shmups.system11.org

Shmups Forum
 
* FAQ    * Search
 * Register  * Login 
It is currently Sat Oct 16, 2021 8:22 pm View unanswered posts
View active topics



Post new topic Reply to topic  [ 379 posts ]  Go to page Previous  1 ... 9, 10, 11, 12, 13  Next
Author Message
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Wed Aug 18, 2021 3:05 pm 



Joined: 15 Jun 2021
Posts: 7
I am very interested in this mod to apply to a (European) SONY KV-25X5E.

ElBartoME wrote:
The resulting PCB is very small (40mm x 25mm).



Would you sell me one of your boards?

One additional question: would it be OK to swap a BF871 transistor with a 2SC3782? I'd like to see how a higher bandwidth transistor affects TVL.


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Thu Aug 19, 2021 8:52 am 



Joined: 07 Aug 2018
Posts: 71
antorsae wrote:
I am very interested in this mod to apply to a (European) SONY KV-25X5E.


I modded a 29X5D so I think it should work without a problem.

antorsae wrote:
Would you sell me one of your boards?


Since you are in Europe I can send you one PCB.

antorsae wrote:
One additional question: would it be OK to swap a BF871 transistor with a 2SC3782? I'd like to see how a higher bandwidth transistor affects TVL.


I don't think swapping transistors is as straight forward. You may also need to change additional passive components to make it work.


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Thu Aug 19, 2021 12:59 pm 



Joined: 07 Aug 2018
Posts: 71
smorsked wrote:
This thread looks very promising. I want to try to adjust G1 voltage to my Sony KV-A2943E scart Trinitron. With thinner scanlines it would be my dream crt. It has the same tube as the KV-29X5E wich people seem to love on these forums.
My question is about how to connect the G1 pins in the case where you have three G1 pins. on my KV-A2943E there are three pins, G1-1, G1-2 and G1-3 all of them are just grounded. Should I cut the three traces and connect all of them to the DIY powersupply or just one of them? Does the circuit change in any way because there are three pins?

Any help is much appreciated <3


In that case you need to cut the ground connection to all three G1 pins. You can see it here in my log: viewtopic.php?p=1441111#p1441111
Since then I'm not using the additional winding anymore and use Luke's circuit. Sonys have a pin on the flyback that is perfect for that. The A2943 has an almost identical chassis like the one I modded in that post (E3431).


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Thu Aug 19, 2021 1:14 pm 


User avatar

Joined: 14 May 2015
Posts: 121
Location: Australia
Just wanted to chip in and add to the enthusiasm for this mod. I have also found some of the criticism to be valid but maybe the delivery not so much. The technical stuff has also gone way above my head, but that’s fine. Maybe I’ll pick up the required knowledge in future and it will make things clear to me.

Is there a BOM and/or a project on OSH Park or similar where I can order this for myself? I’m not sure my Loewe CT1170 is suitable but I might hunt around for a Sony instead.


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Thu Aug 19, 2021 1:20 pm 



Joined: 07 Aug 2018
Posts: 71
bonzo.bits wrote:
Is there a BOM and/or a project on OSH Park or similar where I can order this for myself? I’m not sure my Loewe CT1170 is suitable but I might hunt around for a Sony instead.


When I get my PCBs and am sure they work I can make a BOM and release the Gerber files.


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Thu Aug 19, 2021 11:34 pm 


User avatar

Joined: 14 May 2015
Posts: 121
Location: Australia
ElBartoME wrote:
When I get my PCBs and am sure they work I can make a BOM and release the Gerber files.


Oh sheeeeeeeit, Christmas is just around the corner!


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Fri Aug 20, 2021 12:56 am 



Joined: 17 Aug 2020
Posts: 125
I cut and paste together the Alucard side-by-sides from this thread:
  • top left: my free shitty consumer slot mask CRT TV
  • top right: my modded free shitty consumer slot mask CRT TV
  • bottom left: SamIAm's 700 TVL professional broadcast CRT
  • bottom right: SamIAm's 900 TVL professional broadcast CRT

The pictures were not taken in the exact same area of the map so ignore the background pixels, Alucard's clothing is also different so use your imagination, but regardless, if you compare just Alucard's pixels you can get an idea of how this mod is making the spot size sharper, approximating the sharp spot size of a high TVL professional CRT. Improving amplifier bandwidth is still required to truly increase TVL, but regardless, the results are apparent. If you want something inbetween these extremes, the pot lets you tune for the sharpness and scanline thickness you want.
Image


ElBartoME wrote:
Image
The resulting PCB is very small (40mm x 25mm).


