K4600 w/ flat CRT barrel distortion adjustment *MOSTLY SOLVED*

[deega]

I recently "swapped" a flat crt from a 20 inch RCA TruFlat TV, yoke and all, on to a K4600 chassis.  I picked up the 4600 chassis because it matches this tube's yoke impedances and I really, really wanted to use this tube.  

After much fiddling and learning experiences I've managed to get a fantastic picture on this tube.  Except for the barrel distortion.  I don't intend to put a 4600 yoke on the tube -- I do have one -- because I really don't think it will resolve the barrel distortion and the headache of adjusting the yoke does not appeal to me, especially when I don't think it will solve the problem.  Why?  This is a flat CRT and, from what I gather in my reading of posts here and other sites, it's expected for flat CRT's to have barrel distortion in this sort of a swap.

What I'd like to do is intentionally degrade the K4600 pincushion circuit.  I know this happens organically as C614 fades, but how much degradation of C614 is necessary?  Is it possible to, perhaps, put a POT in place of R612 or R613 and make pincushion correction degradation adjustable?  

I believe the components which may change the pincushion correction characteristics if their values are tweaked are C613, C614, R612 and R613.  Anyone familiar with this circuit who'd like to give some pointers on this?  Is this circuit safe for experimentation?  I'm not going to destroy the monitor by inserting different values am I?

I'm only trying to make pincushion correction "worse".  

Here's a snippet of the schematic showing the circuit in question.

[deega]

Uhh..  Upon closer inspection, I'd leave most of those values alone.  They belong to vertical circuit at large and I'd be afraid to monkey with them.  However, T603 (pincushion transformer) and c614 appear to be isolated from the rest of the vertical circuit and exist only to modulate current to the horizontal yoke.

So, I can think of three options:

1. tap, rewind, or find a suitable replacement for T603
2. experiment with values for c614
3. lift the positive leg on the vertical side of T603 and place a pot in there to attenuate primary current. 

1 ain't gonna happen in my place.  2 seems terribly inconvenient and I'm thinking it's not going to degrade pincushion correction well enough.  3 seems like a feasible way to make this adjustable, assuming the horizontal side is not getting the lion's share of its current from this circuit.  I don't believe that's a wild assumption.

So, I'm going to TRY #3 in the next week.  Anyone care to warn me off a fool's errand?

[lilshawn]

Anyone care to warn me off a fool's errand?

okay. 


give your question some time... a few of the folks who could answer this properly don't log in every day.

perhaps you can give Rick at nieman video displays a call / e-mail. he has been very helpful in the past with puzzling issues with me.

[deega]

Oh, I'm in no rush..  The monitor is off my bench and being placed in my cabinet build right now.  Barrel distortion is lower on the priority list than painting, t-molding, plexi cutting, wiring, and general finishing touches to the cab before Friday (when one of my son's holiday break begins).  I'm building this as a gift to my three boys and need to get it out of the garage hiding spot and into the home hiding spot before Friday. 

I'll probably tackle this issue with more much focus starting this weekend, when the entire cab will be in my "office".

Thanks. 

[deega]

Okay..  The monitor is back online. 

This is a K4600 chassis w/ P317 input board.  All electrolytic caps were replaced a week or two ago.  B+ is 127.

Tube is from a RCA TruFlat TV, the surface is flat and I believe the flatness is more than cosmetic.  Tube number is A51LYZ095X manufactured 12/12/2004.  Yoke readings are 8.5 ohms V and 3 oms H.

I bought a K4600 chassis because of its yoke similarities.

Among the information I've found is this: http://www.repairfaq.org/sam/deflfaq.htm#dsnse with a little nugget of information, "The required field-parabola waveform is derived from the field deflection circuit. Amplitude and DC-level are adjustable, for correcting pin-cushion distortion and setting the screen width. EHT is *not* influenced!"

