My DIY $6.45 CPL filer for my A119

@DavidUK & @RCPilot have you tried your DIY filters on the A119 at night?

Yes. I never remove it. Didn't notice any significant issues using it at night. They may be a little more reflection but nothing bad.
 
We DIY one CPL from M6 Plus,it seems to be ok on A119 dashcam also!
TB2Ft.zdxaJ.eBjSsziXXaJ_XXa_!!2788558980.jpg
 
You probably worded that wrong.

For those that don't know, the test to distinguish a CPL from a linear polarizer is to look through the CPL at your eye reflected in a mirror. If it is a CPL then you will only be able to see your eye if it is the correct way around for the camera to look through it, if it is a linear polarizer then you will see your eye whichever way around it is.

Just discovered my clip-on "CPL" labelled mobile phone filter is actually linear having done your test. And I bought some film ( https://www.ebay.co.uk/itm/191983059127 ) thinking it's linear when it's actually circular.

@Nigel - the reason I only just found out is that I previously thought they were the other way around as the linear polariser goes dark at 180 degree intervals only one side toward my eye when I view my Macbook screen through it, whereas the CPL film goes dark at 180 degree intervals whichever side I look through. I'd assumed that because CPL = one side with a mirror, it followed that the "one side" rule would apply to an LCD screen. Not so, but what's the technical explanation please?
 
Last edited:
not sure what the technical explanation is but the CPL filters will only go dark at 180° intervals from one side when looking at an LCD, won't work if the glass is the wrong way around
 
Just discovered my clip-on "CPL" labelled mobile phone filter is actually linear having done your test. And I bought some film ( https://www.ebay.co.uk/itm/191983059127 ) thinking it's linear when it's actually circular.

@Nigel - the reason I only just found out is that the linear polariser goes dark at 180 degree intervals only one side toward my eye when I view my Macbook screen through it, whereas the CPL film goes dark at 180 degree intervals whichever side I look through. I'd assumed that because CPL = one side with a mirror, it followed that the "one side" rule would apply to an LCD screen. Not so, but what's the technical explanation please?
That's a confusing post!

The light from an LCD display is always linearly polarised since that is how Liquid Crystal Displays control their light output.

The light from a TV is normally polarised in a vertical direction so that polarized sunglasses don't turn it completely black, don't know about Macbooks but I wouldn't be surprised if it was at 45 degrees. 3D TVs are different.

If you put a CPL in front of the LCD display with the linear filter side towards the LCD then the vertically polarised light will go through the linear filter only if you hold the filter so that it is vertical, if you rotate it 90 degress so that it is horizontal then the vertically polarised light from the LCD will be blocked, if you rotate it another 90 degrees so that it is vertical again (but upside down) then the light will pass through again.

If you put the CPL with the 1/4 wave plate side towards the LCD and the linear filter towards your eye then the vertically polarised light will be turned into circularly polarised light before it reaches the linear filter, since it is no longer either vertically or horizontally polarised the filter will not block it. Normally it changes the colour slightly when you rotate it, not sure why but I guess it is due to the filter and 1/4 wave plate being located within a wavelength distance of each other and different colours being different wavelengths.

Hope that helps and isn't too confusing!
 
I'm added some words in red to my previous post to help explain. Also, I'm looking through each filter one at a time, not combined.

In view of what Jokiin says, I'm going to re-do the mirror and LCD tests again in better light (was in a dim hallway). Let you know, plus some stills of what both filters achieve through the A119.
 
OK, re-checked. I have two types of filter:

1/ The clip on mobile phone filter labelled "CPL" Screen Shot 2016-11-28 at 11.10.35.png

2/ The Samsung S6 adhesive polarising fim: https://www.ebay.co.uk/itm/191983059127
Mirror test:
1/ I can see my eye whichever side I look through = PL (not CPL as advertised)
2/ I can see my eye through one side, but the other is dark = CPL

LCD test:
1/ The screen goes dark at 180° intervals whichever side I look through = PL
2/ The screen goes dark at 180° intervals when viewed from the camera's point of view, but when viewed from the other side it doesn't go dark at any point of rotation = CPL

@Nigel @ Jokiin - sorry, my post #63 above was confused and had things the wrong way around (trying to work from memory!).

Here's the PL (left) and the CPL (right) to compare...

Screen Shot 2016-11-28 at 11.25.20.png

And here's the BARE lens (left) with CPL (right) to show how reflections are cut down...

