Maker Pro
Maker Pro

Video signal amplifier.

HANKMARS

Jul 28, 2019
498
Joined
Jul 28, 2019
Messages
498
I am attempting to construct a video signal booster using an LM747 dual op amp. My cable is a 4 conductor shielded type. Length is 500 feet wound on 6 inch spool. I receive enough signal to produce a black and white image but prefer to have a color image. The op amp is working good but I seem to be dropping the plunge to 0 volts, which I believe may affect my horizontal sync. My next attempt to correct this is to capacitively couple the input and output signals and perhaps install a pull down resistor in the output on the away side (of the amp) of the output cap. Any suggestions?

DSCF2227 (2).JPG
 
Last edited by a moderator:

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,700
Joined
Nov 17, 2011
Messages
13,700
For one, the LM747 is completely unsuitable for video applications. It has a bandwidth of typ. 1.5 MHz (datasheet) whereas a SD video requires at least 6.75 MHz of analog bandwidth (reference).
Second you usually cannot simply amplify a video signal. You need to restore the DC level and the sync signal as explained e.g. here.

On this page a few different video amplifier circuits are shown (I have tested none of them).
 

Alec_t

Jul 7, 2015
3,587
Joined
Jul 7, 2015
Messages
3,587
Welcome to EP!
I'm surprised you get any gain at all. A proper video-rated opamp would be better.
What supply voltage(s) are you using? The '747 won't work properly if the inputs are within a few volts of either rail.
My cable is a 4 conductor shielded type. Length is 500 feet wound on 6 inch spool.
That doesn't sound like an efficient antenna. How are the conductors connected? Where did you get the design?
 

HANKMARS

Jul 28, 2019
498
Joined
Jul 28, 2019
Messages
498
For one, the LM747 is completely unsuitable for video applications. It has a bandwidth of typ. 1.5 MHz (datasheet) whereas a SD video requires at least 6.75 MHz of analog bandwidth (reference).
Second you usually cannot simply amplify a video signal. You need to restore the DC level and the sync signal as explained e.g. here.

On this page a few different video amplifier circuits are shown (I have tested none of them).
Thank you for your reply. I agree that the LM747 is an unlikely component for successful results in this application. I happen to have a dozen on hand and was hoping to utilize one. This app is merely using what I am guess would be referred to as a CCTV camera. A security camera let's say. Actually, I suppose this one would be called a lipstick camera. Its horizontal sync pulses appear to have a time period around 30 something μs. Approximately 15K Hz. Well within the workable range of the 747. I am aware that between these sync pulses that much more information is stored. Color burst, brightness, blanking, etc. Therefore, to ensure a high quality image, I should certainly use a device that can easily handle the frequency of what I would refer to as the 'dot clock'. I believe that the dot clock was the highest frequency clock in a CRT type TV. I have rarely worked with video signal processing in my career and am quite unfamiliar with the specifics. In my particular application, high definition is not paramount. But neither is a ghosty, bleeding image suitable. I will try a few more experiments with the LM747 and try to stretch its capabilities. If my results are fruitless, I will acquire more appropriate components. Thank you again for your reply. The reference and explanation pieces were very helpful. MH 7/29/19
 

HANKMARS

Jul 28, 2019
498
Joined
Jul 28, 2019
Messages
498
Welcome to EP!
I'm surprised you get any gain at all. A proper video-rated opamp would be better.
What supply voltage(s) are you using? The '747 won't work properly if the inputs are within a few volts of either rail.

That doesn't sound like an efficient antenna. How are the conductors connected? Where did you get the design?
Thank you for your reply and inquiries. I was thinking that the LM747 had a good rail to rail response. I will check my data sheets. My application does not involve signal radio transmission. Merely a CCTV system. The spooled cable is necessary in order to pay out cable length for a travelling camera. The cable is what the client had already purchased and I am trying to economize since my work on this project is more or less free gratis. Not a critical performance system at present and high definition is not paramount. I do not believe the cable is even twisted pair. I do get decent resolution with a black and white camera but it would be preferred to have a color display. I am always open to suggestions. My power supply is a single sided +12 VDC from a computer power supply. MH 7/29/19
 

aurelZ

Jun 10, 2019
98
Joined
Jun 10, 2019
Messages
98
video_booster.gif Long time ago i was build video signal amplifier with transistors
3 stage C class with ..hmm i am not sure but i know that are in metal case
i think that they are some BF type for CRT TV video signal booster and they are good on 12V PS.
yes it something like this
 

