EEVblog 1458 - Microscope Polarising MAGIC!
HTML-код
- Опубликовано: 27 авг 2024
- Can't see those pesky laser etched part numbers on chips? Dave demonstrates what's happening with the microscope polarising filter lenses on Twitter that's blowing people's minds. Does it actually work?
And other tricks of the trade.
DIY Light box for PCB photography: • DIY Product/PCB Photog...
/ 1494061604161220615
/ 1494343655053832203
Aputure Amaran AL-M9 LED light amzn.to/3H0O5KW
Leica microscope brochure: downloads.leic...
Forum: www.eevblog.co...
Support the EEVblog on:
Locals: locals.com/mem...
Patreon: / eevblog
Odysee: odysee.com/@ee...
EEVblog Web Site: www.eevblog.com
2nd Channel: / eevblog2
EEVdiscover: / eevdiscover
AliExpress Affiliate: s.click.aliexpr...
Buy anything through that link and Dave gets a commission at no cost to you.
Donate With Bitcoin & Other Crypto Currencies!
www.eevblog.co...
T-Shirts: teespring.com/s...
#ElectronicsCreators #Microscope #Polarizing
The reason why circular polarizers are preferred by photographers is subtle and not really adequately explained by "works much better than linear polarizers". The photos you would take with a linear polarizer would look *identical* to a so-called "circular polarizer", but the linear polarizer would send linearly polarized light into the camera, which can confuse and upset the auto-focus/auto-exposure features of the camera. So the circular polarizer adds some wizardy *after* the linear polarizer (namely a quarter waveplate) to take the light which has already been filtered by the linear polarizer part, and scrambles/circularizes the polarization. The filtering-according-to-polarization has already happened, but the polarization is rescrambled so that it doesn't screw with the auto-focus etc.
It just so happens that if you somehow attach the filter *backwards*, what you end up with is a filter that would be selective to circularly polarized light, but that's not the way we attach these filters to our cameras.
So TL;DR, it's really a terrible choice of name. A "circular polarizer" is a linear polarizer with a little hack added to prevent it from breaking auto-focus; and it has practically nothing to do (as far as the end user is concerned) with circular polarization.
100% correct.
Thanks, dude! Explains everything. So basically we can treat it as 2 separate films, not one. Will attenuation get bigger on those? Is there a usecase for linear polarizer in photography (low light conditions or something)?
@@__-nd4hf I'd wager that the attenuation caused by the quarter wave plate is pretty negligible; I can't imagine a reason for actually preferring a pure linear polarizer.
Yes, that what I said in the video, it upsets the autofocus and other systems.
He literally said the same thing in the video.
Fantastic, I'm 3min 30 sec in and I'm off to go buy that filter from the local photography shop. So exited.
I'll watch the rest of the video when I come back. Can't wait!!!
Oh dearrrr!
When he gets back, he's gonna be soooo pissed! ROFL
🤣🤣@@BlondieSL
oh my. you should finished watching the video first. hahaha
Cross polarisation is what's used in geology with thin sections to identify minerals.
Also in biology and crystallology. I made my own setup to look at stress patterns in plastics and glass. Stressing glass in the setup is really trippy
What he has done in fact is to reproduce the cross polarizing system of a petrographic microscope, while also adding a quarter-wave retarder. The problem is that he didn't have a control on the incoming light polarization.
that sounds similar to spectroscopy, thanks for telling.
Saved me $20 today, thanks Dave! Saw the tweet today, nearly bought the magic disc on impulse. Glad I didn't!
As I saw the first half
The video I was like welp I’m sold and about closed the video and went to Amazon - I just kept listening because my hands were full. Glad I did
Same
On the other end Dave almost make me waste money on the first part of the video :P
Polarising sunglasses are a must when driving IMHO. Road paint becomes more neon-like, road signs alike, you can see through the windscreens of other cars and see where the drivers are looking etc.
When I got my first pair seeing the other drivers through the back window glare was the best party trick
I had some for a few weeks, but the tinting on the rear windows of multiple cars I drove at the time would appear all checker board like... Which made it harder to see what was behind.
I should probably get some again since we own zero of those cars now, and I use the centre mirror much less now after several years of euro truck sim.
There was a time when polarising sunglasses were a serious hazzard when driving as the laminations and curves of a car windscreen would show up rainbow patterns all over the place which made it very hard to see the road properly. Even now they can cause problems as all those LCD screens in cars these days also polarise the light so it can be hard to read your binnacle displays (let alone the sat nav). Basically they reduce road glare but CAN create unwanted patterns so must be chosen with great care.
