The picture on your Sentinel looks like the reciprocal of what I have on mine. Very interesting looking distortion.

It is opposite, but this is more normal of these vintage sets. The 19AP4 is set back behind the safety glass a significant distance. So there's some optical distortion from that. And the camera doesn't do it any justice. When watching a program it's not as obvious as these pictures indicate. I expect back in the day no one thought much of the shortcomings. There wasn't any alternative other than a movie theater. Technology advanced significantly over the next 10 years.

A sticky is a topic that stays at the top of the forum section despite other newer topics happening. In this case it's a sticky in the radio TV tech forum section. https://videokarma.org/showthread.php?t=264048

Chris,

Tonight I fired up my RCA TC-166(KCS 40A) chassis with the 5AXP4 test crt and put a DVD test pattern on it. To my surprise it shows the same concave edges you're experiencing. I don't remember it being like this, but I probably never used a test pattern when working with it earlier. This weekend I'll attempt to slide this back in it's cabinet and see if the 16GP4 makes any improvement. Right now I'm not sure what to make of it.

https://live.staticflickr.com/65535/...77f43e5d_z.jpg

Wow!

The slight bend is normal. It is due to the geometry of the CRT. The greater the deflection angle, the greater the bend.

The 15GP22 color CRT has a flat panel screen mounted internally within the tube. Because of this, the pincushion is very noticeable.

Wider deflection angle tube used pincushion correction. For monochrome sets, permanent magnets were used. For color, the introduction of the 25" rectangular CRT in 1965 required dynamic pincushion correction circuitry because permanent magnets would affect color purity.

I still contend the bell of the replacement 16GP4 has become slightly magnetized and removing the tube and carefully going over it with a degaussing coil will alleviate the problem.

Perhaps it was intentional to counteract the outward curve of the CRT face in metal cone CRTs?
https://live.staticflickr.com/65535/...25e2e196_c.jpg

So I should go over the metal bell correct…not do the withdraw circling from the screen

Considering the electron beam scans from a center point behind the crt face, it makes sense there will be distortion more so as it moves outward, predominantly in the horizontal scan. Back at this time RCA must not have considered it a serious enough problem to require correction. I'm not aware of any round screen monochrome set that implemented pincushion correction.

Problem for me is it can be seen when the image has straight lines in it. As you get towards the center it’s not noticeable but it is the last 2” or so of the vertical and horizontal limits of the tube

Is that image a zoomed image from the settings on your converter box Bob? Is that how you got it to fill the screen?

I'm feeding this set a standard 4:3 aspect ratio image. It's a Zenith porthole set in zoom mode. The set is designed to fill the entire screen in this mode by over scanning.

I’m going to try to demagnetize the metal crt bell. As good a place to start as any.

Yes the slight curvature to the faceplate help to reduce the pincushion effect
Only the much later monochrome sets with 110 degrees would you find pincushion magnets.

So the static magnetism to the metal bell is the only cause I can see to cause a pincushion effect on an old round metal CRT. The weird thing is that from Chris's photos the pincushion looked so uniform. Which would suggest the tube at some point in it's life must have gone through a strong uniform magnetic field.

When it was shipped, did it go through a parcel scanner to determine the box's contents?

I picked it up from an owner who bought it new almost 70 years ago. Tomorrow I’m going to do some yoke swaps to see if that makes a difference