It's information similar to what's described on that page that led me to look at the K4600's pincushion transformer circuit.  If I'm reading the information correctly, pincushion correction can be adjusted by attenuating the current from the vertical side.  This would proportionally attenuate the modulation on the horizontal side, right??  And the K4600 seems to have a fairly simple circuit, almost looks like an afterthought.

I'm thinking that before I go hog wild and try mounting a pot in the circuit, I may just cut the lead to the pin xfmr from the V side and see what happens.  I'll have time to do this on Sunday, so hopefully I'll have some good news come Sunday evening..

Here's some pics of the barrel distortion.

[lilshawn]

well from what i understand, (which sometimes isn't much) but older non-digital pincushion is derived by the decay and charge of a capacitor through a resistor. as the voltage drops during the scan of the beam, it lines back up properly with the edge.

as you change the resistor (via the potentiometer) you change the flow  of voltage through cap and influence the start of the beam scan for that line.

i would think if changing the resistors in that circuit (pot either up or down possibly another resistor in line or across the adjustment pot that would limit the maximum or minimum value) doesn't help, you might need to change the cap to a different value.

***my 2 cents might not be worth much, mileage may vary.

[deega]

well from what i understand, (which sometimes isn't much) but older non-digital pincushion is derived by the decay and charge of a capacitor through a resistor. as the voltage drops during the scan of the beam, it lines back up properly with the edge.

Hey, thanks, lilshawn!  I have considered this, but I fear the repercussions.  R612 and C613 fit this description, but the circuit goes on beyond the pin xfmr tap and feeds back into the vertical deflection circuit.  I a little concerned that putting a pot in place of R612 could alter the behavior of the vertical circuit, and I'm pretty pleased with its output.  Except for my review of the pics I posted -- it appears I may have a little barrel on the V side as well.  I don't perceive this distortion with my own eyes, I only see it in the pics.  Cheap camera or accurate representation of the problem??

Perhaps putting a pot at R612 would be easiest first experiment to get a handle on behavior, though.  And if it works, WOOHOO!

What worries me most, though, is the lack of information regarding modifications to pin circuits.  With all the moans about pincushion problems, I'd have assumed someone would have tackled this sort of problem before.  I'm concerned that there's some obvious reason for this which I'm overlooking.  I really don't want to break it! 

[deega]

Well, in short, I shorted the primary, vertical, side of T603, the pin xfmr.  I tried sticking a pot in there as I'd read a patent on attenuation of amplitude modulation which indicated a variable resistor parallel to the parallel coil and cap would attenuate the amplitude of an AM tuner.  Reducing resistance here improved the barrel problem progressively until it was shorted.  There's still some room for improvement, so I may still cut T603 completely out, but the barrel problem is MUCH improved.

Well, now I have some other geometry problems!    Imagine that!  Looks like linearity is off on both axis and the circles in the corner of one test pattern are funky.  I have some "yoke shims", little plastic shivers with some metallic tips, that I can use to try and get rid of that..

Anyways, the test patterns looked a bit scary so I decided to boot up a network copy of MAME and check out a couple of games.  They look good, considering the MAME I booted is configured for the HTPC and not this monitor, and the monitor V/H has not been adjusted after my "fix" yet.

Some pics.

[deega]

Some MAME shots.

Any irregularities are not obvious when in games..  I'm gonna go out on a limb and call this one a success.

[deega]

Where the change was made on the schematic.  Of course, once the hack is permanent, I'll attach documentation of the hack to the chassis in case someone else ever ends up with it.

[lilshawn]

Right on, right on. it looks as though the pulse that the circuit gets from the horizontal trace is being diverted at full power instead of being attenuated. the transformer is still doing it's job, though possibly at a higher voltage now, not necessarily bad...i would keep an eye on the temp of that transformer. it might start running hot after prolonged period of time.

convergence is usually really bad around the corners, especially with static images on old beat up tubes. once you get a moving game video signal on there you will hardly notice it. unless the colors are 1/2 inch out, i wouldn't worry too, too much about it. buuuut... to each their own.