EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: dicky96 on January 05, 2020, 06:00:30 pm
-
Hi guys
This post follows on from a discussion I am having with several members on badcaps forum. So apologies to anyone who also follows that forum. In the end we thought the folks here may have a better idea if our idea is do-able.
We were discussing the practicality and desirability of having a HUD type display built into a binocular or trinocular microscope so that it would be possible to view meter readings and possibly oscilloscope wave forms while diagnosing faults on DUT such as laptop & desktop motherboards or mobile devices, without having to move your eyes from your carefully placed test probes.
We discussed several ideas from wearing smart glasses or some other sort of HUD when using the microscope, to having an OLED or other display integrated into the microscope or a modified eyepiece, or having a scanning laser or other device clipped on the microscope to project a small image directly onto the DUT itself (given that PCBs are relatively flat and reasonably reflective)
We briefly chatted about using blue tooth to get the data from a compatible multimeter onto the microscope display.
That's about as far as we got with it. But we are thinking you eevbloggers could get further ;)
So the questions are...
Is it a desirable/workable feature?
Is it feasible to do this, as an 'affordable plug in add-on' or upgrade to an existing stereo microscope such as the common Amscope and similar models that are all over aliexpress?
Has it already been done?
How would it best be implemented?
Is this a good kickstarter type project for someone?
Cheers
Rich
-
Transparent fluorescent coating that you excite with a UV laser + UV filter?
-
Hi everyone. One of the guys over at Badcaps involved in this discussion is me, so I'm following along here :)
-
https://www.lumineq.com/applications/optics (https://www.lumineq.com/applications/optics)
Could also project a LCD display on a reserved segment of the optics.
-
The biggest hurdle I see is where and how you'd install the display in the microscope as an add-on without damaging the scope. The trinocular port seems to make the most sense, except it would have to work in reverse. I'm no optics expert so I don't know what's going on in there exactly :|
-
Just make it an alternate eye piece.
-
Just saw the insides on one of these microscopes shown here *struggling to work out how to insert links*....is this right ?https://www.youtube.com/watch?v=EndB158l8Z8 (https://www.youtube.com/watch?v=EndB158l8Z8)
-
https://www.lumineq.com/applications/optics (https://www.lumineq.com/applications/optics)
That sort of thing looks like it would work but I'll bet it is also expensive
-
Yes, that looks like the most "professional" and viable option, but the most expensive too, no doubt... :(
-
Nikon D200 has something on this concept
https://www.imaging-resource.com/PRODS/D200/D200VIEWFINDER.HTM (https://www.imaging-resource.com/PRODS/D200/D200VIEWFINDER.HTM)
-
The D200 thingy says it is not really usable as a viewfinder so I wouldn't think it would be usable in place of a stereo microscope
To be honest we were hoping for some sort of home brew solution that could be implemented at a low cost so users of these sort of microscopes could add it for like 50-100 euros or something like that
-
A microscope that comes with this feature for sure will be a good seller, something like D200 solution but already on the microscope. I guess Dlp's will be involved in the solution
http://www.ti.com/dlp-chip/automotive/overview.html (http://www.ti.com/dlp-chip/automotive/overview.html)
-
Oh yeah, that's right: the viewfinder on SLRs is another good way of describing this. I forgot I had one at my old shop which had this in its viewfinder. It could work since you'd only need to display only a few characters at a time, provided you found a way to make it into an add-on as previously discussed, either inside the body somehow so it shows up in both eyepieces or inside the eyepiece itself.
-
If you look at the way the binocular microscope is built, it has two completely separate optical paths, one for each eye. The only time both go through a single lens is where the objective lens is fitted at the bottom (Barlow Lens)
However if the HUD was only present in one eye, wouldn't your brain just compensate for that and you would see it normally anyway? It's not like you depth of vision for the readout itself.
I am also convinced such a device (a microscope that had this feature, or even better an add-on to existing microscopes) would be a very good seller as I doubt it is very expensive to do it. I wish i had the skills, knowledge and time myself. Surely someone would like to take this project on?
Rich
-
My 50 cents:
I don't think it's an easy issue. The D200 viewfinder example is a perfectly controlled implementation and, in optical terms, very different than looking at a bi-trinocular microscope.
