OT, the bigger the better? And which one to choose?

[Richard]

Whatever they are selling is what makes the difference. If they're selling resistors the resistor makes the difference. The Groove tubes guy says the tubes make the difference. Everybody here knows it's really the capacitors that make it happen.

Gibson Historic Bumblebee Capacitors 2-pack | Musician's Friend

[Gregg]

What exactly is so horrible about it?

I wouldn't call it horrible but from this thread you can learn more details how it was made:

http://ampgarage.com/forum/viewtopic...3b7316b6f0f342

Let me put it this way - if you dive into the OT field and start winding your own you'll never do it this way. Obviously it doesn't mean it won't sound good although "good" is entirely subjective thing and is matter of taste. On the other side if you make an OT according to all rules of the craft that won't guarantee perfect (whatever that could mean) sound either.
Another good example of a seemingly (or theoretically) bad design is Marshall C2668 which is maybe the second simplest way to wind an OT but it's a legend among Marshall amp fans.

On a Transformer with two separate primary windings which can be connected in series or parallel (lets say, as found on a mains transformer designed for dual voltages), does the primary inductance differ when connected in series vs connected in parallel? Here is a drawing.

If you have two primaries on the same iron and both measure for example 1 Henry then if you connect them in series you'll get around 4H, if in parallel the inductance will be roughly the same.

[bob p]

Whatever they are selling is what makes the difference. If they're selling resistors the resistor makes the difference. The Groove tubes guy says the tubes make the difference. Everybody here knows it's really the capacitors that make it happen.

Gibson Historic Bumblebee Capacitors 2-pack | Musician's Friend

[ATTACH=CONFIG]40040[/ATTACH]

Sprague Atoms too!



"There's a sucker born every minute."

[Mike Sulzer]

Whatever they are selling is what makes the difference. If they're selling resistors the resistor makes the difference. The Groove tubes guy says the tubes make the difference. Everybody here knows it's really the capacitors that make it happen.

Gibson Historic Bumblebee Capacitors 2-pack | Musician's Friend

[ATTACH=CONFIG]40040[/ATTACH]

You sure they don't cryo treat the chassis after all punching and drilling?

[SoulFetish]

If you have two primaries on the same iron and both measure for example 1 Henry then if you connect them in series you'll get around 4H, if in series the inductance will be roughly the same.

.....now you're just messing with me.

[Gregg]

if in series the inductance will be roughly the same.

I'm sorry. Another typo. Should read "in parallel". It's fixed now.

[Mike Sulzer]

I wouldn't call it horrible but from this thread you can learn more details how it was made:

http://ampgarage.com/forum/viewtopic...3b7316b6f0f342

Let me put it this way - if you dive into the OT field and start winding your own you'll never do it this way. Obviously it doesn't mean it won't sound good although "good" is entirely subjective thing and is matter of taste. On the other side if you make an OT according to all rules of the craft that won't guarantee perfect (whatever that could mean) sound either.
Another good example of a seemingly (or theoretically) bad design is Marshall C2668 which is maybe the second simplest way to wind an OT but it's a legend among Marshall amp fans.



If you have two primaries on the same iron and both measure for example 1 Henry then if you connect them in series you'll get around 4H, if in parallel the inductance will be roughly the same.

My browser wants me to stay away from your link.

However, let's demystify this. Audio transformer design is well understood. For example, there is a design manual here:http://www.dissident-audio.com/Trans...ign_Manual.pdf. As you know, one of the key factors in designing and constructing an audio transformer for 20 Hz to 20 KHz is proper interleave design and careful winding. This gets you about a factor of three over a simple winding technique. So you do not need to do this for guitar. So what would be called a poorly executed design and build in the stereo world would be expected to be no problem for guitar.

There could be factors on the low notes, however. Perhaps getting just the "right" amount of harmonic distortion on the low notes requires the use of just the right right core, that is, much too small for stereo. Oh, wait, that seems to contradict the original idea of this discussion, but never mind, when dealing with factors that affect the sound, it is what you want that counts.

[Gregg]

So what would be called a poorly executed design and build in the stereo world would be expected to be no problem for guitar.

That's right. Almost all those guitar amps that made history intentionally or not had "poorly" or cheaply made OTs (and other parts as well) but they became guitar tone "standards". Some consider the whole thing a matter of accident.

