Microscope camera

Discussion about cameras, microscopes, stains, and gadgets, along with useful tips for preparation of fungi samples
User avatar
adampembs
Frequent user
Posts: 1315
Joined: Wed May 13, 2015 11:40 am
Location: Pembrokeshire
Contact:

Re: Microscope camera

Post by adampembs » Fri Aug 14, 2015 9:06 am

I'm getting horizontal lines which gradually scroll up. It's less noticeable at x400 and x1000 but at x100, its quite bad. I have LED lighting, using a car bulb with 12V adapter. Do I need a better dimmer switch or is there another solution?
Attachments
dimmer.jpg
LED dimmer switch
dimmer.jpg (6.34 KiB) Viewed 2260 times
IS_28-002.jpg
x100
IS_29.jpg
x400
Adam Pollard
Site admin

User avatar
Lancashire Lad
Frequent user
Posts: 744
Joined: Wed May 13, 2015 11:59 am
Location: Red Rose County
Contact:

Re: Microscope camera

Post by Lancashire Lad » Fri Aug 14, 2015 12:43 pm

Hi Adam,

My electrical background is in the industrial/commercial contracting side of things and not electronics, so I can't give you more than the basics of what might be going on, but I think you might have a tricky problem there!

LED's do flicker, generally at twice the input supply frequency (termed the ripple frequency of the output LED driver). - So in the UK, (assuming no filtering is incorporated into the LED circuitry), and with mains frequency of 50Hz, they will have a ripple frequency flicker of 100Hz.
Just in case there are any electronics engineers looking in - I am aware that most LED's (other than those recently developed for commercial lighting situations), only work on direct current, so there has to be some sort of in-built ac/dc conversion/rectification somewhere in the circuitry - but depending on what that consists of, it might only be removing the negative portion of the alternating current waveform.

Such flickering is imperceptible to the human eye in almost all everyday circumstances, so general usage LED dimmer driver circuitry wouldn't need to take it into account. (It would make the products more expensive!).

It's generally more of a problem for video work, when footage is shot in rooms with LED (or fluorescent) lighting. - In those circumstances, it can be removed electronically in post camera video editing suites.
Unfortunately, Digicam type microscope cameras are, in effect, just very simple video cameras, and the still shots from them are simply "plucked" from the streaming video being sent to the computer. (So, if the video contains flicker - the still shot will show the flicker).

Filtering flicker out could be done by introducing a capacitor - but the value of that would need to be closely matched to the components in the rest of the circuitry. The wrong capacitor might botch thing up altogether! (You are in the realms of requiring a specialist electronics engineer).

There are various "Heath Robinson" techniques I've heard of to confirm whether it is in fact LED's that are producing flicker, rather than interference from some other nearby electrical equipment. (Have you discounted that possibility?).
One such being to put the LED at the end of a long(ish) wire, and spin it around over your head as fast as you can in a darkened room. If it is flickering, the circle of light produced will have a strobe effect.
Another method is to use a "spinning top". The disc of which needs to be marked with a large number of lines from centre to perimeter (radius). When spun under the LED lighting (with no other light whatsoever) there will come a point as it slows down, (provided that the LED light is flickering), when the lines appear static, and then start to slowly rotate in the opposite direction to that in which the top is spinning.

Is your dimmer capable of running a halogen lamp of the same voltage? (A halogen lamp would be in the 20-30watts region, as opposed to a couple of watts max. for the LED). If it can, it might be worth a try to see if it improves things.

A better quality dimmer with flicker reducing output circuitry might be available, but I have no idea how you would specify/ensure that what you might buy will definitely work for this particular application.
(If you are handy with a soldering iron, you might be able to get a complete "replacement" LED lighting system for a scope from somewhere like Brunel, and then cobble it together to fit your scope? - they do occasionally have spare parts available).

Might also be worth considering (if you have access to someone else with a scope) trying your camera on another scope that has a proprietary lighting system to see if the problem persists.
Should prove one way or the other whether it is your lighting (or indeed the camera) that is at fault.