Screen Shot 2016-11-28 at 11.26.57.png


A square of CPL film was cut oversize and stuck to the front of a 20.8mm crystal watch glass: https://www.ebay.co.uk/itm/300957615052
Film then trimmed to match glass and a 0.1mm shim washer glued to the rear of the glass after "blacking" the front of the washer out with permanent marker

(matt black paint better)... Screen Shot 2016-11-28 at 11.37.44.png


Two 5x5mm neo magnets stuck either side of the lens housing with clear (removable) glue... Screen Shot 2016-11-28 at 11.35.38.png


Shim washer sticks to magnets behind... Screen Shot 2016-11-28 at 11.29.20.png

This does need a slight revision to stop the weight of the glass pulling it down but I have a few ideas...
 
@Nigel @ Jokiin - sorry, my post #63 above was confused and had things the wrong way around (trying to work from memory!).
It's not easy to keep these things simple, but you've done a very good job of the explanation this time :)

It's certainly a worthwhile improvement even without the sunshine. I think your Porsche windscreen is probably at too gentle an angle angle to get a really good result towards the sides, for the light to get fully polarised the angle of reflection needs to be at around 56 degrees, see: https://en.wikipedia.org/wiki/Brewster's_angle

This does need a slight revision to stop the weight of the glass pulling it down but I have a few ideas...
Normally when attaching things by magnets, I use a magnet on both surfaces, then the magnets are self-centring and don't slide out of position.
 
A cheap CPL lens will always have vignetting. In your case, edges of video is blurry, you can't see license plates etc.
 
It is only circular polarising filters that need to be the correct way around, if the light goes through the quarter wave plate first then it is no longer horizontally polarized when it reaches the linear polariser.

The light from the reflections will be mainly horizontally polarised, so you want your linear polarizer exactly vertical to cut out the reflections. You can test the position by holding it up to a TV screen, most TV's (except 3D ones) are vertically polarised so that you can see them through polarized glasses, if you hold the filter at exactly 90 degrees to what you want it in the car then it should completely cut out the TV image.

Most LCDs used in cars are diagonally polarised as the designers didn't know which way up they would be mounted and diagonally means that they can be viewed through polarized glasses whichever way up they are, although only at half brightness! So if you test it against a car LCD it should cut out the image at 45 degrees to how you want it mounted. If you want to test your TV, use some polarized sunglasses, they are vertically polarized like you want your filter to be.

@Nigel - I ran your test with both of my PL and CPL DIY filters which have a mark to signify the correct rotated position when fixed to the dash cam so as to give minimal reflections.

Using my Macbook Pro Retina screen, when positioned as they would be on the dash cam, both filters were fully clear, but rotating them both 90° fully blacked out the LCD screen as per your test. So, with the Samsung S6 CPL film I now know exactly how to position it for the A119 without trial and error.

With the S6 CPL film, as purchased, in its normal orientation, i.e. short sides at top and bottom, long sides left/right, when rotated 45° anti-clockwise from this, with the adhesive side toward me, it blacks out the LCD screen. Rotated from normal 45° clockwise will be the correct position for the A119.

This being a mobile phone part...
https://www.ebay.co.uk/itm/191983059127...it follows the same 45° rule as for in-car LCD screens.
 
@Nigel - I ran your test with both of my PL and CPL DIY filters which have a mark to signify the correct rotated position when fixed to the dash cam so as to give minimal reflections.

Using my Macbook Pro Retina screen, when positioned as they would be on the dash cam, both filters were fully clear, but rotating them both 90° fully blacked out the LCD screen as per your test. So, with the Samsung S6 CPL film I now know exactly how to position it for the A119 without trial and error.

With the S6 CPL film, as purchased, in its normal orientation, i.e. short sides at top and bottom, long sides left/right, when rotated 45° anti-clockwise from this, with the adhesive side toward me, it blacks out the LCD screen. Rotated from normal 45° clockwise will be the correct position for the A119.

This being a mobile phone part...
https://www.ebay.co.uk/itm/191983059127...it follows the same 45° rule as for in-car LCD screens.
Be careful when using any LCD screen to determine optimal positioning of a polarizing filter as the screen itself has polarization characteristics that affect the results (some work at 45 degrees, some vertical and some horizontal). The only 'correct' way to do it is in the vehicle itself.

There's a lot of discussion, examples, and solutions in this thread: https://dashcamtalk.com/forum/threa...er-manual-adjustment-how-to-adjust-diy.15022/
 
@DT MI - Looking at nico's video in the thread you linked to, I can't work out why 45° between "dark" and "white" is the optimal position?