It would be very difficult to make something DIY that could be adapted to several equipments...
What about smartglasses/optical head-mounted display?
https://en.wikipedia.org/wiki/Optical_head-mounted_display (https://en.wikipedia.org/wiki/Optical_head-mounted_display)
-
Yes, I agree it may not be easy overall. I was thinking an OLED display and somehow "projecting" it into the optical path. I think the only way to achieve this is with a prism (or a semi-reflective piece of glass like a teleprompter), since you HAVE to combine both the image from the display and the one which normally comes through the microscope, so you'd have to "insert" it somewhere, that's why I imagine the third port as a good place to start experimenting before taking stuff apart....just an idea. If I had a microscope myself, I would have definitely looked into it more...
-
I suppose you could put a bright, order 10mm diameter display at the intermediate image plane of the photo port: to project a focused image on to a flat specimen. That way, dust and such on the display would not noticeably degrade the specimen image.
If the specimen isn't flat, you likely want to project the image the other way, to the eyepiece instead of the specimen, so that you are essentially viewing two superimposed specimens. Unfortunately, most microscope systems assume just a single specimen with possibly two or more viewers. I wonder if you could find some sort of optical comparator.
-
I imagine an oled or LCD at right angles to the optic path with an angled piece of flat glass to reflect the displayed image into the eyepiece, kind of like a teleprompter or a Peppers Ghost type effect.
Oled would work well for this as the black is truely black so it won't wash out the image of the PCB, although a neutral density filter would probably help when using an LCD anyway.
The trick will be getting the optics right so both the image from the display and the thing you are looking at are both in focus at the same time.
I suppose it would be easiest to keep both focus mechanisms seperate as the display will never need to be refocussed once it is set. It will then always remain in focus while the thing you are looking at can be brought into and out of focus in the mormal way.
With the correct lenses, you can use a larger (hence cheaper) display, maybe a high DPI iphone display on a breakout board, maybe HDMI or similar to a Raspi or PC running some software to mirror the scope/Test equipment display.
Stick the display in a lightproof enclosure with the correct lenses then attach it to the microscope with that bit of angled glass in the right spot in the original optical path.
I have absolutely no idea about figuring out optics though..... good luck!
-
What about my idea of using a low powered laser and voice coil operated mirrors to produce a scan or raster (like the animation lasers I used when I was using when running mobile discos - some of those you could type something on a keyboard project it onto a wall or even the dancefloor).
The laser and mirror assembly could be clipped to the microscope casing similar to how the LED ring light fits (hey, it could even be incorporated into an LED ring light!), and project a small image onto the PCB under test. You could manually swivel the laser assembly to move the image wherever you want in the field of view. It wouldn't matter if it was projecting from an angle and causing some key-stoning of the image. You would still be able to read it. A graticule and oscilloscope wave form should be easy to project as well as a meter reading.
The PCB under test should be flat and reflective enough to make the displayed image clear enough to read. You could use a diffuser on the laser which would actually make the laser image easier to read and keep it safe to the eye if it was reflecting up the microscope at any time
As it is a laser it does not require focusing.
That sort of addon would potentially fit to any microscope and require no modification of the optics.
-
Just saw the insides on one of these microscopes shown here *struggling to work out how to insert links*....is this right....
The link looks good to me, but you probably do not want that fellow cleaning your microscope. You can actually hear him scrubbing glass. :(
-
If you don't mind looking at an LCD monitor, you could feed everything you want on a PC and display it easy enough. Maybe some better probe and board holders would help.
-
Some googling later .....
