AmScope 3.5X-45X Trinocular Zoom Microscope with Double Arm Boom Stand

[cowasaki]

Well after a lot of looking I have finally decided on the AmScope 3.5X-45X Trinocular Stereo Zoom Microscope with Double Arm Boom Stand. 

This is a link to the same model but with x90, LED light and 5MP camera.

https://www.ebay.co.uk/itm/AmScope-3-5X-90X-Circuit-Zoom-Stereo-Microscope-144-LED-5MP-Camera/401657708141

Does anyone have one of these setups?  Any suggestions re models before I hit buy in the morning? 

There is a vast difference in the LED lighting systems, would the cheap £23 model be adequate at least for the moment?

As for camera, their cameras are VERY expensive compared with others I have found on eBay.  Has anyone experience regarding alternative cameras?

Thanks in advance.

[Howardlong]

The problem with the LED ring lights is that they get in the way of some tools in particular hot air desoldering tools. I don’t use the LED ring as a result, instead I use a pair of gooseneck 3W spots clipped onto the microscope itself.

[cowasaki]

The problem with the LED ring lights is that they get in the way of some tools in particular hot air desoldering tools. I don’t use the LED ring as a result, instead I use a pair of gooseneck 3W spots clipped onto the microscope itself.

Good point....

I've only just found that you can't use the third port AND both eyepieces!  Not happy....

[Howardlong]

Been there, done that. In the end I bought the "proper" three port simul focal job, the trinocular head sits in a cupboard now. I feel it should be made a lot clearer: trinocular is fine if you only want stills, but useless if you're making videos.

These are the gooseneck LED lamps I used: https://www.amazon.co.uk/gp/product/B00H1O1ZF2 They're unavailable now, but you can probably find them elsewhere. Importantly, the beamwidth is only about 30 degrees: most small LED worklights are much wider.

Edit: to be clear you really need at least two light sources, each from a different angle. Both of the gooeneck jobs I have are clipped to the microscope's head so they also move as the head moves.

[cowasaki]

Been there, done that. In the end I bought the "proper" three port simul focal job. I feel it should be made a lot clearer: trinocular is fine if you only want stills, but useless if you're making videos.

These are the gooseneck LED lamps I used: https://www.amazon.co.uk/gp/product/B00H1O1ZF2 They're unavailable now, but you can probably find them elsewhere. Importantly, the beamwidth is only about 30 degrees: most small LED work lights are much wider.

Thanks.  I'm looking on Amazon now. The import duty and postage cost half as much as the whole scope!

I've found the x7-x45 simul-focus model at £559.99 which is a lot more than I expected to pay originally but I am not going to compromise with that! I'm going to need a 0.5x barlow as well which is another £28

[sn4k3]

Light: https://www.eevblog.com/forum/reviews/(review)-polarized-led-ring-light-hdr-70w-mp-for-microscopes/msg1830350/#msg1830350
Camera: https://www.aliexpress.com/store/product/4K-Ultra-HD-60fps-HDMI-output-industrial-microscope-camera-for-SONY-imx226-sensor/1918400_32906410173.html

[cowasaki]

Light: https://www.eevblog.com/forum/reviews/(review)-polarized-led-ring-light-hdr-70w-mp-for-microscopes/msg1830350/#msg1830350
Camera: https://www.aliexpress.com/store/product/4K-Ultra-HD-60fps-HDMI-output-industrial-microscope-camera-for-SONY-imx226-sensor/1918400_32906410173.html

How good is that camera? 

[sn4k3]

I don't have it, but i seen video sample, it very good, the only drawback is lack of pc connection using USB
I have this one: https://www.aliexpress.com/store/product/1080p-HDMI-USB-SD-digital-microscope-camera-with-SONY-sensor/1918400_32464609452.html

They are good to record but not live work, for that you need high framerate camera.

My camera + polarized led ring sample:


So my recomendation is to get microscope separate, and buy camera + led ring of your chooise, the included camera is not good and you also need a good adaptor like these: https://www.aliexpress.com/store/product/0-35X-focus-adjustable-C-mount-adapter-for-New-type-of-Amscope-s-Trinocular-Stereo-microscope/1918400_32830788378.html?spm=2114.12010612.8148356.7.49bd426fOkYljn

[digsys]

I'm using that exact make / model for years now .. my take -
The ring Leds are fine for most jobs, and they don't "clutter" my bench. IF I need any other type of lighting, I have several different options -eg clip-ons
Yup, the camera mount is a complete joke, but I occasionally take a still shot or two. Thankfully, never needed to do a 3d video yet :-)
The unit itself has worked perfectly, with no degradation at all, especially mechanical wearing etc

[cowasaki]

Well I've just ordered it using Amazon Prime so I get it on Friday £559 plus the barlow lens off ebay for £28.  That's £150 more than I expected and budgeted for BUT I really didn't want to lose the vision in one eye just because I am recording it or taking images.  That 4K camera looks quite good.  I would like something I can watch it real time and record as well.