With the mark on my DIY CPL filter positioned for optimal actual viewing on the A119, it takes a 90° turn to make my Macbook screen go fully dark.
In other words, with the filter rotated correctly for the A119, it is in nico's "fully white" position 90° away.

Am I missing something?
 
With the S6 CPL film, as purchased, in its normal orientation, i.e. short sides at top and bottom, long sides left/right, when rotated 45° anti-clockwise from this, with the adhesive side toward me, it blacks out the LCD screen. Rotated from normal 45° clockwise will be the correct position for the A119.

This being a mobile phone part...
https://www.ebay.co.uk/itm/191983059127...it follows the same 45° rule as for in-car LCD screens.
People use mobile phones in both landscape and portrait orientations, so to work with polarised glasses it needs to be at 45 degrees. There was a trend for computer monitors to be usable in portrait mode too, which I think is why nico's screen is also at 45 degrees, normally TV, laptop and most computer monitor screens are vertical like your MacBook since you can then see them full brightness when wearing polarised glasses while the 45 degree screens can only be seen at half brightness. 3D TV screens are different, they can even be circularly polarised!
 
People use mobile phones in both landscape and portrait orientations, so to work with polarised glasses it needs to be at 45 degrees. There was a trend for computer monitors to be usable in portrait mode too, which I think is why nico's screen is also at 45 degrees, normally TV, laptop and most computer monitor screens are vertical like your MacBook since you can then see them full brightness when wearing polarised glasses while the 45 degree screens can only be seen at half brightness.

quite surprised to find a monitor I bought recently (24" HP) was horizontal rather than vertical or 45 degrees
 
My final (?) DIY solution, before the Viofo CPL I ordered today arrives, is much simpler than the magnet versions, and costs £1.79 (if I make ten!)...

Parts:

Ten 19 x 14.1 x 1.5mm self adhesive fibre washer = £1.79 = £0.18 per unit
https://www.ebay.co.uk/itm/251120376539
One 15.1mm crystal watch glass = £1.50 (various diameters, including 14.1mm, but I prefer a slight stretch to the washer, above).
https://www.ebay.co.uk/itm/290957718709
One Samsung S6 CPL film (will make 15 filters) = £1.59 = £0.11 per unit
https://www.ebay.co.uk/itm/191983059127
Total cost per unit = £1.79

Overall cost to make one unit only = £4.88

At 1.5mm thick and 14.1mm internal diameter for the washer there is no vignetting, i.e. the inner edges do not show in the video clips.
A slightly oversize square of self-adhesive CPL film is stuck to the watch glass (outside edge), air bubbles easily removed, then trimmed to the glass edge.
This filter is pushed into the washer so the outer edges of each are flush. A few tiny drops of super glue on a needle are applied to the inner bevelled glass edge where they are sucked into the join with the fibre washer by capillary action.
The washer's adhesive protection paper is removed at the rear, as is the CPL film's at the front, and it's stuck lightly to the A119, centred on the lens housing.
Any final adjustment to rotation can be made by removing and repositioning the washer several times before pressing firmly in its final position.
The A119 lens is fully enclosed and protected from dust (cleaned before application of filter) and the CPL film can be cleaned with a micro fibre cloth without appearing to scratch easily.

Here's one I made earlier, using the 16mm glass element from the cheap mobile phone filter (the watch glass above is better optically, but not here yet).
Washer's a bit battered as the ones I happened to have are a few years old!...

Screen Shot 2016-11-29 at 19.28.56.png

If I could find 3M double sided rings 16x20mm in diameter I'd just stick the CPL film to a 20mm watch glass and stick this to the lens housing with the 3M ring. But I can't.

It's been fun, but I'm fairly sure the proper Viofo CPL will see all my experiments consigned to a drawer in a week or so. EDIT: Cancelled Viofo order due to issues raised 30th Nov. See CPL thread. Back to DIY!
 
Last edited:
quite surprised to find a monitor I bought recently (24" HP) was horizontal rather than vertical or 45 degrees
Being HP, maybe they had a 1,000,000 of them made for use as advertising/information displays in stations/airports/shops where they needed to be in portrait mode and where people where quite likely to be wearing polarised glasses.

Did you find out because you couldn't see it through your glasses, or where you testing CPLs?
 
Latest one. 0.9 x 15 x 15mm watch glass (£1.95) with Samsung S6 adhesive CPL film (£0.10)...

Screen Shot 2016-12-22 at 16.44.53.png

Stuck on (the little blobs behind the glass) using these...

Screen Shot 2016-12-22 at 16.46.44.png

Oriented before applying to the watch glass, as I know which way it goes now.
 
Back
Top