Regards,
Mike.
Common sense is not so common.

User avatar
Lancashire Lad
Frequent user
Posts: 744
Joined: Wed May 13, 2015 11:59 am
Location: Red Rose County
Contact:

Re: Microscope camera

Post by Lancashire Lad » Fri Aug 14, 2015 1:02 pm

Just thought of something that might make my entire reply above complete nonsense - but I can't quite read the details on the dimmer unit photo.

Is your dimmer input supplied by direct current? (i.e from a car battery) or are you feeding it from a mains-low voltage ac/dc transformed power supply?

If it's fed from an ac/dc 240/12v power supply, chances are the output will be via something like a cheap Wheatstone bridge rectifier, which will still give flicker.

But if your supply is pure dc from a12volt battery then the LED's shouldn't be able to flicker at all, which would point to the camera being at fault.

If it is fed from an ac/dc transformer/power supply, you should be able to check the system by temporarily feeding it from a battery of the same output voltage. (Or maybe permanently feed it via battery? - a decent 12v battery should last for weeks/months driving an LED before it need recharging).

Regards,
Mike.
Common sense is not so common.

User avatar
Chris Yeates
Frequent user
Posts: 708
Joined: Tue May 26, 2015 7:01 pm
Location: Huddersfield, West Yorkshire

Re: Microscope camera

Post by Chris Yeates » Fri Aug 14, 2015 6:06 pm

I had very similar problems with a GXCAM 9MP camera on a Brunel SP200 'scope (the earlier version compared with the current one on sale, but fundamentally the same microscope). That was using quartz halogen 6v 20 watt illumination (Kohler). I think that basically the GXCAM and the Tucsen function very similarly. I resolved the problem by tweaking the settings of the camera software. That said I haven't had the problem with the Tucsen: the Nikon 'scope dates from the mid 1990s (that's when they came into production), again using halogen illumination.
Chris
"You must know it's right, the spore is on the wind tonight"
Steely Dan - "Rose Darling"

User avatar
adampembs
Frequent user
Posts: 1315
Joined: Wed May 13, 2015 11:40 am
Location: Pembrokeshire
Contact:

Re: Microscope camera

Post by adampembs » Fri Aug 14, 2015 7:27 pm

Yes, I'm using a 12V main transformer (old router/phone charger) - your idea about the 12V battery is good. I'll buy one.
Adam Pollard
Site admin

User avatar
adampembs
Frequent user
Posts: 1315
Joined: Wed May 13, 2015 11:40 am
Location: Pembrokeshire
Contact:

Re: Microscope camera

Post by adampembs » Fri Aug 14, 2015 8:07 pm

Chris Yeates wrote:I had very similar problems with a GXCAM 9MP camera on a Brunel SP200 'scope (the earlier version compared with the current one on sale, but fundamentally the same microscope). That was using quartz halogen 6v 20 watt illumination (Kohler). I think that basically the GXCAM and the Tucsen function very similarly. I resolved the problem by tweaking the settings of the camera software. That said I haven't had the problem with the Tucsen: the Nikon 'scope dates from the mid 1990s (that's when they came into production), again using halogen illumination.
Chris
Tried "gain", "gamma" and "exposure" but I think Mike is right, the LED is strobing due to low frequency waveforms from the AC/DC conversion: invisible to the naked eye, but like when you see video cameras filing TVs. I understand the rheostats used in modern commercial microscopes are more sophisticated than my budget dimmer.
Adam Pollard
Site admin

User avatar
adampembs
Frequent user
Posts: 1315
Joined: Wed May 13, 2015 11:40 am
Location: Pembrokeshire
Contact:

Re: Microscope camera

Post by adampembs » Fri Aug 21, 2015 4:31 pm

All sorted! Visited our local micrscope expert and there were a few issues.
The dimmer switch (PCM) was causing strobing, even with a 12V battery. Now using a pcm working at ~ 50 KHz so not seeing any strobing.