This is circuit board is great looking! A few questions & comments:
  • Instead of an electrolytic capacitor, can you use a stacked polypropylene or stacked polyester film capacitor like the type that Wima makes? Electrolytic capacitors are a curse for electronics collectors. 1uF is sufficient, and poly film capacitors last forever with sufficient voltage and temperature derating.
  • Where are you mounting the potentiometer? It doesn't look like there is a place on the PCB.
  • What options are you going to provide for mounting the board inside the TV? I have had good success with 3M adhesive standoffs, but being able to mount to the neckboard would be an interesting option if you came up with a technique that works.


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Fri Aug 20, 2021 3:30 am 


User avatar

Joined: 05 Mar 2018
Posts: 1622
Excellent work lads


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Fri Aug 20, 2021 7:57 am 



Joined: 07 Aug 2018
Posts: 71
LukeEvansSimon wrote:
This is circuit board is great looking! A few questions & comments:
  • Instead of an electrolytic capacitor, can you use a stacked polypropylene or stacked polyester film capacitor like the type that Wima makes? Electrolytic capacitors are a curse for electronics collectors. 1uF is sufficient, and poly film capacitors last forever with sufficient voltage and temperature derating.
  • Where are you mounting the potentiometer? It doesn't look like there is a place on the PCB.
  • What options are you going to provide for mounting the board inside the TV? I have had good success with 3M adhesive standoffs, but being able to mount to the neckboard would be an interesting option if you came up with a technique that works.


Thanks!

  • I thought about using a film capacitor. I decided against it because of two reasons. Cost and the load and temperature this capacitor is seeing is very small. The resulting lifetime according to Rubycons usable lifetime calculation is very long (>100,000h). Also the PCB is already in production. :mrgreen: It may be possible to cram a film capacitor into the footprint though.
  • You see the big black box right above the capacitor? That is the potentiometer. It is not one with a big handle as I was not able to find a supplier here for that. It is a normal multi-turn potentiometer with a screw. 0.5W with a working voltage of just above 200V. I have several different isolating screwdrivers that I use to change the voltage on the pot. They are cheap and I recommend everyone to get one.
  • Being able to put the PCB on the neckboard would be ideal to keep the wires to it as short as possible. Until now I have been zip-tieing the perf-boards to an empty space in the TV. But I like the idea on putting it on the neckboard and I think about a solution. Maybe 3D-printed spacer, but that has to be different for every neckboard...


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Fri Aug 20, 2021 9:47 am 



Joined: 18 May 2011
Posts: 15
yoZe wrote:
@LukeEvansSimon: Can you post photos showing a full cps2 grid to see the difference ?
Thank's.



Image


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Fri Aug 20, 2021 6:23 pm 


User avatar

Joined: 08 Jan 2016
Posts: 623
Location: Mountain View, CA
I've ended up doing this to a New Net City Toshiba PF chassis by manually adding a winding to the flyback (about 10-15 turns or so) and then rectifying and filtering the result. Pics soon.
_________________
Image


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Fri Aug 20, 2021 8:41 pm 



Joined: 17 Aug 2020
Posts: 125
mikejmoffitt wrote:
I've ended up doing this to a New Net City Toshiba PF chassis by manually adding a winding to the flyback (about 10-15 turns or so) and then rectifying and filtering the result. Pics soon.


Side-by-sides would be great! I just searched for details on the Toshiba PF D29C051 arcade CRT and it is trisync. It will have enough video bandwidth such that for 15khz video, the TVL bottleneck will not be video bandwidth. The bottleneck will be the smallest spot size you can achieve before moire effect becomes too noticeable since the CRT is a slot mask and not an aperture grille, which is less impacted by moire effect for very small spot sizes.


