I'm working on a Freed-Eisemann Model NR-12, a battery-powered TRF with four '01A tubes. Schematic: http://www.nostalgiaair.org/PagesByModel/880/M0039880.pdf

I got it playing pretty easily, but ran into something interesting.

This set has two phono jacks for the loudspeaker - one jack labeled LOUD, which ties into the output of the 2nd audio stage, and another jack, labeled MEDIUM, which ties into the output of the 1st audio stage.

The LOUD output gave poor frequency response, mostly bass and very little treble. The MEDIUM output sounded good (but with lower volume, of course). After shuffling tubes around with no change, the 2nd audio interstage xfmr seemed to be the culprit, even though its primary and secondary winding resistances checked out OK.

I replaced the 2nd audio xfmr with a Hammond 124A, which is a usual 1:3 Stancor 53C knock-off. The frequency response improved greatly.

I wonder what goes wrong with such a xfmr to cause it to sound lousy? The bad xfmr itself was a replacement (from many years ago), so maybe it was poor to begin with. Construction-wise, it looked well built with plenty of iron and the ratio was correct.

All's well that ends well, I suppose.

Just a guess on my part: Maybe the lamination insulation (oxide or varnish, or whatever else) has gone bad, allowing shorts between laminations? Maybe the air gap has increased due to shifting (if an air gap is used)?

Also, poor impedance matching makes for poor response.

T.

Some transformers are made poorly to begin with. Check to see if there are many really thin laminations, or not so many, like a power transformer. I would think that this would tend to cut out bass, though, not treble, as lower frequencies require more iron segments to cut down on eddy currents (for instance, a 25 cycle power transformer has more iron than a 60 cycle transformer).

Thomas

Y'know I have an interesing somewhat similar thought.
What makes one speaker so much different sounding than another similar style and size?

I have two almost identical radios. A Bendix 526a and a Bendix 110W

They both use almost identical schematics.
The 110w is a few years newer model.. but still the same basic circuit.
I was playing these two raadios when I noticed how totally great the 110 sounded and the 526 was just average ordinary and plain sounding.
The difference was almost astoundingly richer.. brighter, more full... seemed like I could hear things that weren't even there in the 526 model!!
I went about comparing componetns and actually rewiring both to have the "EXACT" same values and components.. but still the 110 sounded richer... I mean REALLY richer..
I finally took the speaker from the 110 and put it on the 526...and that was it!
It was just the speaker.
They looked similar.. maybe even the same bendix stock number.. but I'm not sure of that.
However... that made ALL the difference in the world. Now I wish I knew what elements in that great little speaker go to make it sooo great? It seems to have the same sized magnet and general construction.. just a 4" speaker... but wow!

If the cones are at all different, that can make all of the difference (not to be redundant with words). A cone that is heavy won't respond well to high frequencies. A cone that is stiff won't respond well at low frequencies. Different ribbing in the cones affects harmonics. Do the voice coils have the same impedances? Are the transformers the same? One transformer might be matched better than the other. The glues which are used for bonding the various parts can make all the difference. Extremely rigid glues with light mass make the best bond between the coil and the cone. Flexible glues should be used around the edge of the cone, though stiff glues are fine too, as long as they don't impede cone movement (sloppy application).

I doubt that either of your speakers has an aluminum voice coil, as this was very rare at that time. About the only manufacturer that I can think of that used aluminum was Stromberg Carlson. Most other radio manufacturers used copper. Aluminum responds a ton better to the high frequencies due to its lighter weight.

Thomas

Thomas:
The ribbing is just a tad different and the "better-sounding" cone seems to have just a little steeper angle.
I left the transformers stay with the radios when I switched speakers so only the speaker was moved... and the "good or better" speaker played exactly as good ..great on the other radio.
So it is definately the speaker itself.

Well, a steeper angle makes the cone stiffer (with respect to the driving coil). If anything can make a speaker sound good, it's when the cone moves as a unit. If waves are set up along the cone's surface, from the inside to the outside, faithful reproduction will not be possible (there'll be harmonic waves instead of the original waves).

The other thing that makes a speaker sound good is a light cone, or a really heavy one. For bass only, a really heavy cone is nice, one that has a resonance at the bottom or below the human hearing range. For a speaker that must respond to all frequencies, a resonance in a pleasant area is nice, or, ideally, a resonance above the human hearing range. Don't expect this with a paper cone, though. Paper cones usually have a resonance right in the mid-range area. The better ones will have this resonance where it sounds good.

T.

