Another AO10 LED conversion

Here you can discuss DIY adaptations to the microscope.
Post Reply
Message
Author
dtsh
Posts: 977
Joined: Wed May 01, 2019 6:06 pm
Location: Wisconsin

Another AO10 LED conversion

#1 Post by dtsh » Sat Aug 28, 2021 9:36 pm

There are many examples of people doing very nice jobs of converting the AO 10 to LED, this is probably not one of them, but this one has been working fine for me thus far. I'm no electronics guy, the circuit was designed by a friend and I just followed what he showed me. There's not much to it, but the main point for this was to provide the info and not hijack other threads. If you have any questions, I'll try my best to answer them.


Here is a parts list Not included are two resistors, 1K and 15k, and a 10k potentiometer.

LED substrate:
https://www.mouser.com/ProductDetail/be ... cycode=USD

Cree LED
https://www.mouser.com/ProductDetail/cr ... cycode=USD

LED Driver (0-10v dimmer)
https://www.digikey.com/en/products/det ... 00/2720720

It definitely needs a good heatsink as I cooked one of the LEDs with about 2 hours of use due to not enough thermal contact with the aluminum I had it mounted on. In this revised version, I used a small heatsink to mount the LED to and an aluminum plate I cut to match the mounting bracket of the original light assembly. I chose one of my more beatup stands in case I messed up, but the cosmetics don't affect performance.

Here's the schematic, I'm pretty sure it's accurate.
led_schematic.jpg
led_schematic.jpg (73.75 KiB) Viewed 7658 times
The power jack is a simple 5.5x2.5mm jack for which I had drilled out an appropriate sized hole in the back where the original power lead came out.
AO10_led_power_jack.jpg
AO10_led_power_jack.jpg (59.54 KiB) Viewed 7658 times
I decided to mount the pot on one of the feet as it fit and was in a decent location. The stage is a lefty now, but was a righty one when I did the work. It should probably be setup opposite, but it still works OK.
AO10_led_mounted_pot.jpg
AO10_led_mounted_pot.jpg (81.52 KiB) Viewed 7658 times
Here's a shot showing where I drilled a couple of holes and cut out some of the webbing to make room for the wires.
AO10_led_wiring.jpg
AO10_led_wiring.jpg (88.36 KiB) Viewed 7658 times
3D printed knob and it's good.
AO10_led_knob.jpg
AO10_led_knob.jpg (82.6 KiB) Viewed 7658 times

Dubious
Posts: 426
Joined: Sun May 09, 2021 7:55 pm

Re: Another AO10 LED conversion

#2 Post by Dubious » Mon Aug 30, 2021 6:23 am

Looks quite nice to me, and thanks for posting! I see you chose a neutral-white led. How are you finding the color, compared to the halogen lamp? I am thinking of doing something similar but thought I might try a warm-white led.

dtsh
Posts: 977
Joined: Wed May 01, 2019 6:06 pm
Location: Wisconsin

Re: Another AO10 LED conversion

#3 Post by dtsh » Mon Aug 30, 2021 3:43 pm

Dubious wrote:
Mon Aug 30, 2021 6:23 am
Looks quite nice to me, and thanks for posting! I see you chose a neutral-white led. How are you finding the color, compared to the halogen lamp? I am thinking of doing something similar but thought I might try a warm-white led.
I went back and forth on the color temperature trying to decide, I seem to recall reading somewhere that closer to "daylight" was preferred, so I chose something closer to that end. I'm pretty satisfied with the resulting image, but I'm not sure my opinion matters much from a technical perspective as I lack the skill to evaluate it (and thus chose to go with the recommended). The illuminator this replaced was the tungsten, I have a halogen AO10 as well, but until it fails it will remain as-is. I haven't put the darkfield condenser on it, but the output is so much brighter with the LED that I'm certain it will be a nice improvement there too.

I had considered putting the potentiometer on the back with the power plug and if i were to do it again I would probably give that option a little more consideration. In my mock ups i was using a larger knob and didn't like the placement, but with a smaller one as I eventually settled on I think it would work fine.