HANKMARS

Jul 28, 2019
498
Joined
Jul 28, 2019
Messages
498
View attachment 45769 Long time ago i was build video signal amplifier with transistors
3 stage C class with ..hmm i am not sure but i know that are in metal case
i think that they are some BF type for CRT TV video signal booster and they are good on 12V PS.
yes it something like this
Thank you for your response AURELZ. There are some elements in this schematic that may prove useful. This circuitry is somewhat more involved than I desire, but may have to resort to. I suspect that somebody has put this entire circuit on a single IC chip and I may have to search for such a device. Thank you. MDH 07/29/19
 

WHONOES

May 20, 2017
1,217
Joined
May 20, 2017
Messages
1,217
You will need to impedance match your cable at both ends otherwise you will get lots of reflections that will make a mess of your signal. To do that, you will need to know the characteristic impedance of your cable be it, 50Ω ,75Ω, 300Ω or something else entirely.
 

HANKMARS

Jul 28, 2019
498
Joined
Jul 28, 2019
Messages
498
You will need to impedance match your cable at both ends otherwise you will get lots of reflections that will make a mess of your signal. To do that, you will need to know the characteristic impedance of your cable be it, 50Ω ,75Ω, 300Ω or something else entirely.
I have placed a balun on each end of my cable. Will that negate reflections? I have forgotten the relationship of ohm value to characteristic impedance. I seem to remember that flat antenna wire was 50Ω and coax was 75Ω or visa versa. I am sure that the ohm value was not per foot of cable but maybe related to self capacitance or such? If the baluns do not help with reflections, I did have a project where I could put 2 or 3 winds of ~ 20ga mag wire, onto a torrid coil, ~ 1/2" dia., and adjust the wire spacing until I was within specs. Am I going the correct direction on this? Also, how can I come up with an ohm value for my cable (char imp)?
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
Coax can be of many different impedances, from 25 ohm up. High quality cables will have the type written on the side. In the UK, amateur radio operators usually use 50 ohm but TV and satellite dishes use 75 ohm. I also use 300 ohm twin feeder to connect to a folded dipole.

Measuring the impedance is not simple, there are many ways of doing this.You could try looking at the ARRL Antenna Handbook. The simplest to understand is to put a very short pulse on the cable and look at any reflected pulse with a fast scope. Adjust the resistive termination until there is no return pulse.

Adding complications such as home made transformers will make matters worse.
 

Audioguru

Sep 24, 2016
3,656
Joined
Sep 24, 2016
Messages
3,656
An antique LM747 has such a slow output slew rate that your sync pulse input is gone before the output of the opamp does anything.

What are you trying to boost? The opamp circuit has a voltage gain of only 1 at low frequencies.
 

WHONOES

May 20, 2017
1,217
Joined
May 20, 2017
Messages
1,217
If I remember correctly. Flat cable is 300Ω TV coax is 75Ω and instrument type Coax is 50Ω. You would normally use a source resistor at the transmission end and a load resistor at the receiving end of the appropriate value.
A proper video amplifier such as the AD811 would be beneficial too.
 

HANKMARS

Jul 28, 2019
498
Joined
Jul 28, 2019
Messages
498
An antique LM747 has such a slow output slew rate that your sync pulse input is gone before the output of the opamp does anything.

What are you trying to boost? The opamp circuit has a voltage gain of only 1 at low frequencies.
My understanding is that gain = Rf/Rin. I will look again but a gain of 1 might be maximum gain at ≈ 1MHz. It appears that the horizontal sync pulses are ≈ 15KHz. I'm shooting from the hip here and will later search out actual freqs but 6.45MHz is roughly the dot clock speed. That speed may not even cause an output from a LM747. I appreciate your reply and will have solid figures as I do more experimentation. Cat 5 is relatively an inexpensive product and may help negate line loss of spooled cable. I imagine that there may be a cable with characteristics similar to coax, but physically softer, more flexible, and less bulky and recommended for spool applications. This is a back burner project and finished units will be subjected to harsh environments so damage is to be expected which is reasonable cause to use generic, readily available, economy class components. I am boosting camera video signal only, due to loss of color information. Color camera to spooled cable (500 feet) to digital monitor. No radio transmission of vid sig is involved.
 
Last edited:

HANKMARS

Jul 28, 2019
498
Joined
Jul 28, 2019
Messages
498
My understanding is that gain = Rf/Rin. I will look again but a gain of 1 might be maximum gain at ≈ 1MHz. It appears that the horizontal sync pulses are ≈ 15KHz. I'm shooting from the hip here and will later search out actual freqs but 6.45MHz is roughly the dot clock speed. That speed may not even cause an output from a LM747. I appreciate your reply and will have solid figures as I do more experimentation. Cat 5 is relatively an inexpensive product and may help negate line loss of spooled cable. I imagine that there may be a cable with characteristics similar to coax, but physically softer, more flexible, and less bulky and recommended for spool applications. This is a back burner project and finished units will be subjected to harsh environments so damage is to be expected which is reasonable cause to use generic, readily available, economy class components.
I am trying to boost only camera video output information. There is no radio transmission of this data. Video signal to spooled cable (500 feet) to monitor.
 

HANKMARS

Jul 28, 2019
498
Joined
Jul 28, 2019
Messages
498
You will need to impedance match your cable at both ends otherwise you will get lots of reflections that will make a mess of your signal. To do that, you will need to know the characteristic impedance of your cable be it, 50Ω ,75Ω, 300Ω or something else entirely.
ELECTRICAL IMPEDANCE 1.1.JPG
 

HANKMARS

Jul 28, 2019
498
Joined
Jul 28, 2019
Messages
498
Coax can be of many different impedances, from 25 ohm up. High quality cables will have the type written on the side. In the UK, amateur radio operators usually use 50 ohm but TV and satellite dishes use 75 ohm. I also use 300 ohm twin feeder to connect to a folded dipole.

Measuring the impedance is not simple, there are many ways of doing this.You could try looking at the ARRL Antenna Handbook. The simplest to understand is to put a very short pulse on the cable and look at any reflected pulse with a fast scope. Adjust the resistive termination until there is no return pulse.

Adding complications such as home made transformers will make matters worse.
Memory is hazy on this but it seems that when testing for reflected signal values (expressed in db ?) I plugged a watt meter directly into the output BNC connector. I recall that I had to have the proper impedance connector, in this case either a 50Ω or 75Ω connector. I don't recall if I was loading the output but it seems like I would have had to be loading the signal. The units being tested were satellite modems. Mid to late 1990s. Thank you for your reply MDH 08/03/19
 

HANKMARS

Jul 28, 2019
498
Joined
Jul 28, 2019
Messages
498
If I remember correctly. Flat cable is 300Ω TV coax is 75Ω and instrument type Coax is 50Ω. You would normally use a source resistor at the transmission end and a load resistor at the receiving end of the appropriate value.
A proper video amplifier such as the AD811 would be beneficial too.
I will get the data sheets for the AD811. I am certain that it will be a more appropriate component for my app. Cost is not really a deal breaker but there is no reason for overkill either. Impedance marching of the output to the load allows for maximum power transfer. I'm pretty sure that is fact. I appreciate your input. MDH 08/03/19
 

aurelZ

Jun 10, 2019
98
Joined
Jun 10, 2019
Messages
98
there is no need for any kind of ballon transformer it is not RF signal than video from camera right?
also there is no such a chip amplifier or i don't know which can replace 3 stage transistor amplifier
i don't know why is such a problemetic to build amplifier?
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,878
Joined
Jun 21, 2012
Messages
4,878
@HANKMARS Re: Post #9. A balun! Do you even know what that is, or what it is used for? Show us a schematic of what you are trying to DO instead of how you think it should be done.

If you are trying to move CCTV video from point A to point B, you use coaxial cable, either 75 Ω characteristic impedance or 50 Ω characteristic impedance, and a proper line-driver and a line-receiver amplifier at each end of the cable. You don't use CAT5 Ethernet cable, or ribbon cable of any kind, or whatever cable happens to be available. If your "customer" has purchased the wrong cable, that's his problem not yours. Advise him to take it back and purchase a spool of 75 Ω coaxial CCTV cable. Simplify your life by having him purchase it with coaxial connectors already attached.

These are just buzz words I have thrown at you, for it is obvious you have no idea what you are doing. Ditch the 747 op-amp. It is totally inappropriate. Video amplifiers are generally not a good candidate for bread-boarding. Pay attention to what @duke37, @WHONOES, @Audioguru and perhaps others with real experience have to say.

Me? I'm old and I haven't farkled with this sh!t since they quit broadcasting NTSC video over the air to my rabbit-ear antennas. And BTW, there is no way anyone would EVER use a class-C amplifier to boost ANY sort of video signal. Video was ANALOG until fairly recently. Class-C amplifiers only amplify the peaks of the input signal and are generally used only in radio frequency applications.
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,878
Joined
Jun 21, 2012
Messages
4,878
You don't use CAT5 Ethernet cable
Edit: Sorry for the misinformation in my post #19: you can use CAT5 Ethernet cable, which consists of twisted pairs of insulated wires, for analog video.The pairs must be connected to impedance-matched differential transmitters and receivers to achieve a 500 foot distance without loss of chroma (color) synchronizing information that is encoded in the horizontal sync pulse.
 
Top