"Goose arm" because a goose's arm is way more flexible than a goose's neck 😂😂.
As usual, a great informative video.
For those without ready access to a dumpster where they might find an LCD screen, polarizing film is quite cheaply available on Amazon in sheets of various sizes....
I use my TV on a white screen as a light source, Then using a circuler polariser on my camera. I put a bit of moulded plastic in between them and you get some really wierd and out of this world photos from just plastic. It highlights the stress inside the moulded plastic.
Any example photos?
I've been using a simple technique: a small amount of silicone grease applied with a cotton swab over the chip. When I clean the grease, the inscriptions become very sharp without needing a microscope.
Yep, swap of IPA and the numbers appear instantly!
Yeah, I've found that side lighting is more effective than polarizers. If you have a small, high intensity flashlight you can shine it on the part from the side temporarily to read the markings while leaving the microscope and lighting set-up alone. Most of my microscope viewing is to asses solder connections, which turn to dark grey or black viewed through polarizers.
Cheaper to use both the lenses of a single polarized movie glasses ... a child's polarized movie glasses will set you back around £1
As shown in this video, but use one movie glasses lens over a light source, and other for viewing as shown in the video ... 1st find the lenses correct orientation, place one lens over the another held up to a light, look through, and flip and 360 rotate a lens until you can't see through them both ... when you find the darkest reaction, this is when you know you have the correct orientation ... remembering the lenses correct way around, use one over light source, one for viewing through as shown in the video ... using polarized movie glasses to convert a standard compound microscope into polarized light is common in microscopy
Or, if you have the glasses still in tact, you want to set it up so the light goes through one lens as if it was going towards your eye, and then reflect off the surface and come out through the other lens.
That is so both are acting as linear polarisers, and rotating one lens around will change the brightness.
If you go the other way, so light goes out from your eye, reflect off the surface and then back into the other eye, then rotating the lens will do pretty much nothing.
If rotating the polariser changed it, it either isn't a filter for only allowing circular polarised light through, or you are using it backwards.
A great advantage of circular polarisers is that the orientation doesn't matter.
But that is for use in things like 3D movies.
A circular polariser typically works by combining a linear polariser with a quarter wave plate.
To produce circular polarised light, it is first made into linear polarised light and then the quarter wave plate makes it into circular polarised light.
To then filter it, it goes in reverse, first passing it through a quarter wave plate to make it linear, and then filtering that linear polarised light.
Also, a lot of screens these days are starting to have circular polarisers on them as well, as it helps cut down glare.
Dave doesn't really make the point that when light is reflected from a surface at an angle, it is possible to get an angle where the reflected light is completely polarised. This called "Brewster's angle" (see en.wikipedia.org/wiki/Brewster%27s_angle ). A single polarising filter can then remove that light, so that the black surface of the chip is seen as black again, thus allowing the etching to stand out clearly. You don't need to polarise the incident light for that effect to work, so you don't need two filters. You just need to have the illumination coming from an angle, and the viewpoint at a similar angle on the other side. That's exactly how Polaroid sunglasses work.
Polaroid makes sunglasses?
@@newtonfigley6948 From en.wikipedia.org/wiki/Polaroid_Corporation
"The original Polaroid Corporation was founded in Cambridge, Massachusetts, by Edwin Land and George W. Wheelwright III in 1937. ... Polaroid’s initial market was in polarized sunglasses - spawned from Land’s self-guided research in light polarization. Land, having completed his freshman year at Harvard University, left to pursue this market, resulting in Polaroid's birth."
So, yes.
Nice Video Dave! The effect that the image gets darker by using the polarisation foil with unpolarized light is not due to losses. Because randomly unpolarized light means that 50% of the light matches the direction of the polarisation foil, leads to a blocking of these 50%. So because 50% of the electric fields gets blocked, the light intensity should drop by 1/sqrt(2) if I remember correctly
And when you add a third polariser to the mix, things get really weird.
How are people amazed at this? (or the 'trick') To anyone that passed secondary school basic science, duh...
It's a fun trick though, I had a run in with it just out of college. I had to inspect circuit boards overmoulded with a transparent material. This caused some issues near the injection port because the material was so hot that it washed away the nearby solder and caused little balls of solder to dissipate throughout the plastic. I was tasked with inspecting the outcome, but to improve release upon cooling they had sand-blasted the surface of the mould, the result was a quite effective diffusor which created a hazy surface. While x-ray worked somewhat, it wasn't ideal to find little balls of solder, so in the end out of sheer misery I tried a metallurgy microscope with different filters, the polarizer worked every single time because it neatly removed all the scattered light from the rough surface. What also worked surprisingly well was an immersion objective, but that tended to be quite messy.
I use the "andonstar polarizer film disc" bought on eBay and is very effective.
Essentially you use two discs for side Led illuminators and one for the objective. Once adjusted works like magic
Im a photographer. And every time I show people the effect a polarizing filter has on anything reflective like, windows, water, glass, random surfaces.. basically anything.
Their minds usually become... "BLOWN"
Polarising filters are a ton of fun to play with
You can get circular polarised film from those disposable 3D glasses
I've found many of your videos to be "illuminating," but this one takes the cake! Excellent lighting overview. Thank you.
Sticking a linear polarizer film to a ring light and looking through the middle should do the trick, too.
Also smartphones with non-OLED screens could be used as a makeshift polarized light source.
But wait, does that work or not? I'd have guessed that the polarizer on the lens needs to be at 90 degrees with respect to the polarizer on the light, so what you're describing here would be the worst possible case we saw while spinning the polarizer around?
@@TheHuesSciTech In my understanding you want only the light to enter your eyes that came from the same angle as you are watching. When you stick a polarizer sheet under the ring light it would first filter the light's light and on the way back only let this light return to your eyes. As result you are removing the light from surrounding light sources.
@@MetalheadAndNerd No, it would enhance reflection where polarisation is kept.
But the light from elsewhere will just be filtered as it comes through.
Ideally you want the only light source used to be polarised, or at least the only significant light source. And then you want the return path to be polarised, but in a manner where you can rotate the 2 with respect to each other to allow different polarisation through.
@@jeffreyblack666 Are you hoping that the edged surface changes the light's polarization in a different way than the non-edged surface?
@@MetalheadAndNerd Yes, as the video shows.
Based upon what happens it appears that the main chip surface preserves the polarisation, while the engraved surface does not.
If that is the case, it means without a polariser or with them in the same orientation, the chip surface appears bright and you can't easily see the markings. But when oriented at 90 degrees, the majority of the light reflecting off the chip is blocked as it is the wrong polarisation.
Reflected light is always polarized, therefore use a big mirror to reflect a light source once to create polarized light!
For me the most value part of this video is 8:12 : look how good the trace at the bottom(?) track is to see, on the right picture.
I use a cheap bottle of $1 water based acrylic white paint. Just barely wet your finger with the paint, touch the top of the chip then immediately wipe it off. The paint fills in the etched part numbers and you can then read them with a naked eye from any angle. If you want to remove the paint you can use a cotton swab with alcohol, though it doesn't harm anything and there is barely any paint on the part if you just wipe it back off. Never found a chip where it didn't work and takes seconds to do.
I have a DIY light for my microscope, it has four LEDs spread out and angled towards the middle and I have a little joystick which controls the relative brightness of them all. It means that you can control the direction of the lighting from the joystick.
This is usually called cross-polarized reflected light and is actually often used in macro photography. I've been using if for quite a while to discover otherwise hidden details in meteorites. The difference is indeed striking. The required investment is also insignificant.
Inking the chip with a marker reminds me of the way I mark torx bits. The sizes are stamped but hard to read, so I rub a white wax crayon on them. The wax stays in the groove and makes the numbers stand out.
My trick for reading chips is an angled point source light, like an led torch or my phone. Works great!
you should have done this episode with a 1458 op-amp
I use brake cleaner fluid. Sometimes if there is some grease in or on the laser engraved texts it realy helps to make it visible again. If there is grease on it, sometimes just changing ligthing direction doesn't help. And you just can't wipe that off from the surface, and it is changing light reflection from the engraved areas.
Different kind of polarization and also lighting conditions is very important when setting up machine vision cameras... (could be for reading these chippies on an assembly line too)
I saw that clip and thought about it for a minute and I was pretty sure it needed polarized light as well.
Thanks for the demo.
Thanks for shedding some light on this
If you are on the field and you have a hurry to identify a chip you can actually use a phone with a white image and polarizing sunglasses
High again Dave. I tried the marker trick you mentioned during this video. It worked really well. I used a red Staedtler marker, dabbed some ink all over the IC's surface, and was immediately able to see what I could barely make out with a magnifying glass. Thanks for this trick. It helped me a lot.