Gatorengineer64
Posts: 123
Joined: Sun Aug 22, 2021 12:10 am

Re: Another AO10 LED conversion

#4 Post by Gatorengineer64 » Thu Sep 09, 2021 2:40 am

Do you have any issues with seeing the LED? What I mean by that is on a flash light you can actually see the LED in the light, if you focus it to a narrow beam....

dtsh
Posts: 977
Joined: Wed May 01, 2019 6:06 pm
Location: Wisconsin

Re: Another AO10 LED conversion

#5 Post by dtsh » Fri Sep 10, 2021 1:36 pm

Gatorengineer64 wrote:
Thu Sep 09, 2021 2:40 am
Do you have any issues with seeing the LED? What I mean by that is on a flash light you can actually see the LED in the light, if you focus it to a narrow beam....
The AO10 uses modified Koehler illumination, so there's a frosted surface on the lens just after the LED to help make even diffused light so the LED itself can't be seen. Koehler illumination (modified or not) is designed to ensure the "filament" is out of focus in the image plane.

The LED works quite well and is much brighter than the original lighting; plus the removal of the transformer clears up some desk space.

User avatar
Roldorf
Posts: 574
Joined: Tue Apr 23, 2019 6:58 pm
Location: Northern Germany

Re: Another AO10 LED conversion

#6 Post by Roldorf » Fri Sep 10, 2021 3:11 pm

And of course should last over 10,000 hrs.
Location: Northern Germany

Unknown Brand: Optika SFX 91: Bresser Science Infinity: Canon 4000d
ImageImage

rmb
Posts: 35
Joined: Fri Dec 01, 2017 5:38 pm
Location: scotland near Edinburgh

Re: Another AO10 LED conversion

#7 Post by rmb » Mon Sep 27, 2021 6:31 pm

I have done a number of led conversions on AO 20s, a zetopan, a vickers patholette, a b&l dynoptic, an old leitz ortholux and recently an old leitz Labolux. I am baffled by the complicated efforts people are going to. All you need is your 3w star led, a penny washer (copper or aluminium), some thermal glue, a 100ohm potentiometer and 3 AA RECHARGEABLE batteries. Acquire battery box, wires and connectors to suit. You can get 3x AA battery boxes with a switch. Solder wires to led. Glue led to washer. Put pot in series with both. 3 rechargeables provide 3.6v which is ideal for the led. The 100ohm gives a nice control to the brightness. No flickering. The hardest part is locating the led in the right place I.e. as near to where the lamp filament was. Some cheap metal brackets and more thermal glue helps with this if you put a piece of slightly opaque paper over the light well. You can then move the led about till its optimal. Hold the bits in place with Croc clips till the glue sets. Drill and screw if you must. The glue can be prised off if you aren't happy with the setup. Plus you have a portable setup and one battery box can be used on different scopes if you don't build it in. Can look crude if you are messy but it is all inside the scope. The batteries last for days

apochronaut
Posts: 6272
Joined: Fri May 15, 2015 12:15 am

Re: Another AO10 LED conversion

#8 Post by apochronaut » Tue Sep 28, 2021 12:20 am

Wow! I'd really like to see images! That would be great. High resolution darkfield, phase, anything. I'm always in for enhanced illumination or contrast systems.

dtsh
Posts: 977
Joined: Wed May 01, 2019 6:06 pm
Location: Wisconsin

Re: Another AO10 LED conversion

#9 Post by dtsh » Tue Sep 28, 2021 12:43 am

apochronaut wrote:
Tue Sep 28, 2021 12:20 am
Wow! I'd really like to see images! That would be great. High resolution darkfield, phase, anything. I'm always in for enhanced illumination or contrast systems.
If this was directed at me, I have two images taken with the LED with the brightness at about 1/3 of it's potential output over on my "Newbie learns to take images" thread. They're awful, but that's the fun of learning; seeing progress (somtimes).

Brightfield: viewtopic.php?p=109144#p109144
Darkfield: viewtopic.php?p=109270#p109270
(I'm still figuring out darkfield, it's worse than my usual, which is already not good)

Hobbyst46
Posts: 4277
Joined: Mon Aug 21, 2017 9:02 pm

Re: Another AO10 LED conversion

#10 Post by Hobbyst46 » Tue Sep 28, 2021 9:44 am

dtsh wrote:
Tue Sep 28, 2021 12:43 am
Darkfield: viewtopic.php?p=109270#p109270
(I'm still figuring out darkfield, it's worse than my usual, which is already not good)
Was the DF condenser oiled to the slide bottom ?

apochronaut
Posts: 6272
Joined: Fri May 15, 2015 12:15 am

Re: Another AO10 LED conversion

#11 Post by apochronaut » Tue Sep 28, 2021 12:35 pm

dtsh wrote:
Tue Sep 28, 2021 12:43 am
apochronaut wrote:
Tue Sep 28, 2021 12:20 am
Wow! I'd really like to see images! That would be great. High resolution darkfield, phase, anything. I'm always in for enhanced illumination or contrast systems.
If this was directed at me, I have two images taken with the LED with the brightness at about 1/3 of it's potential output over on my "Newbie learns to take images" thread. They're awful, but that's the fun of learning; seeing progress (somtimes).