It looks like various groups have tried this before:
Novel Microscope-Integrated Stereoscopic Heads-up Display for Intrasurgical OCT in Ophthalmic Surgery
https://iovs.arvojournals.org/article.aspx?articleid=2333389 (https://iovs.arvojournals.org/article.aspx?articleid=2333389)
(https://arvo.silverchair-cdn.com/arvo/content_public/journal/iovs/934157/3514_fig01.gif?Expires=1578381334&Signature=xToBDQBjKQmufNacFBS7YkNQjWECbrCnLpaYshTfgh1B7rrzeVPTXz9tMMRdPL~oT9wugSdTZh8NxAsL5Cdw35oHG5zLVAEDT2fOt2DaFu4tGy2o2JWAkJP7s0SgMKeE-0-QpWNPhqxuDQwMobbS6tAEITBClWHj4T4bw~~Z7q8r-kSzxZxa5a4ojVeCi7fCNyd0yAwtUnQctJ2XSmjQ0clthw0LIYVBRRK0LQOZEbggfWhCaDayl~DAJGmkgrNjq~V8q3xZK-hlyVVtITflOZD~2pKgvgSJnySHGFA~kECOPe8Au6-W6xNt~~1D7et2fZgPvpZ93IHfygY3Xk8tJA__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
They insert a beamsplitter for each eyepath before the collimated beams get to the eyepiece optics and add a lensed image from an OLED display.
Or you could add an external single beamsplitter/45degree glass reflector to an OLED/etc at the same working distance from the objective lens as the PCB/DUT.
e.g.
(https://i.imgur.com/t6qCvRu.jpg?1)
rt
-
Or you could add an external single beamsplitter/45degree glass reflector to an OLED/etc at the same working distance from the objective lens as the PCB/DUT.
e.g.
(https://i.imgur.com/t6qCvRu.jpg?1)
rt
Nice idea. That looks like it could become something workable.......
-
It's a shame there is no open source solution for autostereoscopic eye tracked 3d displays. It's too boutique for any proprietary solution to ever be affordable , but hardware wise it's all cheap and COTS.
-
Nice idea. That looks like it could become something workable.......
Yes, nice idea. BUT, the (hands/tools) working space becomes a big problem.
-
Well yes I thought that, but does the beam splitter have to be at 45 degrees and the projector at 90 degrees to the microscope
Wouldn't it still work if the projector was fitted almost vertically to the microscope, pointed downwards and angled inwards towards the beam splitter by say 20 degrees and the beam splitter mounted close to the objective lens and tilted at 10 degrees in the opposite direction? We only need the meter/oscilloscope readout near one edge of the field of view anyway. The projector could also be behind the microscope and not in front of it
I probably have my angles all mixed up there but hopefully someone clever with optics can figure out what I mean.
Why does the beam splitter need to be the full size of the objective. Can't it be small and to one edge?
At the moment a 0.5x barlow lens is giving about 10cm working distance, losing one or 2cm wouldn't cause a problem, especially if it was just at one edge of the objective
Also I noticed once I set the focus, I can go all the way from maximum to minimum zoom without having to adjust it again. So the meter reading should also stay in focus all the time as well. The projector would not need to be effectively at the same distance from the objective as the DUT. Just somewhere in the range of distance between minimum and maximum zoom
I think the idea is workable and would only lose a small amount of working distance
-
Some googling later .....
It looks like various groups have tried this before:
Novel Microscope-Integrated Stereoscopic Heads-up Display for Intrasurgical OCT in Ophthalmic Surgery
https://iovs.arvojournals.org/article.aspx?articleid=2333389 (https://iovs.arvojournals.org/article.aspx?articleid=2333389)
(https://arvo.silverchair-cdn.com/arvo/content_public/journal/iovs/934157/3514_fig01.gif?Expires=1578381334&Signature=xToBDQBjKQmufNacFBS7YkNQjWECbrCnLpaYshTfgh1B7rrzeVPTXz9tMMRdPL~oT9wugSdTZh8NxAsL5Cdw35oHG5zLVAEDT2fOt2DaFu4tGy2o2JWAkJP7s0SgMKeE-0-QpWNPhqxuDQwMobbS6tAEITBClWHj4T4bw~~Z7q8r-kSzxZxa5a4ojVeCi7fCNyd0yAwtUnQctJ2XSmjQ0clthw0LIYVBRRK0LQOZEbggfWhCaDayl~DAJGmkgrNjq~V8q3xZK-hlyVVtITflOZD~2pKgvgSJnySHGFA~kECOPe8Au6-W6xNt~~1D7et2fZgPvpZ93IHfygY3Xk8tJA__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
They insert a beamsplitter for each eyepath before the collimated beams get to the eyepiece optics and add a lensed image from an OLED display.
