How do old CRTs interlace?
23 January 2013, 11:09 PM
#1
Scooby Regular
Thread Starter
Join Date: Nov 2000
Location: 32 cylinders and many cats
Posts: 18,658
Likes: 0
Received 1 Like
on
1 Post
How do old CRTs interlace?
I have always wondered how an (old analogue, not computerised) CRT display flip flops between odd and even scan lines on alternate frames and have not been able to find out from searching.
What was the original mechanism before the computer age? Anyone have any links to explain how the scanning was controlled electronically not just for the interlacing? I know about the phosphors, mask, electromagnets to steer the beam, just wondered how it was controlled in an analogue world.
Sorry for an odd question, but sometimes I think understanding how something worked in the analogue world helps explain a lot.
Thanks!
What was the original mechanism before the computer age? Anyone have any links to explain how the scanning was controlled electronically not just for the interlacing? I know about the phosphors, mask, electromagnets to steer the beam, just wondered how it was controlled in an analogue world.
Sorry for an odd question, but sometimes I think understanding how something worked in the analogue world helps explain a lot.
Thanks!
24 January 2013, 05:05 PM
#2
This is going back some, but here goes, each line scan is 64u seconds including fly back and blanking (the video portion is 52u secs.) With interlaced scanning line scan and field scan are driving the scan coils on the neck of the tube. The field scan pulls the line scan down the screen while the line scans left to right. With 625 line scan some 50 horizontal lines at the bottom of the screen are used to allow a field ramp pulse to be generated for the field scan flyback. When the field scan fly back is initiated the line scan is pulled back up to the top of the screen. The scan starts again. You effectively have two half pictures per 50hertz. I think that is how it went. The line scan is highly inductive, required tuned circuits. There was sufficient current in line flyback (flyback transformer) to use this for generating low voltages and voltages up to 28kv for final anode tube acceleration. . Field scan due to its lower frequency more resistive so no need for tuned circuits
Last edited by andy97; 24 January 2013 at 05:07 PM.
24 January 2013, 05:21 PM
#3
Scooby Regular
Join Date: Jul 2005
Posts: 12,329
Likes: 0
Received 0 Likes
on
0 Posts
This is going back some, but here goes, each line scan is 64u seconds including fly back and blanking (the video portion is 52u secs.) With interlaced scanning line scan and field scan are driving the scan coils on the neck of the tube. The field scan pulls the line scan down the screen while the line scans left to right. With 625 line scan some 50 horizontal lines at the bottom of the screen are used to allow a field ramp pulse to be generated for the field scan flyback. When the field scan fly back is initiated the line scan is pulled back up to the top of the screen. The scan starts again. You effectively have two half pictures per 50hertz. I think that is how it went. The line scan is highly inductive, required tuned circuits. There was sufficient current in line flyback (flyback transformer) to use this for generating low voltages and voltages up to 28kv for final anode tube acceleration. . Field scan due to its lower frequency more resistive so no need for tuned circuits
On one scan your get half the picture 3xxx lines and on the next scan you get the remaining 3xx lines, but as its interlaced
Scan 1
xxxxxxxxxxxxxxx
---------------
xxxxxxxxxxxxxxx
---------------
xxxxxxxxxxxxxxx
Scan 2
---------------
xxxxxxxxxxxxxxx
---------------
xxxxxxxxxxxxxxx
---------------
Combined result
xxxxxxxxxxxxxxx
xxxxxxxxxxxxxxx
xxxxxxxxxxxxxxx
xxxxxxxxxxxxxxx
xxxxxxxxxxxxxxx
Mart
24 January 2013, 06:37 PM
#4
Scooby Regular
Thread Starter
Join Date: Nov 2000
Location: 32 cylinders and many cats
Posts: 18,658
Likes: 0
Received 1 Like
on
1 Post
