OT placement?

[SLW]

Is there any general consensus on OT placement on the chassis?  It seems like most amps have the PT mounted at one end of the chassis and the OT in the middle with the laminations perpendicular to each other(most of the time).  I have always thought that this was to keep the wire runs short.  I have mounted the PT on one end of the chassis.  I wanted to drill the mounting holes for the OT before I start putting it all together.  The wires on the OT are long enough to mount it at the other end of the chassis if I want.  I had planned on mounting it in the middle but I realized that I was not sure why in the middle instead of someplace else.

[tubeswell]

As long as the magnetic flux around the OT is oriented so that it doesn't get mixed up with the magnetic flux around the PT, then you can put the OT and PT quite close to each other. (So keep the 'cores' at right angles to each other). But all iron is best kept as far away from the Input/V1  end (and from any reverb cable sockets) as possible.

[SLW]

While I had considered the magnetic field directions, I had not considered any issues with V1/input to the amp.  Thanks for taking the time to answer and point that out.

[Fresh_Start]

If you go to the FAQs on the AX84 board, they describe "the headphone test" for OT placement.  You might want to try that, although the results frequently seem to be placement of the OT at the opposite end of the chassis from the PT.

Here's an overly simplified way of viewing the magnetic fields of transformers:  look down on the edge of the laminations and visualize two circles, one going out from each bell, and touching at the center of the laminations.

The classic Fender setup with a lay-down PT has its most of its magnetic field pointing up and down.  That's fine at the power-amp end of the chassis.  The OT has its laminations perpendicular to the PT laminations, and the OT is upright so its magnetic field mostly goes lengthwise along the chassis and doesn't radiate into the circuit board which typically is directly underneath (or above) the OT.

Here's a typical Fender-like setup:

Note that the choke is placed between the PT and the OT and its laminations are parallel to those of the OT.

This is an example of perpendicular placement when both the PT and OT are upright versions

For this situation, you might choose to rotate both the PT and OT 90 degrees so that the PT endbells are facing front-to-back instead of along the chassis.  Then again, it must not matter much in that amp.  Might be more important in a single-ended amp because they don't benefit from the power amp noise rejection you get with a push-pull circuit.  That may explain the AX-84 "headphone trick" - that site's focus was entirely single-ended amps for years.

Both of this photos are from the GALLERY here (see tiny print near the top of the page).  Look at lots of photos of amps, and you'll get the general idea.  Of course, there's a pic of one of Hoffman's amps in the Gallery that ignores the perpendicular "rule" and has the PT, choke and OT with their end bells in line and all laminations parallel.  If Doug made it, I'm confident it works.

Hope this helps,

Chip

[HotBluePlates]

What you planned to do is exactly how I normally mount the OT. My primary reason is simply aping what I've seen in most Fender and Marshall amps. I stick with what works.

Some folks have advocated placing the PT and OT on opposite ends of the chassis. The primary (maybe only) benefit of this choice is that the weight of the chassis is more balanced (it doesn't lean excessively to one side when picked up), and the transformers' weight is moved closer to the vertical sides of the chassis where the whole structure is stiffer.

But I think people overlook the steps taken to make this work. People might cite Hiwatt as an example of transformers at opposite ends of the chassis, but few of these folks have ever seen a Hiwatt in person (I used to own a 70's DR504). The chassis is very spacious. I don't know the exact measurements, but look at a picture of a DR504 and note how much depth the chassis has. A view from the rear shows that the output tubes are larger than typical EL34's, so they are likely American-made 6CA7's, with large bottles. The OT is on the right in the rear-view, near the white jacks. If you pay close attention, the OT is indeed on the "input" side of the amp, with the PT on the same side as the power and standby switches.

The problem with having an OT on the input side of the amp is that the primary wires connecting to the tube plates have large signals present. Big voltages, big (relative to the rest of the amp circuit) currents. Any wire carrying a.c. has a field around the wire; the bigger the a.c. (in voltage or current terms), the bigger and stronger the field. Any nearby wiring can potentially act as an antenna to pick up this field, especially high impedance circuits (such as those found at tube grids). The objection to having the OT anywhere near the input jacks (or other high impedance wiring) is the possibility of oscillation due to coupling the output of a circuit back to the input.

So how does Hiwatt get away with doing this? Look closely at the inside of the DR504 chassis. The input jacks are at the top left of the image, and the OT wiring and speaker jacks are at the bottom left. There is quite a lot of distance between the wiring for the OT and the input wiring, and each are dressed in opposite direction to minimize any coupling. Further, that gray input jack wiring is shielded wire, and the shield is only cut right near the grid pin of the tube socket (and right at the input jack lug).

I'll argue that's much more distance than commonly seen in any amp where folks attempt to place the transformers on opposite ends of the chassis. If we try to estimate the distance based on the landmarks we have (tube socket holes, cap can holes), I would not be surprised if it's 8-10 inches or more. That's likely more distance than found in a tweed chassis.

So transformer placement depends on the chassis you're working with, and the layout you choose to use. But you always want to keep high-level signal wiring away from low-level high-impedance wiring. Chip's comments regarding transformer fields are still applicable, however coupling due to the transformer fields decreases with distance, and there is so much distance between transformers in the Hiwatt example that hum due to this cause is non-existent.