I like this one. It comes close to my original idea actually. Not sure how easy something like this would be to achieve with Aliexpress parts :D
-
It will be a challenge to do it that way with a microscope that has a fixed tube length, like the Greenough-design Amscope the fellow cleans in the above video. Inserting the beam splitter would lengthen the microscope's optical path. You would also need to insert carefully designed twin relay lenses to make the microscope parfocal again.
The beamsplitting HUD was inserted into the infinity (collimated) optical path of a higher end, common main objective (CMO) microscope that was designed to support and simplify such add ons.
-
I suppose you could put a bright, order 10mm diameter display at the intermediate image plane of the photo port: to project a focused image on to a flat specimen. That way, dust and such on the display would not noticeably degrade the specimen image.
Can you translate that into something the layman can understand?
-
I will try. The photo port projects a real image which you can cast and see on a screen. If Amscope or whoever has done their job and made the microscope "parfocal", the port will focus the image at some constant distance above the photo port flange. If you are fortunate, you can cast this "intermediate image" directly onto a camera sensor and Bob's your uncle. Otherwise, you may need a "relay lens" to receive the intermediate image and re-project it onto the sensor, and possibly resize and correct it.
The optics don't care what side the light goes into the microscope and what side it comes out – and work just as well in reverse. If you put a bright LCD where the intermediate image would be, the optics will project an image downward, through the zoom unit and objective – and focus it on the specimen.
-
Not sure I really understand why something like this is needed and would find it really distracting and would compromise the optics.
However, each to their own, but I would think it easier to do the HUD display with a pair of glasses that are worn when the HUD is wanted.
-
@jfiresto
So if I understand that right you could project an image from a multi meter etc down the camera port of a trinocular microscope, reflect it off the PCB under test and view the superimposed image on the DUT through the binocular eyepieces? DannyX wondered if that would be possible.....
-
Nikon D200 has something on this concept
https://www.imaging-resource.com/PRODS/D200/D200VIEWFINDER.HTM (https://www.imaging-resource.com/PRODS/D200/D200VIEWFINDER.HTM)
Not just the D200, but rather most SLR cameras (analog or digital) by any major camera maker released since the late 1980s or so. It’s simply a transparent LCD in the viewfinder assembly, by the focusing screen: https://photo.stackexchange.com/questions/12200/how-are-the-red-focus-point-indicators-displayed-on-a-dslrs-focusing-screen (https://photo.stackexchange.com/questions/12200/how-are-the-red-focus-point-indicators-displayed-on-a-dslrs-focusing-screen)
-
So if I understand that right you could project an image from a multi meter etc down the camera port of a trinocular microscope, reflect it off the PCB under test and view the superimposed image on the DUT through the binocular eyepieces?...
That's right. Here I have stuck an LED into the photo tube of a stereo microscope, so that the illuminated die is at the tube's intermediate image plane:
[attachimg=1]
The plane is about 12mm(?) into the rather old photo tube; later tubes project the image above and outside the tube to simplify the optics between it and the camera. This photo tube steals the view from the left eyepiece.
Here is the view through the right eyepiece at 16X magnification, with the lit up die projected onto the circuit board:
[attachimg=2]
You can see the square bond in the middle and the fringes of the clear lens above it. I held a 2 megapixel camera above the eyepiece to take the picture and was not still enough or precise enough or both to get a really clear picture. You can actually see the backlit bonding wire and more detail when you peer through the eyepiece.
-
Great ! That's what I like to see: practical stuff. Thanks for that ! :D
-
You are welcome and I wish you much success!
-
You are welcome and I wish you much success!
I'll try on mine tomorrow to see if that would work on a simulfocal trinocular
-
Meaning on one side, 50% goes to the camera and 50% to the eyepiece? That should work and appear in both eyepieces, although more dimly.
-
I will be curious how things work out with the "simulfocal" photo tube. A 50:50 beam splitter will reduce the brightness of the viewed LCD image more than four-fold. If you have to really brighten the LCD to make it visible, it may excite reflections within the optics that cloud the view, since the light from the LCD must pass through the microscope, twice.
Wild Heerbrugg, who made the microscope I tested, minimized internal reflections as much as was possible, to support coaxial illumination (but perhaps not within the photo tube). Even then, Wild resorted to polarized light and a quarter-wave plate between the objective and specimen, to win as much contrast as possible. I hope Amscope took as much care!