Just a couple of thoughts. Maybe you have a short or leakage from the windings to the core. You may have a single short in the windings. The resistance check won't reveal a single shorted turn. If you apply an audio frequency sine wave to the defective xfmr you should see the distortion and the output voltage won't be correct . On a good 1:3 xfmr you should see the waveform voltage increase with very little distortion until, of course, you apply too much signal and saturate the core.

MRO

I would think, too, that a single shorted turn (or a few) would tend to resonate better at high frequencies than at low ones.

T.

Hi

Here is a completely different answer since you mentioned the transformer was replaced before. Both AES and PTOP had defective audio interstage transformers. I made frequency measurements along with others and notified both suppliers.

Here is what happened. These were transformers with a center tapped secondary. Both half's of the secondary were wound together, side by side. This doesn't seem like a problem but it increases capacity between outside wires of the windings. Capacity may only increase to .005 mf? I still have information in the computer if someone is interested. This increase in capacity wouldn't be bad except impedance is 90,000 ohms with only tube grid loading. Even .005 mf will greatly reduce high frequency response.

I am almost sure this is the problem you found.

Norm

:I'm working on a Freed-Eisemann Model NR-12, a battery-powered TRF with four '01A tubes. Schematic: http://www.nostalgiaair.org/PagesByModel/880/M0039880.pdf
:
:I got it playing pretty easily, but ran into something interesting.
:
:This set has two phono jacks for the loudspeaker - one jack labeled LOUD, which ties into the output of the 2nd audio stage, and another jack, labeled MEDIUM, which ties into the output of the 1st audio stage.
:
:The LOUD output gave poor frequency response, mostly bass and very little treble. The MEDIUM output sounded good (but with lower volume, of course). After shuffling tubes around with no change, the 2nd audio interstage xfmr seemed to be the culprit, even though its primary and secondary winding resistances checked out OK.
:
:I replaced the 2nd audio xfmr with a Hammond 124A, which is a usual 1:3 Stancor 53C knock-off. The frequency response improved greatly.
:
:I wonder what goes wrong with such a xfmr to cause it to sound lousy? The bad xfmr itself was a replacement (from many years ago), so maybe it was poor to begin with. Construction-wise, it looked well built with plenty of iron and the ratio was correct.
:
:All's well that ends well, I suppose.

Thanks, Norm. Are we talking about the effective shunt capacitance across, say, the primary terminals of the xfmr? If so, I guess there is no way to really measure that capacitance, at least with my equipment -- we'd have to compute it from the winding geometry.

Just for grins, I measured the capacitance between the primary and secondary windings. For the "bad" xfmr, it's 408pF, and for a new Hammond 124A, it's only 57pF. I'm not sure what to make of that.

:Hi
:
: Here is a completely different answer since you mentioned the transformer was replaced before. Both AES and PTOP had defective audio interstage transformers. I made frequency measurements along with others and notified both suppliers.
:
: Here is what happened. These were transformers with a center tapped secondary. Both half's of the secondary were wound together, side by side. This doesn't seem like a problem but it increases capacity between outside wires of the windings. Capacity may only increase to .005 mf? I still have information in the computer if someone is interested. This increase in capacity wouldn't be bad except impedance is 90,000 ohms with only tube grid loading. Even .005 mf will greatly reduce high frequency response.
:
: I am almost sure this is the problem you found.
:
:Norm

Hi Doug

You showed a difference in capacity. Even a ratio of 7:1 is quite a difference. I used HP 4260A. Windings don't have much of an effect with this HP bridge.

I am sure the way wire was wound is your problem. I worked on this for a long time. Someone actually unwound a transformer to prove the problem.

Norm

:Thanks, Norm. Are we talking about the effective shunt capacitance across, say, the primary terminals of the xfmr? If so, I guess there is no way to really measure that capacitance, at least with my equipment -- we'd have to compute it from the winding geometry.
:
:Just for grins, I measured the capacitance between the primary and secondary windings. For the "bad" xfmr, it's 408pF, and for a new Hammond 124A, it's only 57pF. I'm not sure what to make of that.
:
::Hi
::
:: Here is a completely different answer since you mentioned the transformer was replaced before. Both AES and PTOP had defective audio interstage transformers. I made frequency measurements along with others and notified both suppliers.
::
:: Here is what happened. These were transformers with a center tapped secondary. Both half's of the secondary were wound together, side by side. This doesn't seem like a problem but it increases capacity between outside wires of the windings. Capacity may only increase to .005 mf? I still have information in the computer if someone is interested. This increase in capacity wouldn't be bad except impedance is 90,000 ohms with only tube grid loading. Even .005 mf will greatly reduce high frequency response.
::
:: I am almost sure this is the problem you found.
::
::Norm
: