Potted OT and PT alignment

[Bub]

Hi Gent's

I'm doing the layout of my new chassis. The PT and OT are about 3" apart. Do I still have to rotate the PT 90 degrees from the OT to reduce magnetic coupling or does the potting eliminate this?

Thanks

Rob

[jojokeo]

There's no set rule that says you simply have to but if you do, you should minize the problem in using "best practices". Not all trannies are the same so one can't give a concrete hard answer. However, I tried to find something I saw a while back but can't reference it at the moment and have to go by memory (not always the best for me as I get older  ).

Hopefully someone's saw what I'm about to describe? You can take your PT and tape off the secondary leads and hook up the primary lead so that you can carefully plug it in and not get yourself shocked. Do this w/out mounting anything yet. Then take your OT and alligator clip to the secondary w/ headphones. You should be able to "hear" the magnetic coupling noise from the PT as you move things around and find the best positions or orientations for them to be mounted at.

Someone please help if you know the link for this?

[rzenc]

The headphone test works fine for me. Find the sweet spot, mark it down and it's good to go.

Just insulate all leads you are not using. A 2" heatshrink will do.

Best Regards,
Rzenc

[Fresh_Start]

Headphone test should turn up on AX84 somewhere - SEARCH their forum if you don't find it in the FAQs section.

Look at know good layouts.  How are their transformers oriented? 

Personally, I think it's good practice to have the PT and OT at 90 degrees from each other.  A lay-down style PT is even better IMHO but not always practical (like if you're making your own chassis).

BTW the headphone trick probably should be combined with common sense.  IOW don't put the OT on top of the input jack with a bell side pointing toward your first preamp tube.

HTH

Chip

[Bub]

Thanks Gentlemen.

Rob

[jojokeo]

Headphone test should turn up on AX84 somewhere - SEARCH their forum if you don't find it in the FAQs section.

Thanks Chip, I found it later on two diff sites but the info was remembered correctly somehow  :shocked:

[Fresh_Start]

AX84.com FAQs:
http://www.ax84.com/faq.html#hp-trick

And yes, jojo's memory is fine.

Cheers,

Chip

[jojokeo]

I have to admit that I've never actually used it. Have you? Is it really that important to do you think? Curious of how sensitive this issue really is or not?

[dynaman1]

Can't help but wonder if placement matters less when using potted trannys.

[HotBluePlates]

Potted transformers are not necessarily any more immune to noise than a regular transformer. In the absence of other information, assume that the potting can is just for looks.

Some cans may have additional mu-metal shielding inside (often true of small mic-to-grid input transformers) that are rated to reduce hum.

For all transformers, assume that you might have to rotate them for minimum hum pickup.

[Fresh_Start]

I have to admit that I've never actually used it. Have you? Is it really that important to do you think? Curious of how sensitive this issue really is or not?

Honestly, I've only built Fender clones - all with Mojo chassis.  Never saw any point in screwing around with the transformer placement on those.

I do think transformer placement is important, but you have to look at the whole picture.  The "headphone trick" reduces PT interference with the OT, but there's a lot of current in the OT.  Makes me wonder about those amps you see with the OT way down at the input end of the chassis.  No PT interference with the OT, but how much OT radiation gets into other sensitive circuitry?

The Fender Blues Jr. is an amp with poor transformer placement AFAIK.  Bill Machrone used to have an essay on his site about a funky, bent steel shield between the PT and the OT which he said greatly reduced the noise factor.

I'm sure that if you put the PT and OT bell-to-bell, you could create a howling monster   Vintage Fenders all seem to have lay-down PTs at the poweramp end of the chassis and the OT at least 2-3 inches away.

My thinking is that you put the PT as far down on the power amp end of the chassis as possible with its open bells either facing up-down or front-to-back to minimize radiation into the rest of the amp.  The OT should be oriented so that it is perpendicular to the PT on at least one plane.  After that, who knows?