[Howardlong]

0.5x Barlow is what I use permanently, and I go down to silly sizes, 008004 and 0.35mm pitch.

Although the eye optics are fine, the optics in the camera port aren't all that great I don't think, but that may be because I need to spend more time setting up. I use a C mount port adapter, plus a Lumix MFT camera with C mount/MFT adapter. No point going more that 1080p due to my perception of the camera port optics, particularly when zoomed in, even though I do have 4k capability.

You can see some examples in the overhead shots here:





The view you see here is cropped to about half what I see through the eyepieces: you end up with very significant vignetting without cropping.

[TERRA Operative]

I got one of these, sans camera. I got a couple barlow lenses and the setup works well.

I'd also be interested in any info for a camera, particularly a 25 or 30fps @ 1080P capable one.

[janoc]

If you are in the UK, you can order the AMScope scope from this company, saves you the import duties and shipping costs (microscopes cost fortune to ship - big, fragile, heavy):

https://www.gtvision.co.uk/

They sell also on eBay. I have bought a microscope from them some time ago and very happy, first class service. 

[jfiresto]

... The optics in the camera port aren't all that great I don't think, but ... I [may] need to spend more time setting up. I use a C mount port adapter, plus a Lumix MFT camera with C mount/MFT adapter... The view you see ... is cropped to about half what I see through the eyepieces: you end up with very significant vignetting without cropping.

I like that arrangement, but where is the strong vignetting coming from? Does either adapter have reduction optics?

I would expect direct projection onto a MFT sensor should be a very good match, with just a bit of vignetting at the corners. If you are directly projecting, is the camera sensor at the port's flange focal distance? Does the port stay in focus with the eyepieces?

[RoGeorge]

Any suggestions re models before I hit buy in the morning?

I'll go for a simulfocal version.  Those will let you film while looking with both eyes.  If it's not simulfocal, then one eye is switchable between eyepiece or camera.

Also, the brand (AmScope) is not really necessary.  Usually all are just Indian or Chinese generic microscopes with a label on them.  There are not so many models, so if you look carefully at the pictures, you can order the same microscope with a different brand, but significantly cheaper.

[cowasaki]

Any suggestions re models before I hit buy in the morning?

I'll go for a simulfocal version.  Those will let you film while looking with both eyes.  If it's not simulfocal, then one eye is switchable between eyepiece or camera.

I’ve bought a simulfocus model with 7-45x. I’ve then bought a 0.5 Barlow from a different seller as it was half price. I will look at illumination and the camer once I get it. If it will take a DSLR and I can find one that does video then that might be nice. Otherwise I am still open to suggestions.

[KL27x]

OP, I like mine very much. But FYI, in case you don't know.

3.5x-90x zoom means it comes with a 0.5 Barlow and a 2x Barlow.

The base unit is a 7-45x microscope with about a 4-5" distance between the object and the bottom of the lens.

With the 0.5 Barlow attached, it is a 3.5-22.5x zoom with about 6" of clearance

With the 2x Barlow attached, it is a 14-90x microscope with a 2"clearance.

You will maybe never use a 2x Barlow for electronics work.

[cowasaki]

OP, I like mine very much. But FYI, in case you don't know.

3.5x-90x zoom means it comes with a 0.5 Barlow and a 2x Barlow.

The base unit is a 7-45x microscope with about a 4-5" distance between the object and the bottom of the lens.

With the 0.5 Barlow attached, it is a 3.5-22.5x zoom with about 6" of clearance

With the 2x Barlow attached, it is a 14-90x microscope with a 2"clearance.

You will maybe never use a 2x Barlow for electronics work.

If you see my post directly before yours I had worked that out eventually. Why they cause confusion for their customers is beyond me. Why not quote the base model with the cost of each upgrade or accessory. It would make sense.

[Howardlong]

... The optics in the camera port aren't all that great I don't think, but ... I [may] need to spend more time setting up. I use a C mount port adapter, plus a Lumix MFT camera with C mount/MFT adapter... The view you see ... is cropped to about half what I see through the eyepieces: you end up with very significant vignetting without cropping.

I like that arrangement, but where is the strong vignetting coming from? Does either adapter have reduction optics?