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Fri Aug 20, 2021 9:14 pm 


User avatar

Joined: 08 Jan 2016
Posts: 623
Location: Mountain View, CA
No side by side - I don't have two of them, and I won't be putting this one back. Plus, I'm mating it to a different tube than it is typically found with. I've never cared about the "TVL" so much as having a smaller and sharper beam shape for hard edges and better vertical definition, so I do not think I want to jump through too many hoops to try and get a comparison going.
_________________
Image


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Fri Aug 20, 2021 9:16 pm 



Joined: 17 Aug 2020
Posts: 125
mikejmoffitt wrote:
No side by side - I don't have two of them, and I won't be putting this one back. Plus, I'm mating it to a different tube than it is typically found with. I've never cared about the "TVL" so much as having a smaller and sharper beam shape for hard edges and better vertical definition, so I do not think I want to jump through too many hoops to try and get a comparison going.


Understood. My comparison picture above to high TVL pro CRTs, in my opinion, makes it clear that the "high TVL look" is largely due to a small spot size. Higher video bandwidth and a smaller dot pitch help, but a smaller spot size is easier to achieve to drive a visible change.


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Sat Aug 21, 2021 5:25 am 



Joined: 13 Feb 2021
Posts: 19
Lost my password and had to make a new account.

Very cool to see this bearing fruit. I have to admit, I'd be awfully tempted to try this mod if I ever get a chance to tinker with the two 32" JVCs I have in storage. In general, this has been a convincing demonstration of just how much of a difference the chassis side of things can make when creating an image with a CRT.

LukeEvansSimon, is your most recent shot the best one you can get? It looks like the camera is adding a distinct blur to me, as if it's out of focus. If you think you can get a better picture to illustrate your results, I bet it would be worth your while.

Also, based on the way my shots look in your four-way comparison, your image editor is adding some blur of its own, whether from the resizing algorithm or lossy conversion. I made my own lineup based on the 1024x1024 source version of your shot and my source images, which are on page 3 of this thread:

Full size:
Spoiler: show
Image


Scaled 50%:
Image


(Sorry that 900 TVL Alucard is such a chonk. I didn't want to change the width settings on the monitor when I took the shot. The bloom on the whitest whites in 900 TVL Alucard is also simply the camera's doing.)


Last edited by SamIAm MkII on Mon Aug 23, 2021 12:46 am, edited 1 time in total.

Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Sun Aug 22, 2021 5:04 pm 



Joined: 15 Jun 2021
Posts: 7
Although unrelated with the G1 mod I took a picture with physical reference to compare the dot pitch of a 25" Super Trinitron set with a 21" HiBlack one.

25" Super Trinitron tube (M60LCS60X):

Image

21" HiBlack Trinitron tube (A51JXH61X):

Image

The 25" set has more triads than the 21" set (within the ~same physical length).

Based on the above, if improving the video bandwidth of the set (almost independently of the G1 mod), the 25" would see more potential benefit than the 21"


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Wed Sep 08, 2021 6:21 pm 


User avatar

Joined: 10 Sep 2020
Posts: 16
Location: San Francisco, CA
ElBartoME wrote:
Most of the Sony consumers I modded keep the brightness constant when I change G1 and G2. But I'm still gonna verify the brightness.

I recently came across a 9" Zenith with G1 tied to a 24.1v rail. When I took the 24.1v off of it and tied it to ground I noticed a significant change in brightness. I had to turn up screen voltage to compensate, which I think is expected based on what Luke has been saying. The disparity between the "most negative" and "most positive" voltage at the amplifiers gets changed when you modify G1. I lowered G1 and had to increase screen voltage to compensate. Beam size is marginally improved but not much - which isn't surprising since the difference is only 24.1v

Left is before, right is after. You can see the origin of the beam after, and you can see there is slightly less light bleed to adjacent phosphers.
Spoiler: show
Image


ElBartoME wrote:
I modded a Panasonic TV where I also noticed these moiree patterns. The Sonys I modded never showed a pattern like this

Wouldn't this be due to the type of mask used? I would think dot/shadow masks would be the most prone to moire effect and aperture grille the least?


Last edited by treminaor on Fri Sep 10, 2021 8:42 pm, edited 1 time in total.

Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Fri Sep 10, 2021 8:23 pm 



Joined: 15 Oct 2017
Posts: 264
@LukeEvansSimon,

Have you much experience working on or studying PC CRTs?

I know aspects like cathode amplifier bandwidth/design, G1 voltage, etc. for certain TVs, tri-sync arcade monitors and professional monitors have been looked into here, and while this wouldn't entirely relate to the topic of this thread or contribute much to the end goal of modding a CRT to increase its TVL count, I'd be interested to know how this dichotomy between what you've called 'addressable' and 'resolvable' resolution (or at least what I interpret them to mean, maybe differently from how references in the OP define them) presents itself with PC monitors considering how flexible they are at accepting whatever you throw at them, while generally lacking TVL or equivalent measurements from the manufacturer despite that.