Cheers,

Chip

[HotBluePlates]

The OT should be oriented so that it is perpendicular to the PT on at least one plane.  After that, who knows?

The rule is that coupling reduction is proportional to the square of the change of distance.

Plain english: If you double the distance between 2 transformers, you get about 1/4 of the coupling between them. So hum should be reduced by about the same amount. If you move the transformers from touching to 1 inch, there is some amount of hum reduction; moving an additional inch apart (double-distance) gives 4 times the hum reduction. If you move an additional inch apart (3 inches), you have 9 times the hum reduction compared to moving to 1 inch apart.

Remember that transformers on opposite sides of the chassis is a case that is always more about structural issues than hum reduction. That is, the chassis may only be stiff enough to support the weight where the vertical sides stiffen the whole thing. It is also more common on aluminum chasses than steel.

There are 2 types of coupling: electromagnetic and electrostatic.

You might move the OT away from the PT to reduce magnetic coupling. Beyond some point (maybe 3-4 inches or a bit more), hum from this source has been reduced as much as it likely will ever be. Having the OT over preamp circuitry then makes electrostatic coupling more of an issue. Now, high voltage signals at the OT might couple into lower-voltage circuits, and create oscillation.

Point is, don't get carried away solving a perceived problem such that you create a new and different problem.

I haven't tried it yet (no need up to now), but the headphone trick makes the most sense for determining what steps really need to be taken to knock down hum coupled from the PT.

[jojokeo]

On my Marshall-y scratch build below, my tranny locations were sort of dictated by space available, the circuit board, and any signal wiring to be as far away as possible as HBP says about electrostatic coupling. And to eliminate electromagnetic coupling, I purposely offset the OT to be more adjacently set off the PT's corner rather than lined up to the PT's side. It worked like a charm as it's very quiet unless it's fully opened up and even then it's very good compared to many.
This was before I ever heard of the headphone trick and what I meant by "best practices" earlier, trying to be a cautious as possible by using recommended common sense placements as they've been taught. Thanks for your explanations HBP & F_S, they helped to learn even more now. 

[rzenc]

I have to admit that I've never actually used it. Have you? Is it really that important to do you think? Curious of how sensitive this issue really is or not?

I have used this method on many amps, from little 10W to 150W beasts..You can hear the influence of EMF while moving O.T. over chassis. As HBP mention, there is a point where you just can't hear it anymore. Maybe it went down -60dB's?
I noticed that P.T. without shield radiates more EMF, thus O.T. ended up far away from it. Also, I try to align P.T. and O.T. center lines. Helps to reduce hum pick-up when chassis is small.

Bill Machrone used to have an essay on his site about a funky, bent steel shield between the PT and the OT which he said greatly reduced the noise factor.

Once I had to do something alike, however I used 2mm aluminum shield between P.T. and reverb tank to stop hum pick-up. I'm not sure, but it seems that aluminum, has better EMF dispersion then steel.

Hope this helps.
Best Regards,
Rzenc

[HotBluePlates]

I'm not sure, but it seems that aluminum, has better EMF dispersion then steel.

Aluminum is non-magnetic. Therefore, it does not transmit electromagnetic energy the way steel does. But there are situations where you need something that is magnetic in order to have effective shielding.

The headphone trick lets you really hear the difference.

By the way, the common meaning (at least in the U.S.) of "EMF" is Electro-Motive Force. The terms "potential difference," "EMF" and "voltage" are all technically different things, but EMF means voltage more than it means electromagnetic force. But I bet we could find some books where EMF is used for either.

[jojokeo]

Electro -of or pertaining to electrons or to an electron
Motive - something that causes action in a certain way, do a certain thing, etc.; incentive.
Force -The capacity to do work or cause physical change; energy, strength, or active power

Electro - see above
Magnetic - a force exerted between magnetic poles, a force that exists between two electrically charged moving particles.
Energy - The capacity for work or vigorous activity; vigor; power.

*Interesting similarities when breaking down into seperate definitions and appears to be the same meaning

[daven]

I think the term we're looking for (and what folks meant) is [magnetically] induced EMF.

[jojokeo]

Bingo! 

[Shrapnel]

Electro -of or pertaining to electrons or to an electron
Motive - something that causes action in a certain way, do a certain thing, etc.; incentive.
Force -The capacity to do work or cause physical change; energy, strength, or active power

Electro - see above
Magnetic - a force exerted between magnetic poles, a force that exists between two electrically charged moving particles.
Energy - The capacity for work or vigorous activity; vigor; power.

*Interesting similarities when breaking down into seperate definitions and appears to be the same meaning

LOLz.

Some have interpreted EMF to be "Electro Magnetic Field"  especially in the paranormal research communities. I learned it as "force."

Position is important for all the reasons mentioned. If you opt to put it at the input end, "gapping" the OT and the preamp circuits like you do between the PT and the OT is just as important IMHO. It's just harder to test how well you judge until the whole thing is operational.

Also a problem not mentioned is the direction of the windings. Potted trannies are real hard to guess at unless you potted it, or someone graciously provided the info. Most of the time the case is grounded on the chassis which helps.  A Belled transformer isn't a guarantee as to the direction of the windings either, (vertical or horizontal,) although lamination direction is obvious.

[HotBluePlates]

Electro -of or pertaining to electrons or to an electron
Motive - something that causes action in a certain way, do a certain thing, etc.; incentive.
Force -The capacity to do work or cause physical change; energy, strength, or active power

Webster would agree, but Oxford's Unabridged might yield additional options.

Electro - electron, or a unit charge.
Motive - no a reason to commit a crime, but of or related to motion
Force - something that makes you do something

You could call voltage "the force that makes a charge-carrier move." True whether you think in electron current or hole current.

But this is way away from the original point of the thread. It is somewhat more related to the thread about skin effect and frequency; very high frequencies (not the technical term) travel on waveguides with no center. There are E and H forces that move the current along the waveguide. Those forces represent the electrostatic and electromagnetic components of electric current.

[Bub]

Thanks for all the great info. The PT is 5.00 x 4.25 and the OT is 3.31 x 3.88. I'm assuming the longer measurements are the bell ends so I will rotate them 90 degrees at about 3" apart. I'm tight for chassis space because I want to put a tube effects loop board in about 4" away from the OT. I think the point that potted does not really have an effect on containing magnetic fields is what I was looking for.
Who knows, maybe I'll be able to pick up some EVPs when I'm done.

Thanks again

Rob

 Does the skin effect start at about 20KHZ ? Sorry, just had to throw that one in.

[HotBluePlates]

Who knows, maybe I'll be able to pick up some EVPs when I'm done.

Or channel EVH...

[dynaman1]

Who knows, maybe I'll be able to pick up some EVPs when I'm done.

Or channel EVH...

Haha, I tried that but am always put on mental hold...

[jojokeo]

Last night I rigged up the transformer trick set up and found some very interesting things revealing transformer electromagnetically induced hummmm facts vs fiction:
1) The PT & OT needs to be farther apart than I ever realized for total quietness - almost a foot away.
2) The PT orientation's bell ends made no difference when in typically upright mounted position
3) The quietest PT position is definitely when it is layed down ala Fender's position
4) The OT 90 degree lamination rule - at this point for 1 test - is complete myth and makes absolutley no difference
5) The OT layed down when PT is mounted in standard upright position helps equally well as when they are reversed as said above on line 3
6) The OT placed towards the PT's corner &/or offset to it makes no difference.

* Summary - The only thing to reduce the induced humm effects is the PT & OT's distance from each other unless one of them is layed down to the other being upright.

[Fresh_Start]

jo - do your transformers have end bells?
+ thanks for sharing the results of your experiments.  I'm definately going to use this approach before drilling holes in the chassis for my next build.

Cheers,

Chip

[Bub]

Wow! That really seems to throw alot of stuff out the window I have read. I'm definately going to do some testing before I mount my iron. The distance between OT and PT may help explain why my first build is so quite.