This is going back some, but here goes, each line scan is 64u seconds including fly back and blanking (the video portion is 52u secs.) With interlaced scanning line scan and field scan are driving the scan coils on the neck of the tube. The field scan pulls the line scan down the screen while the line scans left to right. With 625 line scan some 50 horizontal lines at the bottom of the screen are used to allow a field ramp pulse to be generated for the field scan flyback. When the field scan fly back is initiated the line scan is pulled back up to the top of the screen. The scan starts again. You effectively have two half pictures per 50hertz. I think that is how it went. The line scan is highly inductive, required tuned circuits. There was sufficient current in line flyback (flyback transformer) to use this for generating low voltages and voltages up to 28kv for final anode tube acceleration. . Field scan due to its lower frequency more resistive so no need for tuned circuits
25 January 2013, 12:52 PM
#5
In Laymans terms,
On one scan your get half the picture 3xxx lines and on the next scan you get the remaining 3xx lines, but as its interlaced
Scan 1
xxxxxxxxxxxxxxx
---------------
xxxxxxxxxxxxxxx
---------------
xxxxxxxxxxxxxxx
Scan 2
---------------
xxxxxxxxxxxxxxx
---------------
xxxxxxxxxxxxxxx
---------------
Combined result
xxxxxxxxxxxxxxx
xxxxxxxxxxxxxxx
xxxxxxxxxxxxxxx
xxxxxxxxxxxxxxx
xxxxxxxxxxxxxxx
Mart
On one scan your get half the picture 3xxx lines and on the next scan you get the remaining 3xx lines, but as its interlaced
Scan 1
xxxxxxxxxxxxxxx
---------------
xxxxxxxxxxxxxxx
---------------
xxxxxxxxxxxxxxx
Scan 2
---------------
xxxxxxxxxxxxxxx
---------------
xxxxxxxxxxxxxxx
---------------
Combined result
xxxxxxxxxxxxxxx
xxxxxxxxxxxxxxx
xxxxxxxxxxxxxxx
xxxxxxxxxxxxxxx
xxxxxxxxxxxxxxx
Mart
29 January 2013, 09:39 PM
#6
Scooby Regular
Join Date: Aug 2003
Location: Sheffield; Rome of the North
Posts: 17,582
Likes: 0
Received 0 Likes
on
0 Posts
This is going back some, but here goes, each line scan is 64u seconds including fly back and blanking (the video portion is 52u secs.) With interlaced scanning line scan and field scan are driving the scan coils on the neck of the tube. The field scan pulls the line scan down the screen while the line scans left to right. With 625 line scan some 50 horizontal lines at the bottom of the screen are used to allow a field ramp pulse to be generated for the field scan flyback. When the field scan fly back is initiated the line scan is pulled back up to the top of the screen. The scan starts again. You effectively have two half pictures per 50hertz. I think that is how it went. The line scan is highly inductive, required tuned circuits. There was sufficient current in line flyback (flyback transformer) to use this for generating low voltages and voltages up to 28kv for final anode tube acceleration. . Field scan due to its lower frequency more resistive so no need for tuned circuits
Trending Topics
29 January 2013, 11:07 PM
#8
Wayhey, this take me back 
I built a raster circuit on some bread board once for my A-level stuff...used it to drive a oscilloscope. I've still got all my notes on PAL TV systems somewhere.
This is a quality website for old school reading: http://www.repairfaq.org/sam/deflfaq.htm (can't belive its still online, has loads of cool in depth stuff on microwaves, laser printers etc. )
I built a raster circuit on some bread board once for my A-level stuff...used it to drive a oscilloscope. I've still got all my notes on PAL TV systems somewhere. This is a quality website for old school reading: http://www.repairfaq.org/sam/deflfaq.htm (can't belive its still online, has loads of cool in depth stuff on microwaves, laser printers etc. )
Last edited by ALi-B; 29 January 2013 at 11:14 PM.
30 January 2013, 01:09 PM
#10
Scooby Regular
iTrader: (2)
Join Date: Mar 2001
Location: Swilling coffee at my lab bench
Posts: 9,096
Likes: 0
Received 0 Likes
on
0 Posts
Maybe one of you guys would be kind enough to suggest how I might fix something...