I know early in the thread, perhaps before work started on other elements to do with the neckboard, there were comments about how for most TVs, the bottleneck was not in the shadow mask or aperture grille and there could be room for 2x gains in resolving power before hitting that. I know in the case of some professional monitors, TVL figures line up almost precisely with how many dot or strip triads could be counted across the screen, e.g. the JVC TM-H150CG is advertised as being able to resolve 'more than 750 TV lines', and dividing its screen width by its 0.27 mm dot pitch would give you ~789 dot triads across 3/4 of a row of them.

You could either take that to mean this monitor (and others) were engineered to be able to alternate between a light/dark line for roughly each triad and the TVL count is a true measure of that, or JVC just marketed a TVL count based on its dot pitch because who's to say high frequency lines of alternating but similar shades of grey don't count if that's the case :P

In the case of a PC CRT which only has a listed dot pitch, is there a rule of thumb or general assumption for the relationship between dot/grille pitch, horizontal deflection and actual resolving power? Plenty seem to assume that a monitor just being able to accept a resolution and display a picture is the equivalent of it being able to fully resolve the details of that image at an acceptable level of sharpness and, unless engineered similarly to expensive professional monitors, we can't say for sure whether the dot/stripe pitch is a reliable indicator either (outside of giving us an upper limit in horizontal resolution), can we?

I found this old blog post relating to the Sony GDM-FW900 that linked an increased triad to pixel ratio to increased sharpness, suggesting a 1:1 match in triads to pixels wouldn't produce acceptable results, which I'm not sure about but is there. It's something I think about quite a bit though.


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Sat Sep 18, 2021 5:49 am 



Joined: 13 Feb 2021
Posts: 19
I'm afraid I made a basic mistake when I posted a certain calculation much earlier in this thread. Since it's referenced in the OP, I'd like to correct it.

LukeEvansSimon was trying to determine the amplifier bandwidth necessary for 1000TVL. I posted this response:

Quote:
One horizontal line of NTSC is 63.5uS, but only 52.6uS of that is active video.

52.6 x 0.75 = 39.45uS
1000 / 39.45 = ~25MHz minimum frequency response for 1000 TVL at 15KHz horizontal refresh.


It's true that 25MHz would be the pixel clock frequency necessary for 1000TVL. In other words, the signal does need to be alternating at the rate of 25 million times per second.

However, when dealing with analogue waveforms, a single cycle, or hertz, corresponds to a full period of the wave:

Image

Basically, you can get one white and one black line (or point, anyway) per cycle. This means that the bandwidth necessary for 1000TVL would be 12.5MHz, not 25MHz.

This makes sense. My 1992 BVM-2015S has a rated frequency response of 6MHz, yet it shows 320 pixel-wide Genesis games extremely well even though those use a 6.7MHz dot-clock. This explanation makes it clear that the analogue waveform frequency of that video would technically be more like 3.35MHz. This also explains why broadcast monitors capable of 1080i still only have a frequency response of about 30MHz when the standard pixel-clock for 1080i (and 720p) is 74.25MHz. Same with the LM2422 that was being talked about before, actually.

It's also good to clarify that NTSC broadcast standards specify exactly 4.2MHz for the video portion of the signal. I was thinking that it was more like an even 5MHz. Interestingly, if you look at ElbartoME's shots from February (page 8) and consider the likely pixel-clock rates (5.3MHz for 256, 7.6MHz for 384, 10.7MHz for 512 per this) and remember to halve those rates for the actual analogue frequency, you can see that his KV-E3431D clearly can't pass 4.2MHz without attenuation.

I might be late to that particular realization, but I wanted to point it out nonetheless. 4.2MHz may have been regarded by many engineers as a maximum rather than a minimum. That's unfortunate in a way...but on the other hand, it does make it all the more convincing that boosting amplifier bandwidth would yield visible improvements, and it lowers the bar for new designs, too. I'm very excited about the potential here, and I ever get the space for it, I'd love to experiment on a guinea-pig CRT myself. Maybe next year...