Brightfield: viewtopic.php?p=109144#p109144
Darkfield: viewtopic.php?p=109270#p109270
(I'm still figuring out darkfield, it's worse than my usual, which is already not good)
No. It was related to the rmb post. I'd like to see images from those systems.

dtsh
Posts: 977
Joined: Wed May 01, 2019 6:06 pm
Location: Wisconsin

Re: Another AO10 LED conversion

#12 Post by dtsh » Tue Sep 28, 2021 1:30 pm

Hobbyst46 wrote:
Tue Sep 28, 2021 9:44 am
dtsh wrote:
Tue Sep 28, 2021 12:43 am
Darkfield: viewtopic.php?p=109270#p109270
(I'm still figuring out darkfield, it's worse than my usual, which is already not good)
Was the DF condenser oiled to the slide bottom ?
I beleive the slide was oiled to the condenser, but I wouldn't bet money on it as I was twiddling a lot of knobs that day; eventually I'll get back to the diatom slide and try again. I should have put scale bars on the images as independently there's little to suggest that the first diatom (G.attenutatum) was significantly larger than the second (G.reimeri). In brightfield, the G. reimeri has been very difficult to see at all; could be my skill, the mountant, specimen, or something else.

My condenser came with a yellowed, but still readable manual which I did read and learned from it that I needed to oil both the condenser and the objective.

dtsh
Posts: 977
Joined: Wed May 01, 2019 6:06 pm
Location: Wisconsin

Re: Another AO10 LED conversion

#13 Post by dtsh » Wed Oct 20, 2021 7:55 pm

I noticed that none of the images showed how I mounted the LED. The LED itself is mounted to a "star" substrate, which itself is glued with thermal glue to a generic aluminum heatsink of appropriate size (I believe it's a 22mmx22mmx10mm).
The aluminum plate is a piece of scrap I had laying around that I cut to fit and drilled mounting holes in the appropriate locations then tapped a single hole to accept a bolt which passes between the middle fins of the heatsink and holds it against the aluminum plate snuggly. I measured pretty carefully to get the LED in a close approximation of where the original filament would have been located. The AO10 has a frosted lens, so positioning doesn't appear to be super critical.
LED_mount.jpg
LED_mount.jpg (78.88 KiB) Viewed 6736 times

dtsh
Posts: 977
Joined: Wed May 01, 2019 6:06 pm
Location: Wisconsin

Re: Another AO10 LED conversion

#14 Post by dtsh » Wed Oct 20, 2021 8:58 pm

dtsh wrote:
Mon Aug 30, 2021 3:43 pm
I haven't put the darkfield condenser on it, but the output is so much brighter with the LED that I'm certain it will be a nice improvement there too.
And I have since used it with the darkfield condenser and it is indeed an improvement. More might be better, but it works better than it did with the original illumination.

dtsh
Posts: 977
Joined: Wed May 01, 2019 6:06 pm
Location: Wisconsin

Re: Another AO10 LED conversion

#15 Post by dtsh » Wed Jan 05, 2022 10:48 pm

I'm testing a cleaner version of this setup. Instead of mounting the pot on the foot, I am going to attempt mounting it in the back. There will be a 5.5mmx2.1mm 12V input where the power cord normally runs, then above it I intend to mount the pot. Due to the thickness of the microscope casing, I will need to mill a small portion of the case down so that I can have enough threads to mount the pot. It should be less invasive than the previous, requiring a lot less modification to the original AO10. The potentiometer I chose this time has a built-in on/off switch too.