A couple of weeks ago I bought a digital storage oscilloscope off Ebay for repair. The CRT system is basically a b&w VGA resolution monitor, powered from +24V dc and fed an analogue video signal from the logic board.
It works fine for an hour or so, but then I turned my back on it for 5 mins and returned to find the image stretched vertically, so the middle 1/3 or so of the image occupies the whole display. The brightness also faded considerably, and after another minute or so the screen was completely black.
I'm not really familiar with how CRT drive electronics work, so short of replacing every electrolytic cap on the board and keeping my fingers crossed, I'm a bit stuck for ideas. I'm hoping someone will recognise the symptoms and tell me that it's something relatively straightforward - or at least, non-terminal!
The full schematic of the CRT board is on the very last page of this manual, and there's a beer in it for anyone who can correctly identify the fault before I do
A couple of weeks ago I bought a digital storage oscilloscope off Ebay for repair. The CRT system is basically a b&w VGA resolution monitor, powered from +24V dc and fed an analogue video signal from the logic board.
It works fine for an hour or so, but then I turned my back on it for 5 mins and returned to find the image stretched vertically, so the middle 1/3 or so of the image occupies the whole display. The brightness also faded considerably, and after another minute or so the screen was completely black.
I'm not really familiar with how CRT drive electronics work, so short of replacing every electrolytic cap on the board and keeping my fingers crossed, I'm a bit stuck for ideas. I'm hoping someone will recognise the symptoms and tell me that it's something relatively straightforward - or at least, non-terminal!
The full schematic of the CRT board is on the very last page of this manual, and there's a beer in it for anyone who can correctly identify the fault before I do
30 January 2013, 01:46 PM
#11
Don't laugh but....
Have you tried hitting it?
Moderate thump with the fist on the top or the sides (whilst its still displaying a picture). If it makes a differnce you are probably looking at a dry solder contact.
Ok, a more insightful stab at it: It sounds like a power supply/voltage drift problem to the transistor/oscillator circuit that controls the horizontal deflection, which is then struggling to maintain its scanning then finally shutting down due to either excessive/insufficent voltage (there will be a protection circuits to prevent x-rays from excessive voltage, however undervolting in other areas can cuase this to trigger as well IIRC).
The cause can be anything, duff cap, dodgy pot, dry solder joints. Thats as much as I can do.
Have you tried hitting it?
Moderate thump with the fist on the top or the sides (whilst its still displaying a picture). If it makes a differnce you are probably looking at a dry solder contact.
Ok, a more insightful stab at it: It sounds like a power supply/voltage drift problem to the transistor/oscillator circuit that controls the horizontal deflection, which is then struggling to maintain its scanning then finally shutting down due to either excessive/insufficent voltage (there will be a protection circuits to prevent x-rays from excessive voltage, however undervolting in other areas can cuase this to trigger as well IIRC).
The cause can be anything, duff cap, dodgy pot, dry solder joints. Thats as much as I can do.
30 January 2013, 02:47 PM
#12
Maybe one of you guys would be kind enough to suggest how I might fix something...
A couple of weeks ago I bought a digital storage oscilloscope off Ebay for repair. The CRT system is basically a b&w VGA resolution monitor, powered from +24V dc and fed an analogue video signal from the logic board.
It works fine for an hour or so, but then I turned my back on it for 5 mins and returned to find the image stretched vertically, so the middle 1/3 or so of the image occupies the whole display. The brightness also faded considerably, and after another minute or so the screen was completely black.
I'm not really familiar with how CRT drive electronics work, so short of replacing every electrolytic cap on the board and keeping my fingers crossed, I'm a bit stuck for ideas. I'm hoping someone will recognise the symptoms and tell me that it's something relatively straightforward - or at least, non-terminal!
The full schematic of the CRT board is on the very last page of this manual, and there's a beer in it for anyone who can correctly identify the fault before I do
A couple of weeks ago I bought a digital storage oscilloscope off Ebay for repair. The CRT system is basically a b&w VGA resolution monitor, powered from +24V dc and fed an analogue video signal from the logic board.
It works fine for an hour or so, but then I turned my back on it for 5 mins and returned to find the image stretched vertically, so the middle 1/3 or so of the image occupies the whole display. The brightness also faded considerably, and after another minute or so the screen was completely black.
I'm not really familiar with how CRT drive electronics work, so short of replacing every electrolytic cap on the board and keeping my fingers crossed, I'm a bit stuck for ideas. I'm hoping someone will recognise the symptoms and tell me that it's something relatively straightforward - or at least, non-terminal!
The full schematic of the CRT board is on the very last page of this manual, and there's a beer in it for anyone who can correctly identify the fault before I do
Look at page 369. The field op IC is a tda1170s, this would be the area to start looking. From memory the Tda1170 chips weren't the most reliable. Check all the caps and soldered joints in this section.
Last edited by andy97; 30 January 2013 at 03:05 PM.
30 January 2013, 05:43 PM
#13
Scooby Regular
iTrader: (2)
Join Date: Mar 2001
Location: Swilling coffee at my lab bench
Posts: 9,096
Likes: 0
Received 0 Likes
on
0 Posts
I won't say the thought didn't cross my mind! But given the way the screen faded out over time, it doesn't feel like a work/not work fault which could be explained by a broken joint.
I'd guess that it's actually a vertical deflection problem, given that the image retained the correct width but became stretched vertically, so the top and bottom of the image were lost off the edges of the screen.
I take the correct width to be a good sign, in that if the beam were actually weakened (by a low anode voltage, perhaps), it would be easier to steer and would therefore be deflected too far in both axes - does that make sense? Given that the image retained the correct width, I don't believe this can be the problem.
I have an order nearly ready to go in to Farnell for a whole bunch of replacement caps. The scope is nearly 20 years old, so a refresh has to be due soon anyway.
Will do. The TDA1170S is obsolete now but easily available on Ebay.
If I replace it, do you think I'd also have to readjust any of the pots on the board that control geometry correction? Or are those more likely to be a function of the tube than the IC driving it?
k, a more insightful stab at it: It sounds like a power supply/voltage drift problem to the transistor/oscillator circuit that controls the horizontal deflection
I take the correct width to be a good sign, in that if the beam were actually weakened (by a low anode voltage, perhaps), it would be easier to steer and would therefore be deflected too far in both axes - does that make sense? Given that the image retained the correct width, I don't believe this can be the problem.
Sounds like a field output drive fault. Being a digital storage scope I would assume that the drive is an integrated circuit design, but could use individual components for the field drive. A simple way to do a bit of fault finding before fully dismantling to get access is to use freezer spray. Gently spray the freezer on electrolytic capacitors to see if the symptoms rapidly change. Capacitors would be the first item to suspect.
Look at page 369. The field op IC is a tda1170s, this would be the area to start looking. From memory the Tda1170 chips weren't the most reliable. Check all the caps and soldered joints in this section.
If I replace it, do you think I'd also have to readjust any of the pots on the board that control geometry correction? Or are those more likely to be a function of the tube than the IC driving it?
02 February 2013, 04:52 PM
#15
Scooby Regular
Pots should only need adjusting based on input resolution (once set and all things being equal), so you shouldn't need to adjust any geometry if you're replacing like for like components.
I have 8 CRT's in the house (well.. a couple in the garage) for use with my retro and arcade game set-ups. I also have getting on for 10 video scalers, processors & line doublers, to try and get appropriate low resolution reproduction on modern display panels. My house is a homage to CRT and video tech.
I love this kind of stuff.
I have 8 CRT's in the house (well.. a couple in the garage) for use with my retro and arcade game set-ups.
I also have getting on for 10 video scalers, processors & line doublers, to try and get appropriate low resolution reproduction on modern display panels. My house is a homage to CRT and video tech. I love this kind of stuff.
Thread
Thread Starter
Forum
Replies
Last Post
Brendan Hughes
Computer & Technology Related
9
14 July 2004 11:15 AM