I would expect direct projection onto a MFT sensor should be a very good match, with just a bit of vignetting at the corners. If you are directly projecting, is the camera sensor at the port's flange focal distance? Does the port stay in focus with the eyepieces?

The port's C mount adapter has optics (0.5x), and I suspect that's where the image degradation comes from. The C mount to MFT has no optics and is this https://www.amazon.co.uk/gp/product/B007XAJAOC . If I use the full sensor with the adapters, then I get the full circle in frame. With a native 1080p sensor crop it's a 2.7x crop (aka "Ex. Tele Conv") with the GX9 camera. I've just realised why I was previously using an older G7: it had a sensor crop of 2.4x at native 1080p sensor crop due to its 16M pixel sensor (vs 20.3M pixel for the GX9).

Following your post, you've re-peaked my interest in resolving this, and I just had a very brief five minutes to revisit it. If I deconstruct the microscope's camera head to some degree, then without any adapters, optical or otherwise, I can get a fairly reasonable 1x crop picture but I'll need to experiment more and fabricate some sort of adjustable enclosed adapter tube.

[cowasaki]

... The optics in the camera port aren't all that great I don't think, but ... I [may] need to spend more time setting up. I use a C mount port adapter, plus a Lumix MFT camera with C mount/MFT adapter... The view you see ... is cropped to about half what I see through the eyepieces: you end up with very significant vignetting without cropping.

I like that arrangement, but where is the strong vignetting coming from? Does either adapter have reduction optics?

I would expect direct projection onto a MFT sensor should be a very good match, with just a bit of vignetting at the corners. If you are directly projecting, is the camera sensor at the port's flange focal distance? Does the port stay in focus with the eyepieces?

The port's C mount adapter has optics (0.5x), and I suspect that's where the image degradation comes from. The C mount to MFT has no optics and is this https://www.amazon.co.uk/gp/product/B007XAJAOC . If I use the full sensor with the adapters, then I get the full circle in frame. With a native 1080p sensor crop it's a 2.7x crop (aka "Ex. Tele Conv") with the GX9 camera. I've just realised why I was previously using an older G7: it had a sensor crop of 2.4x at native 1080p sensor crop due to its 16M pixel sensor (vs 20.3M pixel for the GX9).

Following your post, you've re-peaked my interest in resolving this, and I just had a very brief five minutes to revisit it. If I deconstruct the microscope's camera head to some degree, then without any adapters, optical or otherwise, I can get a fairly reasonable 1x crop picture but I'll need to experiment more and fabricate some sort of adjustable enclosed adapter tube.

What do you need? It might be a simple job with fusion 360 and the 3D printer.

[Howardlong]

... The optics in the camera port aren't all that great I don't think, but ... I [may] need to spend more time setting up. I use a C mount port adapter, plus a Lumix MFT camera with C mount/MFT adapter... The view you see ... is cropped to about half what I see through the eyepieces: you end up with very significant vignetting without cropping.

I like that arrangement, but where is the strong vignetting coming from? Does either adapter have reduction optics?

I would expect direct projection onto a MFT sensor should be a very good match, with just a bit of vignetting at the corners. If you are directly projecting, is the camera sensor at the port's flange focal distance? Does the port stay in focus with the eyepieces?

The port's C mount adapter has optics (0.5x), and I suspect that's where the image degradation comes from. The C mount to MFT has no optics and is this https://www.amazon.co.uk/gp/product/B007XAJAOC . If I use the full sensor with the adapters, then I get the full circle in frame. With a native 1080p sensor crop it's a 2.7x crop (aka "Ex. Tele Conv") with the GX9 camera. I've just realised why I was previously using an older G7: it had a sensor crop of 2.4x at native 1080p sensor crop due to its 16M pixel sensor (vs 20.3M pixel for the GX9).

Following your post, you've re-peaked my interest in resolving this, and I just had a very brief five minutes to revisit it. If I deconstruct the microscope's camera head to some degree, then without any adapters, optical or otherwise, I can get a fairly reasonable 1x crop picture but I'll need to experiment more and fabricate some sort of adjustable enclosed adapter tube.

What do you need? It might be a simple job with fusion 360 and the 3D printer.

Indeed, my 3d printer has been gathering dust for about five years after I never used it!

[jfiresto]

Following your post, you've re-peaked my interest in resolving this, and I just had a very brief five minutes to revisit it. If I deconstruct the microscope's camera head to some degree, then without any adapters, optical or otherwise, I can get a fairly reasonable 1x crop picture but I'll need to experiment more and fabricate some sort of adjustable enclosed adapter tube.

