the condensers in the radio are faulty. The big ones are all in metal cans, and I'm going to leave them alone for now. They work perfectly. The micas are all fine. Also, none of the resistors have drifted one bit. They're all perfect. Because of this, I am certain that the detector only distorts because that's simply how it works. I placed a 3.3 Meg resistor from the detector grid to B-, and this cleared some of the distortion. It also improved the fidelity greatly. Changing the value of the grid leak made no improvement. I suppose I could bypass the grid leak and condenser, and make the stage into a normal RF amplifier. I could then have a diode detector hidden in one of the cans.

All in all, for a 1929 radio, it's a fine one. I use it for hours, and it's great to listen to when there's good music to listen to. It'll be beautiful when the cabinet is refinished. It uses many different fine veneers.

Thomas

Thomas, until recently, I had a Majestic Model 50. Similar in many ways to your set, but it was a superhet.

It was actually a little too selective - the treble would get chopped off with the outer portion of the sidebands. To get decent treble, I had to detune slightly. I had been meaning to stagger-tune the IF xfmrs, but never got around to it.
Doug

Doug/Thomas,
The 2 TRF's I've done Edison 7R and G/G Majestic 90B did not perform as well as the more recent superhets I've done. Even with 3 stages of tuned RF amplification, sensitivity/selectivity left much to be desired. I personally feel that the answer lies in tuned bandwidth issues and AVC, and would be interested if anyone has used a sweep gen. to observe the BW of one of these old sets, compared to a superhet.

marv

:Thomas, until recently, I had a Majestic Model 50. Similar in many ways to your set, but it was a superhet.
:
:It was actually a little too selective - the treble would get chopped off with the outer portion of the sidebands. To get decent treble, I had to detune slightly. I had been meaning to stagger-tune the IF xfmrs, but never got around to it.
:Doug

Hi, Marv. I've got way more TRFs than I should. Quite a few of them sound quite good, with decent selectivity. The lack of AGC is an issue that has to be worked around.

I've got (or had) a couple of early superhets that are too darned selective. The Majestic 50 that I mentioned is one.

But TRFs can be too selective also. I have a TRF Stewart-Warner 950 that fits that description. Here is what appeared in a March 1930 commentary in "Radio" on that set:

"...fidelity is sacrificed somewhat for selectivity....quite a bit of the attenuation at the high frequencies may be blamed upon the cutting of the side bands due to extreme selectivity."

My TRF Stewart-Warner 950 has bass galore. I ought to get in there and slightly de-synchronize the four RF tuning caps.

This set has three RF stages (four RF tuning caps on a single shaft), detector, 1st audio, 2nd audio, and two 45s in push-pull for the final. Plenty of power!
Doug

:Doug/Thomas,
:The 2 TRF's I've done Edison 7R and G/G Majestic 90B did not perform as well as the more recent superhets I've done. Even with 3 stages of tuned RF amplification, sensitivity/selectivity left much to be desired. I personally feel that the answer lies in tuned bandwidth issues and AVC, and would be interested if anyone has used a sweep gen. to observe the BW of one of these old sets, compared to a superhet.
:
:marv
:
::Thomas, until recently, I had a Majestic Model 50. Similar in many ways to your set, but it was a superhet.
::
::It was actually a little too selective - the treble would get chopped off with the outer portion of the sidebands. To get decent treble, I had to detune slightly. I had been meaning to stagger-tune the IF xfmrs, but never got around to it.
::Doug

Doug,
Yes, my quick calculations indicate that wider BW or stagger tuning would favor higher fidelity with a sacrifice in selectivity, whereas peaking (narrow BW)would favor the bass, and improve selectivity. I once knew a rule of thumb about the number of sidebands captured in the IF/Tuned circuits required for good fidelity/volume, and it has since disappeared into a maze of gray matter, but I'm sure many old/new timers can still recite it. In relation to my 2 TRFs, I don't recall any tuning mechanism other than the caps across the multi-gang tuning condenser. The inductors were fixed and inside an aluminum cover, so stagger tuning would have to be accomplished by de-tuning at the ganged condenser.

