3rd Gen started

[Stephen1966]

So, it's hard to quantify this. First thing worth mentioning is that the DMM measured about 3V (plus and minus), I put this down to the resistive load of the choke. If I had grounded my negative lead - to actual ground - I expect the voltage would have read around 450V.

Setting the amp in clean only, and the choke in parallel first, there's a clean delineation of notes on the open strings but inverted, the notes appear to sing more readily. This appears more apparent in the bass register. It seems a more harmonically textured palette in the inverted switching. The notes appear to have a little more bloom. it's a very subtle experience though. I can't be sure I'm hearing something, because I want to hear something! When I tried switching the choke when open notes were played and still ringing, I was hard pressed to detect any difference. Still, there is the impression, that it sounds better inverted. Better for me, that is. I'm pretty sure as well that as the current draw rises with harder playing and chords, so too will the impedance and beyond the realm of a weak single open string, it will make more of a difference.

I'm going to swap the choke leads over to the inverted position and live with a while. Give it some juice and see what comes of it. Am I just hearing things that aren't there? That's the Question. It's over to you.

One last safety note. Before touching the box with the amp under power I checked there were no volts running through it. I also checked there were no arcs or volts running through it as I switched it over using insulated grips. it was fine. It was only when I was satisfied that it didn't pose an electrocution risk that I handled the box with my bare hands. It's best to be afraid!

[rootz]

I'm a bit lost for words looking at this build. It is incredible. Unbelievable craftsmanship and tenacity to obtain unobtainium parts.

[ijedouglas]

I'm a bit lost for words looking at this build. It is incredible. Unbelievable craftsmanship and tenacity to obtain unobtainium parts.

Thanks for the kind words, I really appreciate it

[Stephen1966]

I'm trying to understand what difference it could make with the "polarity" of the choke. I mean, everywhere else I see it referenced at all, it says it doesn't matter; that correct polarity is ensured however you wire it up. There is plenty of evidence to suggest that that is more or less the case. Such a small difference, if there is one at all (I'm hedging my bets) is likely so inconsequential to not matter at all... in most cases. However, when you spend that extra $100m to subtract that extra 1/100th of a second per lap from a formula one engine, it isn't the big improvements you are going for.

So. It's a small difference.

Then how could it be...? I have a hypothetical I would like to run by you.

The choke can be modeled as a linear resistance in parallel with an inductance. Bear with me, this is more an intuition than actual education! But, if the resistance is linear then a specific impulse of the AC sine, just one, represented by the peak of just one AC cycle then as the charge enters the choke at one end of a coil of wire, it meets the maximum resistance, and then, as it travels along the coil towards the end of the wire, the resistance gets smaller and smaller. Heat might alter the resistive and inductive properties of the coil but let's ignore that for now. Common sense tells me that the resistance is maximum at the beginning of the wire and minimum at the end, and the resistance in a perfect coil with absolutely consistent thickness and conductivity along its length is linear.

For this hypothesis to work though, I predict that inductance is not linear, but more "S" shaped. If I recall my first impressions '...the notes appear to sing more readily,' could this be because the impedance curve is not just "S" shaped but asymmetrical as well?

This may be mere fancy. I could just be imagining all of this but the coil in a choke is a physical element - I imagine passing from outer to inner windings. But the windings in the choke's core may be generating their inductance more efficiently than those in the outer windings. A product of core permeability which may also be non-linear in operation perhaps. Resistance assumed as linear though could possibly have a different inflection point with inductance depending on which direction you sent the charge - from outer windings to inner, or vice versa.

This is really the impression that struck me: "...the notes appear to sing more readily". Very subjective, for sure, but please tell me I am not going mad!

IMG071.jpg

[bepone]

The choke can be modeled as a linear resistance in parallel with an inductance. Bear with me, this is more an intuition than actual education

R and L in series, C in parallel with them

[bepone]

about the choke "polarity".. you have no polarity on choke, both ends are floating on 450V.. the only thing what differs technically is capacitance from beginning of winding (close to the core) vs capacitance on the end of the winding..
ie via parasitic capacitance this node is somehow shunted to the ground / core is grounded / but this should be some small value and negligible to the audio signal..

[Stephen1966]

The choke can be modeled as a linear resistance in parallel with an inductance. Bear with me, this is more an intuition than actual education

R and L in series, C in parallel with them

Thank you. I appreciate your clarification.

about the choke "polarity".. you have no polarity on choke, both ends are floating on 450V.. the only thing what differs technically is capacitance from beginning of winding (close to the core) vs capacitance on the end of the winding..
ie via parasitic capacitance this node is somehow shunted to the ground / core is grounded / but this should be some small value and negligible to the audio signal.

Could this parasitic capacitance create an effective polarity? Not actual, but effective. Transformer theory is a new venture for me. Please forgive my ignorance.

[bepone]

If we are speaking about output transformer , lets say closest to the choke in questio, means single ended with air gap .. then two transformer wires if are connected to anode and Ub+, and vice versa, there IS a difference in behaviour on frequency characteristic.
So anode must go to beginning of the winding, and Ub+ to the end.

But on DC choke, both ends of winding are shunted to the ground via Electrolytic caps..(C1 and C2 usually) so i cant see what is happening there (if so)

[sergio]

beautiful build!!, congratulations for all the components used, I assume that you have done a great research. very good!
Sergio

[ijedouglas]

about the choke "polarity".. you have no polarity on choke, both ends are floating on 450V.. the only thing what differs technically is capacitance from beginning of winding (close to the core) vs capacitance on the end of the winding..
ie via parasitic capacitance this node is somehow shunted to the ground / core is grounded / but this should be some small value and negligible to the audio signal..

I have to strongly disagree with you on this. The difference in tone while swapping the leads was significant and certainly not "negligible". To make sure I wasn't suffering confirmation bias, I asked my wife to take a quick listen as I swapped polarity without telling her what to listen for. As soon as I flipped the switch she immediately pointed and said "the other way sounds like a blanket on the amp".

I have performed this similar test (without the wife) on all my amps and there is a significant difference between the choke polarity/direction. As I mentioned earlier, I did not discover this, Chris Merren mentioned this in passing and I have a great respect for his knowledge on transformers.

All the chokes tested were vintage Shumacher 022699 and TMI TF-155

[ijedouglas]

beautiful build!!, congratulations for all the components used, I assume that you have done a great research. very good!
Sergio

Thank you

[WhopperPlate]

Uh oh , things happen and people don’t know why … spooky…

[bepone]

I have to strongly disagree with you on this. The difference in tone while swapping the leads was significant and certainly not "negligible".

All the chokes tested were vintage Shumacher 022699 and TMI TF-155

i was elaborating technical aspect, and that i couldnt see any difference in regular choke, didnt directly deny it..your choke is wound on paper and waxed.. test should be performed also on modern plastic bobbin chokes, maybe there is less difference.

[Stephen1966]

I have to strongly disagree with you on this. The difference in tone while swapping the leads was significant and certainly not "negligible".

All the chokes tested were vintage Shumacher 022699 and TMI TF-155

i was elaborating technical aspect, and that i couldnt see any difference in regular choke, didnt directly deny it..your choke is wound on paper and waxed.. test should be performed also on modern plastic bobbin chokes, maybe there is less difference.

You know, I get the feeling we might be looking at this backwards Caps filter out the AC but block the DC and regarding the caps either side, they are not usually equal. That's probably to factored in... If we look at the choke in isolation though, what would we need to do to alter its resonant peak?

Edit: my choke is a Hammond 194B.

[Stephen1966]

I'm posting this link so I can get back to it more readily: https://resources.system-analysis.caden ... n-inductor

I'm thinking the construction of the choke is creating different amounts of parasitic capacitance depending on which way it is wired up to the supply and that this is altering the resonant peak frequency of the choke. I also think, I notice it more when the notes are attacked because of the almost percussive amounts of current that are drawn during those fleeting moments. Less noticeable, insensible even, when switching as a note is already starting to decay. The blanket effect, suggests the presence of lower frequencies and the absence or reduction of such when the choke is inverted. My first impressions led me to prefer the richer tonal character of the blanket effect. But it really only seems noticeable in the attack of the note. Most chokes and inductors seem to be purposed for a more steady-state operation, and in that role the "polarity" seems to be neither here nor there.

Ian, did you notice it more during the attack as well?