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Mon Sep 20, 2021 3:32 pm 



Joined: 13 Feb 2021
Posts: 19
I made a quick and easy "TVL test pattern generator" for anyone using CRT Emudriver. Which is to say, I churned out a bunch of images you're all free to use.

All you have to do is open Arcade OSD and set your desktop to the standard 2560x240p60 super resolution. Then, open this pack of images, view them full-screen, and just flip through them like a slideshow. Every image is exactly 2560x240 in size, so no scaling or filtering should occur. However, be aware that some image programs may do something you don't want regardless, so I suggest using irfanview if you're not sure. It's a lightweight program where all you have to do is double-click an open image to make it go full-screen, and press left and right to advance through the slideshow.

Each image features alternating black and white lines of one consistent pixel width: the image with the largest width has this at 15, and the image with the smallest has it at 1. Though we don't get to adjust in fine increments quite like we'd like to, this still lets us go through a very useful range via the 2560 pixels per line.

Here is an example of an actual image.
Spoiler: show
Image


Here is a quick snapshot of what this looks like on a real CRT:
Image

You can also see all of the images in this imgur post. When using them, you'll probably find this table handy:



(I also added a "0" image that shows 100% white in the upper half and 50% white (grey) in the lower half. Since "1" is so extremely high-frequency, I think most TVs are going to wind up turning it into a flat grey signal, so "0" is there for comparison.)

If you have a CRT Emudriver PC set up already, I think you'll be hard pressed to find an easier and more thorough way to test TVL than this, at least for now. I also think you'll have a blast just going back and forth through the whole set a few times on any CRT you can connect.

That monitor in the shot is an Ikegami HTM-2050R2, which is what was used for the "900 TVL" Alucard shot. That's in quotes because it's just what the manufacturer advertised. How does this monitor do with the whole set of images? Well, as you can probably make out in the scaled down shot above, even when displaying #15, the white lines are a little fatter than the black lines. As you go down through the set, not only do all the lines get more cramped (obviously), but the space-ratio of the white lines to the black lines gets less and less equal.

As has been talked about in this thread before, the definition of TVL is too general and vague to really pinpoint a cutoff. Having said that, to me, the biggest threshold was crossed between #3 and #2 (640 and 960 TVL, respectively). When displaying #3, even though there isn't much in the way of black lines left anymore, you can still clearly see the individual white lines as separate and countable dots along the scanline. When displaying #2, even though you can detect a ripple in the scanline, the whole line looks connected. Meanwhile, #1 shows a grey screen so solid that you'd never guess an alternating pattern was being input; sure enough, it's almost exactly the same intensity as the 50% white in #0.

I suspect that if this monitor were in 1080i mode, and each scanline were being drawn in less than half the time, the equivalent of #3 would look even better, and maybe #2 would have a fighting chance. Also, in the schematic for the RGB amp, there is an SD/HD switch that the system's computer toggles, and it looks like it has an effect on bandwidth. I don't care enough to try to hack it, but I take it as evidence enough that 900 TVL is probably not a completely unreasonable rating.

Anyway, I hope this is useful for someone. :)


Last edited by SamIAm MkII on Tue Sep 21, 2021 8:28 am, edited 2 times in total.

Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Tue Sep 21, 2021 6:46 am 



Joined: 17 Mar 2018
Posts: 200
Location: Germany
Very cool, thanks for taking the time to do these!
I might have to try these on the D32 for fun :lol:


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Tue Sep 21, 2021 7:29 am 



Joined: 13 Feb 2021
Posts: 19
I'd be very interested to hear your impressions! :D

The R2 version of the HTM-2050 doesn't have support for 720p, so there is no way for me to test that. If you are familiar enough with modelines to create your own 2560x720p desktop resolution, I'm sure it would be interesting to compare that with 2560x240p. At fully triple the horizontal scan rate, the amplifier bandwidth is all the more likely to be a bottleneck before spot-size and dot pitch.

I'll probably convert these to use at other resolutions, too. First, I want to confirm whether going above 2560 will break anything for anyone (including me). If 5120x240p is possible, that would give us some really nice additional screens (768, 548, 426, 349 TVL).


Last edited by SamIAm MkII on Tue Sep 21, 2021 8:26 am, edited 1 time in total.

Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Tue Sep 21, 2021 7:48 am 



Joined: 17 Mar 2018
Posts: 200
Location: Germany
Okay, I'll put it on my ToDo list!