marv

:Hi, Marv. I've got way more TRFs than I should. Quite a few of them sound quite good, with decent selectivity. The lack of AGC is an issue that has to be worked around.
:
:I've got (or had) a couple of early superhets that are too darned selective. The Majestic 50 that I mentioned is one.
:
:But TRFs can be too selective also. I have a TRF Stewart-Warner 950 that fits that description. Here is what appeared in a March 1930 commentary in "Radio" on that set:
:
:"...fidelity is sacrificed somewhat for selectivity....quite a bit of the attenuation at the high frequencies may be blamed upon the cutting of the side bands due to extreme selectivity."
:
:My TRF Stewart-Warner 950 has bass galore. I ought to get in there and slightly de-synchronize the four RF tuning caps.
:
:This set has three RF stages (four RF tuning caps on a single shaft), detector, 1st audio, 2nd audio, and two 45s in push-pull for the final. Plenty of power!
:Doug
:
::Doug/Thomas,
::The 2 TRF's I've done Edison 7R and G/G Majestic 90B did not perform as well as the more recent superhets I've done. Even with 3 stages of tuned RF amplification, sensitivity/selectivity left much to be desired. I personally feel that the answer lies in tuned bandwidth issues and AVC, and would be interested if anyone has used a sweep gen. to observe the BW of one of these old sets, compared to a superhet.
::
::marv
::
:::Thomas, until recently, I had a Majestic Model 50. Similar in many ways to your set, but it was a superhet.
:::
:::It was actually a little too selective - the treble would get chopped off with the outer portion of the sidebands. To get decent treble, I had to detune slightly. I had been meaning to stagger-tune the IF xfmrs, but never got around to it.
:::Doug

Well, this particular TRF has less than normal superhet selectivity, though not as poor as a two stage TRF. It's about comparable to a Philco 60 as far as selectivity goes, though perhaps it's a bit better (or worse...I'm not sure). It's really good at singling out stations, though it's not as good as some of the best pin-point superheterodynes I own. If you used it, though, you'd be amazed. Other than having to adjust the antenna and volume controls, it behaves very much like a superheterodyne. I doubt that most people could tell much difference. Stations sometimes blend together, but it really is darn good.

Sensitivity on this set is incredible. It also uses triodes everywhere. Fidelity is great, and the thing pulls in stations from all over the nation. Of course the volume control must be turned up for the weak ones. The AVC in a superheterodyne would do this automatically. It really is an incredible radio.

Perhaps I will convert the set so that it has AVC. The detector works better when it is driven harder, so keeping the RF tubes running high would help the audio (instead of turning them way down in order to turn down the volume). It distorts at low volumes. I'm also contemplating with adjusting the grid leak capacitance, as this might stabilize the bias voltage. Bias for the grid leak detector, if you didn't already know, is developed from the incoming signal, as well as the electrons that pile up on the grid from electron flow. The capacitor has to be just the right value so that it'll retain proper bias for various voltage fluctuations. Bass notes tend to fuzz the audio and make it sound kind of garbled at low volumes.

Anyway, if I go the AVC route, I'll omit the volume control and just leave the 600 ohm resistor (which is the same as turning the volume all the way up). I'll have a small solid state diode hidden somewhere, and develop AVC like the Crosley 1117 develops it. A separate diode is used, and the signal is taken prior to the detector with a small value condenser. There isn't really a way to get AVC from a grid leak detector, as far as I know, so I'll have to use the Crosley 1117 method. Perhaps I'll have a switch hidden under the set that'll allow me to switch from a small value AVC filter condenser to a large value, so that I can control the rate of AVC. I kind of like how the radio, as it is now, doesn't AVC the audio at all. AVC tends to make the audio sound like a wall of music, with no dynamics. A lack of AVC with distant stations, however, as well as going from station to station, is an annoyance.