My apologies to the purists, I'm butchering another instrument.
ao10_led3.jpg
ao10_led3.jpg (111.39 KiB) Viewed 6165 times
I'm not entirely sold on how/where I have the LED placed. It's about the same distance from the frosted lens as the original tungsten filament, but I've considered milling a slot into the aluminum base and making a matching T-nut to ride in the slot to allow for adjusting the distance. I suspect it's overkill and might not do it as I haven't had any issues with the first one.

Right now it's going through a burn-in; after 60 minutes of runtime the temperature of the heatsink is 21c (19c in the room).
ao10_led4.jpg
ao10_led4.jpg (95.58 KiB) Viewed 6165 times
With the 10W Cree LED, it's plenty bright even at it's lowest setting (note the 10% transmission neutral density filter is engaged, greatly reducing it's apparent brightness).

DrPhoxinus
Posts: 316
Joined: Mon Jan 28, 2019 5:17 pm
Location: Rochester Hills, MI

Re: Another AO10 LED conversion

#16 Post by DrPhoxinus » Thu Jan 06, 2022 1:16 am

Very Interesting.

I have had good luck with a PWN but I haven’t tried photography. $7 at Amazon

https://www.amazon.com/PWM-Dimmer-Contr ... C86&sr=8-7

Hobbyst46
Posts: 4277
Joined: Mon Aug 21, 2017 9:02 pm

Re: Another AO10 LED conversion

#17 Post by Hobbyst46 » Thu Jan 06, 2022 8:36 am

DrPhoxinus wrote:
Thu Jan 06, 2022 1:16 am
Very Interesting.

I have had good luck with a PWN but I haven’t tried photography. $7 at Amazon

https://www.amazon.com/PWM-Dimmer-Contr ... C86&sr=8-7
Only that such dimmer might yield banding in photography at short exposure times. A higher frequency PWM will do better in this respect.

DrPhoxinus
Posts: 316
Joined: Mon Jan 28, 2019 5:17 pm
Location: Rochester Hills, MI

Re: Another AO10 LED conversion

#18 Post by DrPhoxinus » Thu Jan 06, 2022 4:40 pm

It does do a great job with halogen bulbs

dtsh
Posts: 977
Joined: Wed May 01, 2019 6:06 pm
Location: Wisconsin

Re: Another AO10 LED conversion

#19 Post by dtsh » Thu Jan 06, 2022 8:09 pm

I chose this driver specifically as it had an internal analog dimming circuit (0-10v), making it almost entirely self-contained and easy to setup. I was avoiding PWM as the off/on nature can cause banding, but I readily admit that my understanding of the electronics involved is marginal.

The entire parts list is around $25; sure I could have gone with countless other solutions, but this one is a nice and tidy little package that's easy to assemble; a replacement transformer and rewiring the old socket would have cost more and I'd still have to fabricate the plug, unless anyone is aware of a replacement 2 pin plug of the type used for these.

I have a couple of other drivers with a 0-5v dimming circuit which, try as I might, I can't seem to get to work; my understanding almost certainly the limiting factor. They have PVM dimming as well, but I need to learn how to implement that first. So many things to learn and only one life to try and pack it all in.

DrPhoxinus
Posts: 316
Joined: Mon Jan 28, 2019 5:17 pm
Location: Rochester Hills, MI

Re: Another AO10 LED conversion

#20 Post by DrPhoxinus » Thu Jan 06, 2022 8:23 pm

Indeed,

I have powered the bulbs by using the female cable that was
there to power a second head. That cable is wired to the bulb socket. What are those connectors?
Amphenol?

If anyone knows please tell me.

Gerard

dtsh
Posts: 977
Joined: Wed May 01, 2019 6:06 pm
Location: Wisconsin

Re: Another AO10 LED conversion

#21 Post by dtsh » Wed Oct 05, 2022 1:39 am

Another one built and ready for testing...
LED1.jpg
LED1.jpg (94.36 KiB) Viewed 3671 times
The LED and driver mounted on an aluminum plate
LED2.jpg
LED2.jpg (84.09 KiB) Viewed 3671 times
I tried very hard to make this as a drop-in that required no modification, but I was not happy with double-sided tape, so I tapped two small holes.
LED3.jpg
LED3.jpg (66.69 KiB) Viewed 3671 times
A 3D printed box for the DC input and 10k potentiometer
LED4.jpg
LED4.jpg (84.54 KiB) Viewed 3671 times
And 3D printed.knob.

All working great, this one is slightly less output than the previous ones, guessing somewhere in the 6-8w range.

Post Reply