[Mike Sulzer]

That's right. Almost all those guitar amps that made history intentionally or not had "poorly" or cheaply made OTs (and other parts as well) but they became guitar tone "standards". Some consider the whole thing a matter of accident.

I agree that that accident does play a significant role in desired guitar tone, but I am saying something different about the high frequency response. It is not needed because the speaker falls fast at 5 MHz or so. Also the speaker is inductive in the multiple KHz range. The transformer leakage inductance just adds a bit to that.

[nickb]

That's right. Almost all those guitar amps that made history intentionally or not had "poorly" or cheaply made OTs (and other parts as well) but they became guitar tone "standards". Some consider the whole thing a matter of accident.

I couldn't read the ampage site either due to malware.

What bothers me about all this is the speaker does't do much above 5KHz so leakage inductance isn't an issue. At the low end, the amps often roll off quickly below 100Hz so the primary inductance is not critical either. Therefore, I would suggest that the speaker and the amp have a greater impact than the transformer in general.

Edit: Bother, I just read that Mike said more or less the same thing. I should have read to the end.

[Gregg]

My browser wants me to stay away from your link.

Mine as well but I ignore it and my internet security package doesn't detect anything.

What bothers me about all this is the speaker does't do much above 5KHz so leakage inductance isn't an issue. At the low end, the amps often roll off quickly below 100Hz so the primary inductance is not critical either.

This is true but there is this other school of thought that says it's better to have a full range OT. Some amps have such wider range transformers.
So maybe some amps would benefit from a Hi-Fi OT (or were designed around one) but some wouldn't and are not going to sound "that" way. Again this is mostly a matter of taste. Neither one is "good" or "bad".

[Mike Sulzer]

Mine as well but I ignore it and my internet security package doesn't detect anything.



This is true but there is this other school of thought that says it's better to have a full range OT. Some amps have such wider range transformers.
So maybe some amps would benefit from a Hi-Fi OT (or were designed around one) but some wouldn't and are not going to sound "that" way. Again this is mostly a matter of taste. Neither one is "good" or "bad".

Have you looked at the frequency response of a guitar speaker? The extra high frequencies from a full range OT do not matter. I do not think this is a matter of sound.

As for a design requiring a full range OT, this seems very unlikely. There is one possible issue if you use feedback around the transformer: it is at least a remote possibility that the extra phase lag resulting from a less than full range transformer might induce instability in some circuits. But I do not think that this matters with the relatively low open loop gain in a system that feedbacks to the PI (the way it is done in guitar amps).

[SoulFetish]

Have you looked at the frequency response of a guitar speaker? The extra high frequencies from a full range OT do not matter. I do not think this is a matter of sound.

As for a design requiring a full range OT, this seems very unlikely. There is one possible issue if you use feedback around the transformer: it is at least a remote possibility that the extra phase lag resulting from a less than full range transformer might induce instability in some circuits. But I do not think that this matters with the relatively low open loop gain in a system that feedbacks to the PI (the way it is done in guitar amps).

I was just going to make this point. To me this is what makes sense, at least conceptually. The plots for speakers I have looked at confirm precisly what you and nick have been saying, along with a large resonant peak somwhere in the 80-96Hz (give or take). The rolloff at 5kHz provide the limiting factors for high frequency band width. To me, this offers a convenient filter, but I worry that designing a transformer with this as the only or ultimate consideration may end up with a design which creates unpredictabe phase shifts in high frequencies in the audible range and unstable conditions in global feedback circuits and lead dress in the chassis.

[Gregg]

Have you looked at the frequency response of a guitar speaker? The extra high frequencies from a full range OT do not matter. I do not think this is a matter of sound.

I know the frequency response of a guitar speaker however still some would insist on a full range OT. Also even the "poorly made" OTs have higher response than that. Actually it's not that difficult in an OT to go above the audio range.

As for a design requiring a full range OT, this seems very unlikely.

Very often amp designers used what they had by hand or what was cheap. Fact is some guitar amps have full range OTs. Whether this was necessary for the sound desired or not is another question.
The speaker has a 5kHz rolloff but if the full range OT will pass the high frequencies "louder" then the rolloff of the speaker in dB could be expected to be less than usual. What is interesting some well known high gain amps have full top range OT's although the opposite could be assumed.

[pdf64]

When comparing the amp's characteristics when suitably loading the 4 ohm output (compared to the 16 ohm output), might poorer coupling increase the magnitude of the back emf spikes resulting from heavy overdrive?