As encouragement, here is what I get with direct projection onto the sensor of a 12 megapixel Olympus E-PL1 camera with 0.7X objective magnification (7X at the binocular eyepieces):



The uneven framing and unevening vignetting is due to the old Motic phototube. They may be China's best microscope maker, but their stuff can hard to keep in alignment.

Here is a 100% crop:



Imagine what you can do with more modern gear.

[cowasaki]

Following your post, you've re-peaked my interest in resolving this, and I just had a very brief five minutes to revisit it. If I deconstruct the microscope's camera head to some degree, then without any adapters, optical or otherwise, I can get a fairly reasonable 1x crop picture but I'll need to experiment more and fabricate some sort of adjustable enclosed adapter tube.

As encouragement, here is what I get with direct projection onto the sensor of a 12 megapixel Olympus E-PL1 camera with 0.7X objective magnification (7X at the binocular eyepieces):



The uneven framing and unevening vignetting is due to the old Motic phototube. They may be China's best microscope maker, but their stuff can hard to keep in alignment.

Here is a 100% crop:



Imagine what you can do with more modern gear.

My OCD wouldn’t be happy with that resister :-)

[cowasaki]

Well as at least a temporary lighting solution possibly medium term I decided to raid my parts bin....

I bought a couple of Ikea bendy LED spot lights a few years ago which had a large crocodile clip at the end in order to attach them to a desk or shelf. Both of them broke at the clip so rather than waste them they got chucked in the parts bin.

10 minutes with Fusion 360 and a 3D print later here we have my temp solution.

I'm really surprised just how good the lighting is!

EDIT: Just found the light on Ikea's site although I couldn't find it in black - JANSJÖ @ £10

[Howardlong]

I have some of those same Ikea clip on spots, they’re OK but the beam width is quite wide so bear that in mind, they’re nowhere near as good as those Lloydtron 3W jobs I mentioned.

Yesterday I modified the stock camera port adapter to directly take an MFT camera wihiut extra optics, and it’s definitely a usable solution, although it still doesn’t seem as clear through the camera port as kt is through the eyepieces. I’m still working on that part.

[cowasaki]

I have some of those same Ikea clip on spots, they’re OK but the beam width is quite wide so bear that in mind, they’re nowhere near as good as those Lloydtron 3W jobs I mentioned.

Yesterday I modified the stock camera port adapter to directly take an MFT camera wihiut extra optics, and it’s definitely a usable solution, although it still doesn’t seem as clear through the camera port as kt is through the eyepieces. I’m still working on that part.

The Ikea light is just to keep me going till I get an alternative.  :-)

[agehall]

You probably want a ring light too but those lights make great extras. I have one attached on my scope so that I can get a bit of extra light and some more yellow light in the view.

[SteveyG]

Thanks for this post, I was looking at the same model - didn't know about the inability to capture video at the same time 

In terms of cameras, I've been testing a couple of low cost items out on my channel. I'll shortly be doing a video on a "21MP" camera (it offers 2K recording anyway), but so far the this one is performing well and the manufacturers are actually updating firmware.

I used this camera in my latest soldering video with a cheap 180x lens. (https://www.banggood.com/custlink/KmDmyuomIK)

https://youtu.be/bGkbXPUzcmE

[SteveyG]

Been there, done that. In the end I bought the "proper" three port simul focal job, the trinocular head sits in a cupboard now. I feel it should be made a lot clearer: trinocular is fine if you only want stills, but useless if you're making videos.

These are the gooseneck LED lamps I used: https://www.amazon.co.uk/gp/product/B00H1O1ZF2 They're unavailable now, but you can probably find them elsewhere. Importantly, the beamwidth is only about 30 degrees: most small LED worklights are much wider.

Edit: to be clear you really need at least two light sources, each from a different angle. Both of the gooeneck jobs I have are clipped to the microscope's head so they also move as the head moves.

For some reason I missed all your recent videos, not sure why YouTube didn't put them in my feed. Great close-ups - what microscope are you using please?

[coppice]

The problem with the LED ring lights is that they get in the way of some tools in particular hot air desoldering tools. I don’t use the LED ring as a result, instead I use a pair of gooseneck 3W spots clipped onto the microscope itself.

Even if you find a ring light doesn't get in your way, you should consider supplementing it with some directional lighting. Ring lights avoid dark corners, but used on their own they can result in a so little shadowing that there is no contrast in the image, and details are hard to make out.

[SteveyG]

I noticed this recently. Some lights 45° to the board give much more contrast.

[cowasaki]

I noticed this recently. Some lights 45° to the board give much more contrast.