My LED also had a built in pcm, so that has been removed. I'm also running off a 12V battery.
Attachments
P1110096.JPG
LED lighting for microscope
P1110097.JPG
New pcm device
Adam Pollard
Site admin

Sheffield Lass
User
Posts: 5
Joined: Thu Jul 02, 2015 8:56 pm

Re: Microscope camera

Post by Sheffield Lass » Sun Oct 18, 2015 3:35 pm

I've just bitten the bullet and bought myself a new microscope camera as my old one was getting more and more problematic as it got increasingly difficult to connect at USB2 speed.

Due to budget constraints and the way I use my microscope and camera I opted for a 5Mp Amscope MU500-ck model, which should now be winging its way to me from USA courtesy of Ebay, for a mere £152 incl all import taxes, postage etc. This is about £60-80 cheaper than similar capability / same model I could have bought in the UK. If budget had not been an issue I'd have gone for a 5Mp USB3 camera, probably the Touptek U3CMOS camera (which is £375+)

As I did plenty of research before I opted for it, thought I'd share the results of my research of what is available.

Seems there are 3 main camera manufacturers in the budget market that we can get in the UK. Scopetek, Tucsen and Touptek. There are plenty of very expensive manufacturers also available, if you've got £1000s to spare!

Scopetek and Touptek do a very basic range (DCM for Scopetek, SCMOS for Touptek), which are usually shipped without measuring software. Absolute minimum resolution needed for fungi microscopy is 1.3Mp. Very cheap and cheerful at c£60+. For those on a very tight budget they just about scrape through as usable for fungi id - I know, my old one was a 1.3Mp Scopetek with measuring software - but make sure that they have measuring software or that you can use 3rd party measuring software with it, otherwise they're useless, But for £100, you can get into a better quality range if you keep your eye open for bargains (ebay, Amazon)

Economy Camera range for USB2:
Touptek and Tucsen do a very similar range of cameras, often using the same sensors (Aptima / Micron), but in their own bodies and they come with their own different software (drivers and image capture/manipulation).
Touptek range is the UCMOS range (from 1.3Mp-10Mp) http://www.touptek.com/product/product. ... &class2=98
Tuscen is the ISH range (3Mp-10Mp). http://www.tucsen.com/products/ISH1000_ ... SH300.html

Brunell Microscope cameras are Touptek, those supplied by Nicks Science Supplies are Tucsen. Amscope repackage Touptek and Tucsen under their own name, and you can get some through Amazon UK and Ebay UK. Their MU range is Touptek, their MA and MT ranges are Tucsen. The difference between the MA and MT is the software. The MA range has much more limited usability with 3rd party software, the MT range makes this much easier - but can cost £20-30 more, but the MA/MT hardware is the same.

Looking at customer reviews, the software packaged with Touptek seemed more user friendly than Tucsen's. There is also an issue with Mac and Linux compatibility with both - generally appears that capability of the software is much more limited than if used on Windows. Touptek seems to be TWAIN compatible, but Tucsen needs closer scrutiny - the MT version has it, I believe, but the MA version does not. So do your research on those. I'm Windows only, so I only skimmed those details.

Camera ranges for USB3:
Touptek do USB3 ranges from 3Mp up to 18Mp.
Tucsen do as well, but under different range name.

I'll do a second post on some of the things consider.

Sheffield Lass
User
Posts: 5
Joined: Thu Jul 02, 2015 8:56 pm

Re: Microscope camera

Post by Sheffield Lass » Sun Oct 18, 2015 4:16 pm

cont.

The cameras usually need a C-mount adapter, which includes a lens which brings the camera image view similar to what you see down the eye piece. This most usually is a 0.5x lens - but go on the recommendations of the camera manufacturer. Its other purpose is to make the focal point of the camera very similar to that of the eye. If you are using a trinocular, and the camera for photo capture more than live view, then this is important, as the least amount of refocussing the microscope from the eye view to getting the camera in focus to take a sharp photo the better. Some of the adapters are focusable, some are fixed. So for a trinocular a focusable might be a good idea, as this can be set at the optimum position which minimises the need to refocus the microscope. If like me you only have a monocular, and use the camera all the time for live view, and eyepiece never, then fixed is fine. ToupTek have a choice of fixed and focussable. I'm not sure which Brunell Microscopes supply, so you need to ask. The Amscope ToupTek packages seem to have the fixed. I don't know which the Tucsen ones are.