Totally worth getting two polarising filters (one on the light source and one just before the lens) if you are interested in mineralogy, geology, gemmology, cell biology, optical physics, materials sciences, seeing stresses in transparent or translucent materials, etc, etc. especially if you already have a good microscope to play with. In my experience the polarising filters from old lcds are pretty good especially if you can get a few with the same orientation (some screens actually have more than one per side) and you get them to stay flat.
You don’t have to use a transmitted light microscope (where the light goes through the sample) to see the effects most samples of anything which isn’t clear the whole way through will reflect enough light to give you a good look. If the samples are totally clear you can use a mirror under the sample angled very slightly so it doesn’t shine light straight into the lens. (Transmitted light microscopes have a condenser under the stage which focuses the light making it so only one light path goes through the same bit of sample into your eye which makes divisions like cell walls and stuff so they will not be nice thin lines without one)
Circular and linear polarising filters will actually produce different effects when looking at some materials.
I had to giggle. Photographers have known this trick for years. Cool Vid.
i was standing between two piers on a sunny day. my pixel could only pick up the reflections on the surface of the water so i held my polarized sunglasses in front of the camera and the shots of everything going on underwater were pretty amazing.
Awesome, I peeled some polarizing film from a broken LCD TV and have a camera polarizer.
On the other hand, it's well known that the best way to read those pesky barely-readable markings found on EPROM chips is taking a close-up photo of them with a digital camera that has a xenon-based flash. Accept no substitutes, a smartphone's LED flash won't work...
Very interesting ... and your answer is even better.
Really, this is all about glare. I bought a ring light with my microscope and was rather disappointed for 2 reasons:
1) It was WAY too dim. The dimmer knob was useless, since it always needed to be at Maximum. Perhaps I could have bought a more powerful one?
2) The ring produced a HUGE amount of glare, making it difficult to see and giving me headaches. This is because of the small diameter of the ring. Basically, the light goes almost straight down, and bounces straight back up. Horrible. This becomes worse as you raise the microscope. A larger ring would help ... but there's a practical limit to the size, and a higher cost.
My solution: I made a simple bracket, attached to the lens with a ring adapter, on which I placed 2 ordinary (100W equivalent) LED bulbs, which are offset 100mm on either side of the lens. The arrangement is similar to lights on a copying stand. Also, I added opaque reflectors, so the bright lights are not 'in my face.' I haven't bothered with dimming. I could, but it seems unnecessary. There is LOTS of light to see with, and the glare is completely gone. As you correctly suggest, reducing the glare is all about placing a light off to the side. I've never used my ring light again ... my solution works much better.
Any difficulty reading text, I just rotate the chip/board until it becomes visible.
Thank you Dave
Super fantastic! I use my smartphone (have no microscope) with torch on. Usually I take a picture and then enlarge it. But couldn't read the part numbers on a few mosfets. All I had to do was to take the picture from a sharp angle - thanks!
I work on a lot of boards coated with laquer or conformal coating type of gel and I find a uv light helps make the numbers glow....
I just put a drop of Isopropyl alcohol from my wash bottle. Works better than a sharpie.
Polarized light and and a filter are useful for inspecting faint and hard to see numbers. It's also good for eliminating solder glare in some of your videos Dave, so keep it on the camera for future use.
you got me good at the start with this one
Wow this is pretty interesting thanks for sharing Dave
He needs it because of the ring light. So you need cross polarization to block the reflections. You are using an off axis light so you can simply position your lights ource such that it is not producing an incident angle reflection on the chips. For real full put a piece of "transparent" plastic between the polarized film on the light and your polarizing filter. Rotate the polarizing filter until the light looks black and enjoy all the pretty colors of the light being bent by the birefringence of the plastic! With the two polarizers 90 deg opposed they block all the light (this was an old school way cinematographers would do a fade to black without any post processing... Two polarizing filters and you turn one during the shot.) The plastic alters the polarization of the light (differently at different frequencies) and makes pretty prismatic colors in the plastic.
Note that with a polarizing filter half the reflections on your solder joints and reflections on the edges of the raised tracks also disappear, and make the image better to judge. Downside is that you need double the light, so you could miss out on depth of field.
I used to use different colours of electrical tape over the lens of small LED torches to see different details of objects.
I rub white thermal grease on the chip then wipe it off. No scope required !
Side note: A thing which may be useful are alternate light sources, e.g. infrared and ultraviolet. I remember a clip a coupla years back where they had discovered that exploding caps left a residue easily detectable under UV light.
N.B. Protect your eyes and use appropriate filters to protect your scope or camera.
I've always just got in the habit of slightly tilting what I'm working on if I need to read a part number. Would love to eventually get a good adjustable polarizing light setup at some point, but what I have works well enough for pretty much everything.
If the object lay flat on the table, I usually turn my ring light off, put a light source (soft light is better) from six o'clock direction from the object, rise it up so the light falls about 45° to the object. It works quite well for me to read those laser markings.
“Quality Inspection lab requests that all chips be oriented in the same direction so they don’t need to adjust the polarising filters so much.”
Application of the magic spit is always handy
Anybody who spends a lot of time inspecting boards with these magnifiers, rarely puts the board down. Always tilting and moving it around
And this would work equally well in those conditions. Tilt of the board is completely irrelevant.
Whoa! Dual Daves!
I learned a very simple way, just rub some white thermal paste into the surface of the chip, it will penetrate into the etching and, well, make it whiter. =)
You win by 20 minutes.
- 12:45 You don't need a sharpie, licking your finger and rubbing it will often suffice. Sometimes I resort to smearing a bit of white thermal-grease into the etchings.
- 14:23 In my experience, polarizing film does _not_ "peel off nicely" (at least from TVs), they're pretty horrible and messy to get off. 😕
what helped me in the past is using a pencil and drawing some lines on the chip, the itched part rub the pencil more than the other part of the chip and its more visible..
Try looking at some random rocks (minerals) through that scope. ;)
These things have been common in geology forever.
Polarizers from old cameras are dirt-cheap if you don't care about size and mounting options and don't feel like taking apart old LCDs. Glue one to a cheap flashlight and you can do all kinds of fun things.
This is basically the same way you inspect optical components. You can't just shine a light on a lens from one direction and then decide that it's fine. You have to shift the light source or the lens around (whichever is easier) to illuminate things from different angles.
Love that 6809 reveal. Working on my homemade Color Computer project and that's the CPU for it. But of course I'm using the 63C09 instead because it supports faster speeds and has a few more features while being 100% backwards compatible.
if you want to test things with polarizers you can find them in a defective monitor .. just check the backplate reflective setup. made up of all kind of polarizing (at least two filters with 90° offset), reflecting and diffusing light filters to get a more even and straight back lighting for the pixels.. (and be careful in old monitors so you dont break the HV cfl tubes)
And this is generally why Pick N Place machines will have LED illumination at a few different angles to the part.
That lets them better differentiate between the main outline of the part and reflective "pits" etc.
Some fancy PnP machines will even read the part number, and compare it to the BOM.
It is quite magical using a polarizing filter on a camera, which you see the reflections disappear from water, windows etc.
You can try it on your own eyeballs, too. lol
TLDW: a polariser is only useful if you are using a ring light, because light firing straight down reflects straight back into the lens
I have a metalurgical microscope with a built in polarizer and it makes a big difference. Never though to us it to read chips.
Thank you for calling a torch a torch, Dave. It's the little things that matter! Fed up hearing about bloody flashlights. Never understood why a torch that doesn't flash is called a flashlight in some areas. Ah well...
Use ethanol (or isopropanol), this works better than a whiteboard marker. With a q-tip, the dust (basically old human skin) can be removed easily.
Revealed! Now the manufacturers will employ different scribing methods to make the polarizer useless.
I think that the difference is going to be much bigger on a cheap microscope and/or a black/white pcb.
Those pcbs are very hard to inspect due to reflection with the cheap microscopes
The best solution is to use 2 pairs of 3D glasses for theaters ("real D 3D") and only the right of left eye part. You put one (in reverse) in front of the light and one in front of the objective.
The one in front of the light must be used in reverse to create the right polarisation. The light is where your eyes are.
This is because the filter is not symmetric because it actually use a linear filter and a part that delays photons (quarter wave plate) depending on their polarisation.
Then you use the second one to capture the image.
I don't know how the photo filters are used in the video but maybe the one on the light is in the wrong way.
If you don't use the filter properly it's just doing linear polarisation, not circular!
Check how those filters are made to understand why both sides are different and why the one in front of the light must be used in reverse:
en.m.wikipedia.org/wiki/Polarizer#Circular_polarizers
The reason you need twice the same part of the 3D glasses is because you want to create for example right hand rotating waves that become left hand after specular reflection and filter that out with a left hand filter.
I didn't think this was a thing, but I always got away with my cell camera with the light on until the right angle is ok. Take the pic, zoom in et voila!!
Now look through that film with polarised light on some transparent plastic any bend that plastic. It's quite beautiful.
Can I just wear polarised sun glasses and turn my head? XD
I use my iPhone camera with the zoom at max, I tilt the board until I see the IC marking and take a picture. After that, I zoom in on the picture and I can see very well the marking.
Seems like a good technique to quickly visually inspect lots of solder joints without having to change the lighting setup multiple times.
Polarizers are expensive and a pain in the butt. You can get the same effect for reading chip labeling by shining a flashlight from the side onto the chip. Much simpler method and the result is pretty much the same.
this is great. i use an old smartphone as a microscope and it must have a polarized lens on it because iv no problems at all reading etchings on chips. but if i ever up grade to the real thing i'l know what to do. thank you Dave, & thank you to the guy who found this to work in the 1st place... VERY COOL/HANDY : )
Anglers have been taking advantage of this for decades...
Wow. Thank you!
A bit too late, but I absolutely love polarizing magic. Last year I bought some polarized sunglasses from a gas station, about $20, nothing fancy, and I'm not ever buying regular sunglasses again. They are so good that I rarely go out without them, mostly on very cloudy days or at night.
And I totally forgot I can get one for my camera too, since I am into photography a bit too...
Double D's starts of great from the start!
hahaha. very nice. make sense. how busy we are focusing our electronics stuffs forgetting some important things too.
Dave, that's great example of deceptive marketing, I'm positive someone noticed increased sales ^^ - glad You stopped it ^^
My microscope has a polarizer pair as a basic option... It's a medical/scientific microscope, though... It's great to see various transparent crystals inside cells...
I also had Polaroid sunglasses in the past... They made all the puddles on the road and windshields of cars almost invisible...
It will depend on the light source you are using. Led light has a different wave length of light to say the sun's light , so not likely to work well. Oh, your light source should be at 45 degrees to the board surface.👍
What a roller coaster, this.
Maybe the trick is to have a ring-light where the polarization changes with the angle (along the ring). Then a polarizing filter basically filters the incident-angle and you dont need to fiddle around with an additional light.
This is soooo cool. Thanks! 😍
I suspect the real use case for cross polarized light source/filter is when you've got chips that have been abraded or otherwise worked over to obscure the part numbers.
Will polarized light help in those cases?!
@@stefanmayer444 That is what i'd like to know
@@ka9dgx I'm not sure what methods are used in forensics to recover "removed stuff", but maybe there are some methods for that.
No chance. The abraded surface includes whateva was etched there less than 0.001" deep. Laser etching doesn't impress the material that may have an effect at depth of crystaline structures in the substrate material, like say serial numbers stamped on a metallic component that can be detected after careful polishing of the abraded surface.
I do the same, play with the light. Also my Eakins microscope has some nice software feature like WDR that is reducing light reflection.
Nice video, i need this!
"polarizer won't help unless you also have a matching light".. Absolutely, that is why photographers don't use a CP filters all the time. It's usually for sunset, or water bodies or sky, all of which scatter the sunlight which polarize the light. The CP then allows to filter out polarized panes of light allowing to show water reflection VS making it darker for example.
I remember seeing polarizer on microscopes to see smaller things in biology (number of years ago). Never quite understood how it worked but you could see smaller cell with it.
You had me at the beginning, thought I might need to go get one of them filters. Even trying to angle my light source sometimes I just have a hard time reading some chips. On occasion I've put some drops of isopropyl alcohol on the chips and that sometimes helps.
The so called circular polarization is the effect of two filters, a plane polarized filter (also historically called a Nicol), and a quarter-wave plate also called a retarder. The retarder effect is two shift the phase of one of the two orthogonal polarized images with a quarter of the full range of visible light.
The trick with the angle of illumination was marketed by Leitz in the Ultropak package. The image for difficult to obtain samples was indeed a game changer. An Ultropak would eat the guts of any microscope illumination system today.
What is happening when you move the external light is that you add light with polarized directions different from the ones your system was tuned to exclude. So when one buys polarization filters they should be always in pair. One filter to control the polarization of the incoming light and the other to control the outcome of the reflecting subject.