I've seen your channel

[Howardlong]

Great close-ups - what microscope are you using please?

It's a bit of a Frankenstein setup TBH.


Originally I had SM-8TW2-144S. This is a trinocular, not a simul-focal head. The articulating arm is great. I don't use the LED ring because I found that it gets in the way of some tools like hot air irons. I do use the 0.5x Barlow permanently. Instead of the ring I use a pair of gooseneck Lloytron 3W L1507SV narrow (~45 degree) spots clipped onto the microscope's head that are still available on ebay. The problem with desk based goosenecks is that you'd have to repeatedly adjust them as you adjust the microscope head's position. Having them clipped to the head itself saves you this inconvenience.

A couple of years ago, I swapped the trinocular microscope head for a simul-focal one (SM745TP): simul-focal is pretty much essential if you do video. The trinocular and simul-focal heads look almost identical from the outside, and all the ports are the same.

Originally for the camera I used a 0.5x reducer with C mount and a third party HDMI 1080p camera. Although the latency isn't bad, it's still useless trying to solder using a screen compared to having a 3D view that a fully optical stereoscopic microscope affords.

I also modified a 4k60p Yi 4k Plus action camera to take a C mount, but the results weren't great. As a result I have been using a Micro four-thirds camera (MFT) on the mount for the past year or so, still with the 0.5x reducer, but with a C mount to MFT adapter, and using the in camera cropping at native sensor resolution. The camera I've been using is either a Panasonic G7 or GX9 in 1080p. Over the past few days, I've modified the camera port to directly take an uncropped MFT sensor without the 0.5x reducer. Although it's a bit better, I'm not totally happy with the results. At present there still seems to be little point in going to 4k as initial indications suggest that the results don't seem to improve compared to 1080p, but I probably need to play with it a bit more. My ultimate ambition is to be able to take decent 4k60p video on the camera head: I do already have a couple of MFT cameras with that capability.

This brings me to whether or not these microscopes are parfocal or not, i.e., do they maintain a reasonable focus throughout the zoom range? The answer is yes, but you do need to spend time setting them up properly in a systematic and iterative way. The reason I mention this is that I believe this is at least part of the reason that I'm having difficulty with the camera port giving the same kind of image quality that I see through the eye pieces particularly over the complete zoom range. I can get close, but it's still a little disappointing at the moment, I need to spend more time at it. For example, as well as the aforementioned parfocal aspect, focus still isn't maintained across the frame.

Finally, with the 0.5x Barlow there is about 8" of working distance which is good, but you need to make sure you can adjust your chair height for this comfortably. I have to set my chair at its highest setting for this to be comfortable. Not a big deal, assuming you have a height adjustable chair of course!

[jfiresto]

.... I've modified the camera port to directly take an uncropped MFT sensor without the 0.5x reducer. Although it's a bit better, I'm not totally happy with the results....

Perhaps just a silly question: if you move an eyepiece from a binocular tube to over the photo port, is the view just as good?

For example, as well as the aforementioned parfocal aspect, focus still isn't maintained across the frame.

Is the focus good over the central 60% of the field of view, by area?

[Howardlong]

.... I've modified the camera port to directly take an uncropped MFT sensor without the 0.5x reducer. Although it's a bit better, I'm not totally happy with the results....

Perhaps just a silly question: if you move an eyepiece from a binocular tube to over the photo port, is the view just as good?

It's not at all bad, although I suspect my eye is doing some correcting.

For example, as well as the aforementioned parfocal aspect, focus still isn't maintained across the frame.

Is the focus good over the central 60% of the field of view, by area?

Some 0603 parts and a 0.5mm pitch QFM on a 0.635mm pitch prototyping board.

Original large area image (20Mpx, 11MB)
Original zoomed area image (80Mpx, 31MB)

[jfiresto]

Thanks for posting the pictures. I downloaded the originals and believe that the camera is capturing enough pixels. More than enough. I can't view the second, zoomed image without slowing my computer to a crawl while it pages out memory.

Do you have a Greenough design microscope? If you do, I would expect the usual intra-microscope aberrations will be radially symmetric. The gradual, left-to-right focus problem, here, is not. Is one side focusing too high and the other too low?

Have you tried tilting the specimen 5-6 degrees to make it perpendicular to the eyepiece/optical path the camera is viewing?

Does the focus gradient(?) disappear and the focus greatly improve when you remove the 0.5X auxiliary objective (assuming you are using one)?

Could anything be tilting the camera's image sensor with respect to the camera port?

Has the camera port image somehow gotten tilted or moved off center?