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Tue Sep 21, 2021 2:43 pm 



Joined: 13 Feb 2021
Posts: 19
Here is a quick and cheap video showing how easy it is to change TVL screens. All I'm doing is pressing left on my keyboard.

https://www.youtube.com/watch?v=kQTaMGY2Q6g

Another thing occurs to me: If you had an Extron box with high bandwidth that could convert RGBHV to RGsB, you could feed it the RGBHV from your PC and simply take the "Gs" portion to feed to a composite video input or luma (Y) of S-Video or component input. Since chrominance doesn't matter, it would be fine...though I suppose the Y/C separation for composite might be affected by the pattern cycling at certain frequencies.


Last edited by SamIAm MkII on Tue Sep 21, 2021 3:42 pm, edited 2 times in total.

Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Tue Sep 21, 2021 3:24 pm 


User avatar

Joined: 16 Jan 2014
Posts: 1604
This is super cool, thanks for this! I cant wait to try on my sets at home. Any way you could do a 480p version for 31KHz only monitors?


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Tue Sep 21, 2021 3:35 pm 



Joined: 13 Feb 2021
Posts: 19
You know, as long as the horizontal width is still 2560 pixels, I don't think your image program will do anything too offensive when it stretches the image straight down. Irfanview doesn't seem to. But, for peace of mind, it might be better to simply use 2560x480 images in the end.

If people are interested, I'd probably like to make a few sets for various resolutions. Now that I have a system down, it doesn't take me much time.

At least PNG formatting gets these down to just a couple of kilobytes each.


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Tue Sep 21, 2021 3:46 pm 



Joined: 15 Jun 2021
Posts: 7
I did some experiments this weekend on a Sony KV-25X5E (European Super Trinitron, FE-1 chassis).

I did a drop-in replacement of BF871 with 2SC3503.

This is a dot pattern (15 Khz w/ SNES 240p Test Suite) after the mod:
Spoiler: show
Image


Then I connect a (150Mhz) probe: CH1 (yellow) to output of neck (red cathode), CH2 (blue) input to neck board (red channel), the probe/scope adds some parasitic capacitance and it results in a degraded signal (see red trail):

Spoiler: show
Image


This is the view from the scope:

Spoiler: show
Image


Detail:

Spoiler: show
Image


I was expecting to measure the rise/fall times however I am not sure if I can do it with the probe I have since it distorts the signal.

Any ideas?


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Tue Sep 21, 2021 3:55 pm 



Joined: 13 Feb 2021
Posts: 19
antorsae wrote:
I was expecting to measure the rise/fall times however I am not sure if I can do it with the probe I have since it distorts the signal.

Any ideas?


I remember that when I was scoping a neckboard long ago, I got a similar result on-screen that largely went away as soon as I switched the probes (and scope) into 10x mode. I wasn't looking at rise/fall times, though, so this might not help.


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Tue Sep 21, 2021 4:17 pm 



Joined: 15 Jun 2021
Posts: 7
SamIAm MkII wrote:
antorsae wrote:
I remember that when I was scoping a neckboard long ago, I got a similar result on-screen that largely went away as soon as I switched the probes (and scope) into 10x mode. I wasn't looking at rise/fall times, though, so this might not help.


Probe was already in 10x mode.


Top
 Offline Profile  
 
 Post subject: Re: Mod a CRT to increase its TVL
PostPosted: Tue Sep 21, 2021 7:20 pm 


User avatar

Joined: 16 Jan 2014
Posts: 1604
SamIAm MkII wrote:
You know, as long as the horizontal width is still 2560 pixels, I don't think your image program will do anything too offensive when it stretches the image straight down. Irfanview doesn't seem to. But, for peace of mind, it might be better to simply use 2560x480 images in the end.

If people are interested, I'd probably like to make a few sets for various resolutions. Now that I have a system down, it doesn't take me much time.

At least PNG formatting gets these down to just a couple of kilobytes each.



Count me in, I'd like to see sets for 480, 600, and 768 resolutions. Also maybe 900 and 1200 if you are really bored. :mrgreen: I think that would cover the majority of CRTs.


Top
 Offline Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 379 posts ]  Go to page Previous  1 ... 9, 10, 11, 12, 13  Next

All times are UTC


Who is online

Users browsing this forum: Google [Bot], shmupsman, Tempest_2084 and 12 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  
Space Pilot 3K template by Jakob Persson
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group