Again, check your microscope tube internal diameter size - you may need some additional adaptors which may or may not be included in the kit you buy.

Some also come with a calibration slide, others do not. A calibration slide is necessary to set up the camera software to do accurate measurements. They can be expensive if bought separately, though some from China through Ebay can be had for as little as £10.

cont.

Sheffield Lass
User
Posts: 5
Joined: Thu Jul 02, 2015 8:56 pm

Re: Microscope camera

Post by Sheffield Lass » Sun Oct 18, 2015 6:48 pm

cont:

Resolution v frame per second

This is especially important if you use live view a lot.
1.3Mp (1280 x 1024) @ 11 frames per second (fps) is the minimum you can get away with for sufficient detail using a x400 lens for typical fungi id, and acceptable lag between focussing and image on the screen.

The higher the resolution the lower the fps. Below 10-11fps makes live view problematic - due to time lag. Check the manufacturers / suppliers data for the camera for the resolutions that it can offer.
For the 5Mp USB2 camera I've chosen I get: 5@2592x1944, 18@1280x960, 60@640x480.
The 2592 x 1944 would be too slow for live view, but gives me the option to do a still photo at the higher resolution. The 18@1280x960 gives me good speed and sufficient detail for live view, and fits my screen well.

If I'd gone for a 5Mp USB3 camera with USB3 computer I'd have got 14.2@2560x1922, 38.3@1280x960, 101.2@640x480. So I'd have good speed at 2560x1922, though I'd only seeing part of the view on my screen as my screen is not that size! Yes, I'd have liked that camera!

If you use the camera mainly for stills then higher resolutions (8Mp, 10Mp) may have advantages, but there are also other limitations with larger resolutions.

At greater magnification the microscope lenses tend to be the limiters of quality of view, so greater resolution doesn't necessarily improve detail. At high magnifications the resolving power of such high resolution camera can higher than that of the optical system. The image produced would be larger, but not better. Instead of resolving additional detail, the camera just produces more data! x40 lens generally gives a better quality than x100. Sometimes using a lower microscope magnification with a higher camera resolution can give a more detailed picture than a higher magnification with a high resolution.

Light dynamics. As resolution of camera increases, light sensitivity can decrease. Higher resolution ones can struggle at lower light levels. The dynamic range of a camera specifies the ratio between the brightest and darkest part of a specimen that it is still capable of capturing. Dynamic range should not be confused with the contrast of the picture. Cameras with a low dynamic range have a problem of accurately recording both bright and dark parts. The brighter parts of the specimen, such as fine structures that blend in with the background lit, will disappear completely. A reduction of the exposure time would make these bright structures visible, but now the darker parts of the specimen will be reproduced as completely black. It does not matter how we adjust the exposure time or light intensity, there will always be some loss of information either at the bright or at the dark end of the spectrum.

Light dynamics depends on a few things such as pixel size - the smaller the pixel, the less light reaching it. A 3Mp Toupek USB2 camera has pixel size 3.2x3.2um and light sensitivity 1.0 V/lux-sec, The 5Mp has pixel size 2.2x2.2um, and light sensitivity 0.53 V/lux-sec, 10 Mp has pixel size 1.67x1.67um and light sensitivity 0.31v/lux-sec. By comparison their 5Mp Toupek USB3 camera has pixel size 2.2x2.2um and light sensitivity 1.76v/lux-sec - so their newer technology is better, more sensitive for the same pixel size! Another reason for going for the USB3 model over the USB2 model.

Hope this helps a bit in making decisions on which camera will suit use and budget best. My choice was based on live-view use and a small budget. It would have been the same if it had been for quality stills (not live view) and small budget - as the budget was the limiter. But a bigger budget would have steered me towards USB3 rather than higher resolution, as live-view is more important for me.

Melanie

Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests