Thread specifications of Swift trinocular port

[crb5]

Has anyone figured out the thread specifications on the Swift 350T trinocular port (possibly similar to the Swift 380 port)? The tube is designed to accept 23 mm eyepiece USB cameras which it does – but a bit loosely, so I drilled a hole in the side and superglued a nut to it, so I can use a rubber-tipped screw to stop the camera rotating under the weight of the USB cable. This tube is attached to another by a 25 mm thread, possibly 0.75 mm pitch – it is a loose fit to a filter holder with this specification. The middle tube ends in a long 1 inch (25.4 mm) male thread with a collar which allows the tube length to be adjusted to make the camera parfocal with the eyepieces. I assumed this was a C-mount thread but it did not fit the C-mount camera thread which came with a Watec camera. On the other hand, the male thread was a loose fit in the SM1 thread of Thorlab tubes (https://www.thorlabs.com/newgrouppage9. ... up_id=3307) which have a 1.035”-40 thread. Finally, the whole unit attaches to the trinocular head via a 33 mm diameter thread of unknown specification. I have been trying to figure out ways of attaching various eyepieces and cameras to the trinoc port, using these threads without much success. However, I have found a work around that allows me to attach a dSLR directly or via a wide-angled eyepiece in afocal mode, if anyone is interested.

[LouiseScot]

Hi

I connected mine via a 3D printed adapter to a helical focuser. Then a 25mm astro eyepiece (=10x), and adapters to a Canon 18-55 EFS kit lens attached to my Canon 1100d:

AfocalSetup_15pc.jpg (35.81 KiB) Viewed 2322 times

I'm not certain it's optimal as I get a smaller image field of view compared to what I see via the oculars using the same eyepiece. It is parfocal with the oculars. I probably need to tinker with it some more but the image quality is quite good. I posted some darkfield images recently and other brightfield ones.

I'm afraid I can't tell you exactly what the trinocular port thread was - I just measured it with a digital caliper and did some trial and error. I'll have the fusion360 and stl files somewhere.

Louise

[LouiseScot]

Looks like I used ~M25.4 x 0.75 - within the precision limits of my 3D printer!

[dtsh]

That's a weird mix of threads, I would double check the pitch of the 25.4mm thread, that might be Imperial as 25.4mm = 1 inch.

[LouiseScot]

That's a weird mix of threads, I would double check the pitch of the 25.4mm thread, that might be Imperial as 25.4mm = 1 inch.

It might be, but you can adjust the tolerance to get a good enough fit. I counted 15 threads in 20mm -> 0.75mm pitch. 3D printing is often not so precise anyway - at least, not on my printer!

ps the female part of the M25 thread noted above fits a Nikon cfi60 objective which is M25 x 0.75

Louise

[Scarodactyl]

Inner diameters always come out a little tighter than specced on an fdm printer so that tracks.

[crb5]

To get around the problem of non-standard threads on the Swift trinocular tube, I sandwiched a 37mmF to 52mmM adapter ring (2 mm thick) between the bottom collar bearing the unknown 33 mm Swift thread and the microscope itself. With a 52 mm coupling ring and a set of step-up and step-down filter rings and a reversing ring, I ended up with a 37mmM filter thread concentric with the original Swift 25 mm tube. The 37mm thread fitted a EagleEye eyepiece I acquired around 20 years ago for digiscoping with a terrestrial telescope. Used alone this eyepiece provided a much bigger field of view (22 mm field number cf. supplied 10x eyepiece with 18 mm field number). With the digiscoping adapter, I could attach my Nikon D3000 with a 55 mm lens or a JVC camcorder, to image in the afocal mode. With the EagleEye ocular removed, I could access the inner 23mm ocular tube for its intended use with a USB camera, so switching between arrangements was easy

A modification of this arrangement using just the 52mm to 58mm step-up and step down rings, plus a 52mm to 42mmT2 adapter and T2 to Nikon bayonet, allowed me to position the Nikon D3000 APS-C sensor close to the primary image plane. While this achieved a full field of view, the image was poor at the edges. This is likely due to using a cheap non-plan objective lens, but as the afocal images looked better, it also reflects the need for an eyepiece to correct for aberrations in the objective. A lens-less USB camera positioned at the primary image plane works fine because it only records the central part of the image (around 1/5 the eyepiece view). I realize this thread makes heavy reading, but hope the pictures below clarify the arrangements.

[LouiseScot]

Hi

I previously found by experiment that an afocal setup seemed to work better than a direct projection one. The afocal principle is that with the camera lens just above the EP then the camera sees what the eye would see. There is no correction involved with the Swift. There is a simple formula for afocal setups which relates projection EP magnification and camera lens focal length and which gives the 'relay' magnification:

Relay Mag = (EP mag x Cam Lens fl)/250

ideally, you want the relay mag to be about 1.6 for an apsc dslr e.g.

1.6 = (EP mag x 50mm)/250

i.e. EP mag = 400/50 = 8x if using a 50mm lens.

The optimum obviously has to be adjusted for smaller sensors.

A 10x EP will still work but will give you a smaller fov, but a smaller fov may be preferable to having vignetting.

You may need some fine adjustments to get the camera sensor parfocal with the oculars

Needless to say, you have to make sure there's no sensor tilt..

Louise