LVDS cable problem

Thread Starter

abradatanu

Joined May 10, 2014
10
Hi All,

New on this forum and I don't know if this is the right place for this question, so can an admin please move it if so

I have a DIY projector with a Pixelworks PW338 LCD controller driving a 15.4" LP154WU1 WUXGA screen and all has been working fine, but I now need to extend the LVDS cable from the controller to the screen by about 300mm.

Initially I tried Ethernet UTP pairs and tried to twist them at the same rate the LVDS cables were twisted and the screen was displaying an image, but it had horrendous colours and a lot of noise.

I thought it was because of interference as the cables weren't shileded so I opened up a VGA cable and I replaced the UTP pairs with individually shielded cables inside the VGA cable, I also used foil and braided shielding on the connection spots between the LVDS cables and extension VGA cables, no luck.
I still get exactly the same results as I did with UTP.

I tried different colour depth and resolution settings, the result is the same. As can be seen in the images the extension is only about 300mm long so that shouldn't be the problem as I heard people extending LVDS cables for longer distances than this.

The resolution I'm displaying on the projector screen is 1920 by 1080 through HDMI at 24 Hz but I also tried VGA with the same (actually a bit worse) result.

Does anyone know what might be the cause of this or what I should check?






 

bertus

Joined Apr 5, 2008
22,278
Hello,

LVDS cables are twisted pair cables.
See this wiki page for more info:
http://en.wikipedia.org/wiki/LVDS

A coax cable as in the VGA cable is not.
See this page of the wiki for info on coax cable:
http://en.wikipedia.org/wiki/Coax

The twisting of the cable will give the wanted noise suppression from outside.
That is how it is possible to have high speeds in the CAT series cables for ethernet connections.
The CAT6 cable has the highest speed.(upto 250 Mhz).
See this wiki page for more info on the CAT6 cable:
http://en.wikipedia.org/wiki/Cat6

Bertus
 

Thread Starter

abradatanu

Joined May 10, 2014
10
Thanks Bertus

If I think about it, the cable I tried initially wasn't actually UTP cable as I remember having this cable around and it being very old and it is the solid core type, not like CAT5 or 5e UTP which is made of strands.

Also from what I understand par of cables are twisted to attenuate interference from each other and having individual cables with foil and braided shielding should have the same effect on interference without having to twist the pairs.

So you think I should give it a try with CAT6 UTP cable. Can I just use the CAT6 cable as it is or do I have to take it out and try and match the number of twists to the LVDS cable twists?

And does this look like a result of interference?
 

bertus

Joined Apr 5, 2008
22,278
Hello,

Looking at the images, it looks like oversaturation, wich can be caused by overshoots in the signal.

The CAT6 cable has 4 pairs of wires in it with each an other numer of twists per lenght.
This "shields" the pairs from eachother.

Bertus
 

Thread Starter

abradatanu

Joined May 10, 2014
10
Oh, I get it now, the twists shield the pairs from each other but also the individual wires in the pairs. I agree that having different twist ratios for each pair in CAT5, 5e or 6 shields the pairs from each other, but in my LVDS cable all the pairs have the same twist ratio apparently.

also what I don't get is why do I get the same results with the individually shielded VGA cables as the shields, from what I read, should be much more effective at blocking electromagnetic interference emitted from each cable.

Anyway I don't have a CAT6 cable but would a CAT5e one work? Or should I wait and buy a CAT6?
 

Thread Starter

abradatanu

Joined May 10, 2014
10
OK, been real busy lately, but I decided to have a go at this with the small thin wires from the VGA cables I spliced.

I carefully measured the number of twists in the LVDS cable and then I twisted the VGA extensions using the same number of twists and the end result is MUCH better.

There is still some noticeable noise, and to get rid of this I will try to shield each twisted pair with aluminium foil and see if that works.
 
Top