Perhaps this TRF is where I got the notion that TRFs are so good. This one sure is. I've seen some two stage sets that leave a lot to be desired, though. I guess the technology has the potential to be good, if you don't mind having a boat anchor for AM. The cabinet is beautiful, though, and I'll settle for the boat anchor chassis inside. The globe tubes and the 27 screen plates make for interesting viewing anyway. Speaking of screen plates, I've noticed that they are much more susceptable to interference (AC interference, such as holding your hand near the detector tube, which causes hum) than the later style solid plate 27 tubes. I like the globe look, though, so I may switch the detector G-27 for a G-56, which has the solid plate. The set hums terribly with a G-27, and there's no provision for a shield can over the detector tube. The three tubes prior to the detector have shields.

T.

Thomas,
My G/G Majestic doesn't have a shield over the detector nor did the Edison 7R, but the 7R had a shield base, and I was able to find one that fit perfectly at PTOP.

marv

:Well, this particular TRF has less than normal superhet selectivity, though not as poor as a two stage TRF. It's about comparable to a Philco 60 as far as selectivity goes, though perhaps it's a bit better (or worse...I'm not sure). It's really good at singling out stations, though it's not as good as some of the best pin-point superheterodynes I own. If you used it, though, you'd be amazed. Other than having to adjust the antenna and volume controls, it behaves very much like a superheterodyne. I doubt that most people could tell much difference. Stations sometimes blend together, but it really is darn good.
:
:Sensitivity on this set is incredible. It also uses triodes everywhere. Fidelity is great, and the thing pulls in stations from all over the nation. Of course the volume control must be turned up for the weak ones. The AVC in a superheterodyne would do this automatically. It really is an incredible radio.
:
:Perhaps I will convert the set so that it has AVC. The detector works better when it is driven harder, so keeping the RF tubes running high would help the audio (instead of turning them way down in order to turn down the volume). It distorts at low volumes. I'm also contemplating with adjusting the grid leak capacitance, as this might stabilize the bias voltage. Bias for the grid leak detector, if you didn't already know, is developed from the incoming signal, as well as the electrons that pile up on the grid from electron flow. The capacitor has to be just the right value so that it'll retain proper bias for various voltage fluctuations. Bass notes tend to fuzz the audio and make it sound kind of garbled at low volumes.
:
:Anyway, if I go the AVC route, I'll omit the volume control and just leave the 600 ohm resistor (which is the same as turning the volume all the way up). I'll have a small solid state diode hidden somewhere, and develop AVC like the Crosley 1117 develops it. A separate diode is used, and the signal is taken prior to the detector with a small value condenser. There isn't really a way to get AVC from a grid leak detector, as far as I know, so I'll have to use the Crosley 1117 method. Perhaps I'll have a switch hidden under the set that'll allow me to switch from a small value AVC filter condenser to a large value, so that I can control the rate of AVC. I kind of like how the radio, as it is now, doesn't AVC the audio at all. AVC tends to make the audio sound like a wall of music, with no dynamics. A lack of AVC with distant stations, however, as well as going from station to station, is an annoyance.
:
:Perhaps this TRF is where I got the notion that TRFs are so good. This one sure is. I've seen some two stage sets that leave a lot to be desired, though. I guess the technology has the potential to be good, if you don't mind having a boat anchor for AM. The cabinet is beautiful, though, and I'll settle for the boat anchor chassis inside. The globe tubes and the 27 screen plates make for interesting viewing anyway. Speaking of screen plates, I've noticed that they are much more susceptable to interference (AC interference, such as holding your hand near the detector tube, which causes hum) than the later style solid plate 27 tubes. I like the globe look, though, so I may switch the detector G-27 for a G-56, which has the solid plate. The set hums terribly with a G-27, and there's no provision for a shield can over the detector tube. The three tubes prior to the detector have shields.
:
:T.

Yeah. A shield would be nice for the detector, though I really like being able to see the tube. I'll try a G-56 first, and see if it helps. It should, since a newer 27 doesn't hum much at all. I'll also try to see if I can balance things under the chassis so that there aren't any stray hum fields. The set will hum even if I'm nowhere near it, so something isn't balanced. If I touch the chassis and hold my hand near the detector, the hum goes away, so something's out of balance. Another thing I can do is turn on the phonograph motor, which cuts the hum. Unplugging the phonograph motor, however, does not reduce the hum. Hmmmmm..... where is it coming from? I'll have to see.

T.

:Thomas,
:My G/G Majestic doesn't have a shield over the detector nor did the Edison 7R, but the 7R had a shield base, and I was able to find one that fit perfectly at PTOP.
:
:marv
:
::Well, this particular TRF has less than normal superhet selectivity, though not as poor as a two stage TRF. It's about comparable to a Philco 60 as far as selectivity goes, though perhaps it's a bit better (or worse...I'm not sure). It's really good at singling out stations, though it's not as good as some of the best pin-point superheterodynes I own. If you used it, though, you'd be amazed. Other than having to adjust the antenna and volume controls, it behaves very much like a superheterodyne. I doubt that most people could tell much difference. Stations sometimes blend together, but it really is darn good.
::
::Sensitivity on this set is incredible. It also uses triodes everywhere. Fidelity is great, and the thing pulls in stations from all over the nation. Of course the volume control must be turned up for the weak ones. The AVC in a superheterodyne would do this automatically. It really is an incredible radio.
::
::Perhaps I will convert the set so that it has AVC. The detector works better when it is driven harder, so keeping the RF tubes running high would help the audio (instead of turning them way down in order to turn down the volume). It distorts at low volumes. I'm also contemplating with adjusting the grid leak capacitance, as this might stabilize the bias voltage. Bias for the grid leak detector, if you didn't already know, is developed from the incoming signal, as well as the electrons that pile up on the grid from electron flow. The capacitor has to be just the right value so that it'll retain proper bias for various voltage fluctuations. Bass notes tend to fuzz the audio and make it sound kind of garbled at low volumes.
::
::Anyway, if I go the AVC route, I'll omit the volume control and just leave the 600 ohm resistor (which is the same as turning the volume all the way up). I'll have a small solid state diode hidden somewhere, and develop AVC like the Crosley 1117 develops it. A separate diode is used, and the signal is taken prior to the detector with a small value condenser. There isn't really a way to get AVC from a grid leak detector, as far as I know, so I'll have to use the Crosley 1117 method. Perhaps I'll have a switch hidden under the set that'll allow me to switch from a small value AVC filter condenser to a large value, so that I can control the rate of AVC. I kind of like how the radio, as it is now, doesn't AVC the audio at all. AVC tends to make the audio sound like a wall of music, with no dynamics. A lack of AVC with distant stations, however, as well as going from station to station, is an annoyance.
::
::Perhaps this TRF is where I got the notion that TRFs are so good. This one sure is. I've seen some two stage sets that leave a lot to be desired, though. I guess the technology has the potential to be good, if you don't mind having a boat anchor for AM. The cabinet is beautiful, though, and I'll settle for the boat anchor chassis inside. The globe tubes and the 27 screen plates make for interesting viewing anyway. Speaking of screen plates, I've noticed that they are much more susceptable to interference (AC interference, such as holding your hand near the detector tube, which causes hum) than the later style solid plate 27 tubes. I like the globe look, though, so I may switch the detector G-27 for a G-56, which has the solid plate. The set hums terribly with a G-27, and there's no provision for a shield can over the detector tube. The three tubes prior to the detector have shields.
::
::T.

Interesting. Actually, with a proper alignment job, this thing works just as well as a typical superheterodyne. I forgot that the last time I aligned the set I didn't own a signal generator. It's REALLY good at separating stations. Unfortunately, with the alignment, some of the audio quality has been lost (treble). It's still very good, though. Now if I can only get the distortion out of the audio at low volumes, and figure out how to neutralize the circuits. It doesn't howl at full volume, so I guess I should be satisfied. I just want to know how to make it work as well as it can. It did squak at full volume, but I adjusted one of the neutralizing trimmers, and the squaking went away.

T.

And here's how:

http://www.philcoradio.com/tech/neutral.htm