Thanks

Rob

[jojokeo]

jo - do your transformers have end bells?
+ thanks for sharing the results of your experiments.  I'm definately going to use this approach before drilling holes in the chassis for my next build.

Yes, Chip even the OT. Another thing I did tonight was use an NOS Stancor flat mount PT w/ end bells and it confirmed that same things, although this one was a bit "louder" or "hummier" 
In moving the OT around, it also needed to be on the end about 10" or so to become silent. Interesting also was when I moved it closer - as I rotated the OT so that it's end bells were also laying or pointing downwards - this was it's best position. I could get pretty close, ~4" or so w/ it being pretty darn quiet! There's no way to mount it this way (unless inside the chassis on one of the sides) but things were good when both were "laying down".
*A question for myself while doing this is, I wonder how much of what I hear would actually be heard after the amp is made and idling? Maybe a very small amount of humm is acceptable? But, common sense says that you'd want it silent if at all possible. 

Wow! That really seems to throw alot of stuff out the window I have read. I'm definately going to do some testing before I mount my iron. The distance between OT and PT may help explain why my first build is so quite.

Yes I agree, your build is just like what I'm talking about and recommend for best case mounting I'm discovering. This is answering questions that I have attributed to grounding in the past and has had me study and learn as much about grounding as possible, thinking that this had to be "it". Also I've been elevating my heaters on every build too. I've been very concious of layout and everything else when the whole time it looks to be that my iron is too close to get that last little tiny bit of noise to go away. 

[Fresh_Start]

jo - I've got to believe that there is a point of diminishing returns closer than 10".  Virtually all commercial amps have the PT and OT way closer than that.  My Super Reverb build had them much closer and it turned out scary quiet.  Lay-down PT and end bells on both transformers.

I'm also wondering if a push-pull topology with negative feedback around the poweramp might cancel out some of the noise you're hearing - or at least reduce it below the overall noise floor.  Makes me think a bigger spread between the PT and the OT might be more important for a single-ended amp.

Thanks again,

Chip

[jojokeo]

I know what you mean but the thing about the Super Reverb confirms my findings that the lay down PT allows the OT to be much closer to maybe w/in 4" to 5" or so from the tests.

I've pulled a couple good ol' amp books out and guess what? Many reputable manufacturers have them mounted on the chassis ends as in Bub's picture. The selected few showing the chassis from behind I could see illustrating this; Garnet, Groove Tubes, HiWatt, Hughes & Kettner, Matchless, Mesa Boogie, Orange, & Soldano.

The ones that mount them close together - Fender & Marshall. Fender atleast uses the laydown PT which allows them to be quiet. Leo had to have known this.

[HotBluePlates]

Fender atleast uses the laydown PT which allows them to be quiet. Leo had to have known this.

But Leo made combo amps. In the case of the tweed amps, there is often *very* little space allowed for the PT once you consider the OT and speaker/tube placement. And a laydown PT allows for easier power supply wiring.

Your results are very interesting, but also points out that different transformers can give unexpected results. We generally view toroid PT's as the "quiet ones" because the windings cover the core and have a shielding effect. But they can spew noise where the leads exit the windings.

Also, magnetic energy seems to want to jump from sharp corners. In any event, it's hard to know beforehand exactly what will be the quietest position.

I do have a suspicion that the induced hum will likely be drowned out by any applied signal in the OT. I'm also not certain what effect the unloaded secondaries might have on this whole exercise. I'll have to do some reading/testing to see if the is larger radiated field than when energy is being drawn out of the PT by loaded secondaries.

[Fresh_Start]

HBP - thanks for that note about torroidal transformers.  The leads usually exit the windings in the center, right?

Chip

[HotBluePlates]

No, usually from the edge of the "doughnut."

[jojokeo]

The only sure way to know is if we all make an amp & perform this test first noting our results or findings. But I don't want to make one with them close anymore if I don't have to.