-
Amscope are pretty much $200 OEM (with double boom stand) Chinese imports, so doubtful.
-
OK I tried with a green 5mm LED. The camera fits onto the tube with the sensor above the end of the tube so in my case the image point or whatever you called it is projected above the tube
When I held a green LED there I got a bright green wash but it wasn't focused, though if I moved the LED up and down slightly I could get a brighter green blob in the centre of the wash.
I then closed each eye in turn and discovered that the bright green wash was only in the left lens. In the right lens I got a dim but focused image of the LED reflected off the workbench.
My microscope is not Amscope, it is actually this one, though I got a considerable discount on the price when I bought it on Singles Day 11.11
https://www.aliexpress.com/item/32928742330.html?spm=a2g0s.9042311.0.0.27424c4dabYU1M (https://www.aliexpress.com/item/32928742330.html?spm=a2g0s.9042311.0.0.27424c4dabYU1M)
-
Rats. I wonder if the reflection happens quite early and involves the beam splitter.
Suppose the microscope is on its side so that the objective supplies 1, the eyepiece receives 2 and the camera port receives 3:
(https://upload.wikimedia.org/wikipedia/commons/1/1f/Beamsplitter-1.png)
If the LED supplies light to 3, 50% will reflect to 1 (the objective) and 50% will pass through and out the top of the beam splitter. If a surface above the splitter reflects light back, 50% of that will then pass through to 3 (the LED) and 50% will reflect to 2 (the eyepiece).
I will shine light into another microscope that has a 50:50 beam splitter and see what happens.
-
I shined an LED into a 50:50 photo tube, so that it illuminated a PCB. I could see, superimpose and sharply focus both the PCB and LED in the photo tube side eyepiece, but the LED had to be so bright that the microscope's internal reflections noticeably fog the PCB image and reduce its contrast.
I could considerably reduce the fogging by shining in the LED at an angle. I suppose if you used a bright and really high resolution display, you could angle it and feed it a distorted, compensating image. In the end, however, I think displaying graphics and text through a 50:50 photo tube will be less or less than satisfactory.
If the overlayed text and graphics only need to block the PCB image, as with an eyepiece reticle, could you use a transparent LCD as the reticle inside an eyepiece?
EDIT: Struck the text of an irreproducible comment.
-
Hmm well if your reflection theory of how the light reflects at the beam splitter is correct then if I move the LED much further away from the camera port then maybe I can get a focused image in that eyepeice?
Also maybe be if I focus the LED through another objective lens and down the camera port then that would work and give a sharp bright image in that eye? I have a spare objective because the microscope came with a 0.5x and a 2.0x Barlow lens and i am currently using the 0.5x on the microscope
One thing I did notice, I could not tell the green was only visible in one eye until I shut that eye.
I guess for test the above idea the 2x Barlow should be mounted at the same distance from the beam splitter as the 0.5x objective is? And then the LED mounted at 25% of the focal length of the 0.5x objective (which I can measure)? Or I could fit the 2.0x objective to the microscope and measure the working distance compared with the 0.5x so I would know how far away to put the LED?
As you understand the maths of this much better than me could you suggest some useful distances at which I could check this out using the spare 2.0x Barlow lens and an LED?
Cheers
I could experiment with some cardboard tube from a kitchen roll?
-
OK i had a bit of time today so I had a play around with the LED and the 2.0x objective lens but I didn't get anywhere. I couldn't get the LED to produce any sort of image. :(
-
If you project the image onto the what ever it is that you are working on, it seems like you would constantly be trying to find a flat spot to where you could read it.
The HUD glasses may be alright if they make one that would still not get in the way of using the microscope. Is there such a product? I have not looked.
Not sure I really understand why something like this is needed and would find it really distracting and would compromise the optics.
However, each to their own, but I would think it easier to do the HUD display with a pair of glasses that are worn when the HUD is wanted.
Fully agree that any distortion would make it worthless. I have been in situations due to old aging eyes, where I will be wanting to probe a signal/s by hand and use the microscope to position them. I then look away trying not to move my hands. I can work faster this way than by using a probe holder but I've also slipped.... I could see it being useful but I'm not so sure it would be better than having a fast, high-res camera and doing it digitally.
