RGB to PAL video encoder with MC1377 works but no with everything...

[Axxel]

Hi everybody :grin:

I´m Axel, French and living in Barcelona and I love electronics :lol:

Sorry for my English I will try to explain it well ;-)

I searched for a solution during a lot of time and... I think I need a little bit of help...

I will try to explain you my problem.... :-?

I just made on my breadboard a RGB to PAL video encoder, it works very well on a CRT classic TV, but with a BT878 capture card, I see the picture but in B&W... And with a car DVD portable set with AV input, I can see only some parasites...

First I use a French/European Nintendo NES to get an RGB output. I extract the composite synchro with a LM1881 to enter into the MC1377 pal encoder, based en this scheme :



Here you can see a photo of the breadboard:



At the upper left, there is the MC1377 encoder, with a 3.58Mhz quartz to have an european PAL-BGHI output signal. I tried to amplify the output with a BF494C transistor and I get the same negative results.

At the upper right, there is a bands generator PIC based. (LINK -> https://www.edaboard.com/threads/4597/ ) In this case we won't use it because we are entering directly the RGB signal of the NES.

At the left down, there is another try of video amplifier, results to a fail too... This video amplifier use this scheme :



At the right down, there is the synchro-separator, based on LM1881:



Now the results in pictures: (The game is Megaman 4 and it is a very good game :wink:)

On the classic TV, perfect



On the computer with a BT878, we loose the colour...



And on the car tv, only few white lines...



So, i tried to compare the output signal with a correct PAL signal. I use a SEGA Megadrive to do that.

Now I make me sure everything is working correctly :







We can see everything working perfectly on the 3 devices.

Now let's compare the output signal with oscilloscope:

MEGADRIVE OUTPUT : **broken link removed**

BREADBOARD OUTPUT : **broken link removed**

I see a difference of the middle continuous range of about 2 volts. So I think my signal is very high? Or is it the MC1377 that create this problem?

I tried to amplify o reduce the signal, without any positive results. Sometimes I can have some changes but I never get it working on the 3 devices...

Thanks a lot for reading me and for your attention :grin:

Cheers :grin:

Axel.

 

[chuckey]

To remove the DC offset, you could fiddle with offset voltage on the bread board, but an easier way is to use a clamp circuit. Connect the output via a large capacitor. Now you have AC coupling so the DC level will bounce about according to the picture level. Connect a diode (Ge preferred) across the output to earth, with the anode to earth. this should then clamp the most negative of the video (sync pulse tips) to -.3V, if you use Si diode then to -.6V.
Frank

 

[Axxel]

To remove the DC offset, you could fiddle with offset voltage on the bread board, but an easier way is to use a clamp circuit. Connect the output via a large capacitor. Now you have AC coupling so the DC level will bounce about according to the picture level. Connect a diode (Ge preferred) across the output to earth, with the anode to earth. this should then clamp the most negative of the video (sync pulse tips) to -.3V, if you use Si diode then to -.6V.
Frank

Thanks for the answer! So basically I have to clamp the output video signal to get the -0.3V peaks and to quit the DC offset?

I have to put this on the output:

https://en.wikipedia.org/wiki/File:Negative_Voltage_Clamping_Circuit.svg

The condenser acts like high-pass filter to quit the DC offset and the diode is used to have the -0.3V peaks right?

I saw this scheme too, based on TDA 5850:



Is that the right solution?

Thanks

Axel.

 

[chuckey]

The wiki circuit ignores the diode voltage and uses the diode "the other way around", but yes, that or your TDA5850 would be a good place to DC restore your video. I hope it cures your problem.
Frank

 

[Axxel]

The wiki circuit ignores the diode voltage and uses the diode "the other way around", but yes, that or your TDA5850 would be a good place to DC restore your video. I hope it cures your problem.
Frank

Hello Frank! Thanks for your help ;-)

First, I looked after the TDA5850 but it is expensive and hard to find:-? so if I don't use it, I think it can be better for the moment...

Today I tried to investigate more about this problem and trying to follow your instructions:

I made this on the breadboard:



Now I totally lost the DC offset! When I connect the output signal to the car TV set, I can get a picture, but there is no color... but I get a picture



I made some tests and when I turn down the power supply at 7.5V instead of 12V, I get a picture color, a little noisy and distorted but I have color! The same thing happens with the BT878 capture card



Here we have the oscillograms to compare both with the PAL megadrive output:

:arrow: This is the megadrive output:



:arrow: This is the breadboard output:



I can see the levels of the megadrive are a little bit higher than the breadboard output... And the zero level is more little on the megadrive -0,1V and we have a 0V with the breadboard.

I think i'm following the correct path, I bought a germanium diode today to make a intent to make a clamp circuit but I didn't get any sucess... I can't see how this could work and be mounted... I need more help xddd

Regards

Axel.

 

[chuckey]

i can not se the reason for the 220 and 75 ohm resistors, other then attenuating your signal. If your cable matches the input impedance of the next device, there will be no reflections for you to terminate!
Frank

 

[Axxel]

i can not se the reason for the 220 and 75 ohm resistors, other then attenuating your signal. If your cable matches the input impedance of the next device, there will be no reflections for you to terminate!
Frank

Hello :grin:

I looked after MC1377 schematics and I found those:

The video output part of the SEGA Master System 1 Schematic wich use a MC1377, where you can see a similar schematic as I used:

I Tried and I get a BW picture on the car LCD monitor...



And a velleman KIT RGB Processor wich use too a MC1377:



I Tried too and unfortunatly and I get a BW picture on the car LCD monitor...

I asked to a freind his master system 1 to make a test and it was a frech version so it has only and RGB output :-( The MC1377 is absent on the french board :-?

 

[emresel]

Hi;
I want to remind one point.
Color loss is generally caused by burst freq error. If it is deviating too much, video decoders ignores color and only shows luminance (B&W).
Can you check MC1377 xtal. Try to measure color burst freq on video signal. If it is not 4.43Mhz try to change load caps of the xtal and place xtal closer to your chip.
The other point is burst amplitude. It should be around 280mV as i know. Measure it at the receiver side, check if its level is enough. (If its amplitude is not good, video decoder may not lock to correct freq).
From your explanations i feel that burst level is your problem.

Hope helps.

 

[chuckey]

I think its all down to the levels, if the colour is there on one sort of device then its there!!! Could be that the monochrome device needs more level to stop its colour killer or its xtal is not on frequency or we are back to the actual DC levels. As far as I remember, standard video has black level at 0V, sync tips - .3 V, peak white +.7V and with the colour subcarrier sitting on top(for yellow) I think it can go to + 1.2V.

 

[rfredel]

Hello Axxel,

At the upper left, there is the MC1377 encoder, with a 3.58Mhz quartz to have an european PAL-BGHI output signa

is it only a writing error, or do you use a 3,58MHz quartz? European Pal needs 4,43MHz, 3,58MHz is for NTSC.

I think your TV will be Multi-Norm and so you get color.

Regards

Rainer