-
I could see it being useful but I'm not so sure it would be better than having a fast, high-res camera and doing it digitally.
But then you would lose the stereoscopic vision which (for me at least) makes a microscope practically useless apart from inspection use.
-
I could see it being useful but I'm not so sure it would be better than having a fast, high-res camera and doing it digitally.
But then you would lose the stereoscopic vision which (for me at least) makes a microscope practically useless apart from inspection use.
That is true, but I can't think of a time where having the stereo was all that important to me when I was just trying to probe a board. One problem I can think of that really limits my use for it is that normally when I am free probing a board under the microscope, it's with the scope. Normally, I will need to change settings and need to look away to do it. If I were buzzing out a board with a meter I just need to hear the beep, so it doesn't help me in that case.
Maybe there are uses that relate to electronics that I am not thinking of. Maybe if I had it, I would find uses for it outside of what I mentioned.
******
I should add as well that a lot of times when I am using the scope, I am using the readouts. I'm not sure that even if the image were crystal clear that I could actually read it.
-
Another reason I liked the idea of the HUD is because I wear glasses and have to take them off to use the microscope (though I am going to buy some of the high eyepoint eyepieces to see if I can use it with my glasses)
So I am looking through the microscope - I have my test probe in the right place, I look up - but I can't possibly read my meter or scope without my glasses. So I need to put them on. With one hand while keeping the probe in the correct place on a 0.5mm qfp pin with the other hand and taking care not to short anything....... hmmmm
-
This is where some decent probe holders come in. A few years back someone had posted about their homemade probe holder that had a camera mounted to it. It was a fairly simple setup made from wood. Something like that may help you out.
-
This one?
https://www.eevblog.com/forum/testgear/oscilloscope-probe-holder-with-vision/ (https://www.eevblog.com/forum/testgear/oscilloscope-probe-holder-with-vision/)
-
Interesting idea, I like it, but it will only work on flat, even boards. If you've got stuff like heatsinks or large components that elevate the board from the table in any way, it's useless...just today I wrestled with a board out of an active speaker set which had a heatsink on one side for the TDA ICs, but nothing on the opposite end, so it sat at a 45 degree angle all the time, making it a nightmare to work on....
-
This one?
https://www.eevblog.com/forum/testgear/oscilloscope-probe-holder-with-vision/ (https://www.eevblog.com/forum/testgear/oscilloscope-probe-holder-with-vision/)
Yes, that's the one. I ended up picking up a couple of these low cost cameras. Turned out they were made by the same mfg, just supplied with different packaging and software. I ended up using one for some thermal testing. They were certainly cheap enough to integrate into some sort of holder.
Interesting idea, I like it, but it will only work on flat, even boards. If you've got stuff like heatsinks or large components that elevate the board from the table in any way, it's useless...just today I wrestled with a board out of an active speaker set which had a heatsink on one side for the TDA ICs, but nothing on the opposite end, so it sat at a 45 degree angle all the time, making it a nightmare to work on....
I made a probe holder out of one of these from Harbor Frieght:
https://www.harborfreight.com/hand-tools/measuring-marking-layout/calipers-micrometer/clamping-dial-indicator-63656.html (https://www.harborfreight.com/hand-tools/measuring-marking-layout/calipers-micrometer/clamping-dial-indicator-63656.html)
This has worked out fairly well. I have thought about making a stage that the probe and camera would mount to.
-
Interesting idea, I like it, but it will only work on flat, even boards. If you've got stuff like heatsinks or large components that elevate the board from the table in any way, it's useless...just today I wrestled with a board out of an active speaker set which had a heatsink on one side for the TDA ICs, but nothing on the opposite end, so it sat at a 45 degree angle all the time, making it a nightmare to work on....
In that sort of situation it's generally easiest to solder a short piece of wire to whatever point you want to probe if it is not easy to get to or you are at risk of slipping. You wouldn't get (or need to get) a board like that under a microscope either. Though actually you could with the double boom stand as you can also rotate the microscope head to the required angle
-
These guys have a multimeter HUD for microscopes. Looks promising!
https://www.clinicadelamanzana.com/1i-hud (https://www.clinicadelamanzana.com/1i-hud)