MicrobeHunter.com Microscopy Forum

I am trying to adapt this led 20W 32V to replace the 20V 20W bulb without altering the original housing .


Regards

Your first concern would be the dissipation of heat. You need an efficient heat sink. The plate on which the LED is laid in the bottom photo is NOT a good heat sink.
Sorry, I am not familiar with AO microscopes, but a lot of info about LED conversion has been posted on this Forum, at least over the 3 past years. Try to search "conversion". If you insist on investing time and labor in this scope, the backup source for LED might be retroDiode of Texas. They sell LED units for at least some AO scopes, the price could be about $200 (need to ask).

Thank you, I have seen them. You sent me the info before.

I like electronics and I had this led at home and I was wandering if I could adapt it into the housing.

True it needs a heat sink, I have a few ones.

Regards

Another point, less important, to consider is how to dim the illumination intensity. A constant current source is best as you probably know. I am using a PCM, which limits my selection of camera shutter speed because of banding.

I've had good luck dimming low power LEDs with a potentiometer. It works quite well, because the voltage-current curve of the LED is very steep.

Your approach - once you have a proper power supply and heat sink looks good. Hard to tell from the photo if your LED is a multi-die unit? Ideally you'd want something (the light emitting diode part) about the size of the original filament -- and something like 2-3 watts (actual) should be plenty of light.

I've adapted one of those Series 10's by removing the back part (socket) and finding a LED flashlight that fits. My recollection is the front end had to be removed to get the LED die in the same position as the original filament. Don't recall if I had to turn down the barrel or bore out the holder to get a snug fit -- think I found something around 100 lumens that fit just right. Don't have that scope any more to check.

Pro was that the scope became portable. Con was that the rechargeable batteries needed recharging every hour or so. You could also wire it to a power supply.

I have a 3W LED kit that I have used successfully on quite a few varieties of scopes. The 3W is sufficient to replace a 6V30W halogen. The power supply needs 110V and has variable intensity control. Some of these have been in professional use for five years.

lorez

The emitter surface of your LED is much bigger that that of the bulb filament. Does your microscope have a Köhler illumination? LEDs have a low efficiency and need really good cooling. It is usually best to use a massive Aluminium heat sink. This should always stay cold enough so you can touch it.

MicroBob wrote:The emitter surface of your LED is much bigger that that of the bulb filament. Does your microscope have a Köhler illumination? LEDs have a low efficiency and need really good cooling. It is usually best to use a massive Aluminium heat sink. This should always stay cold enough so you can touch it.

Doesn't have a Köhler illumination. Yes I will adapt a heat sink. Any way this led do not get too hot.

Regards

lorez wrote:I have a 3W LED kit that I have used successfully on quite a few varieties of scopes. The 3W is sufficient to replace a 6V30W halogen. The power supply needs 110V and has variable intensity control. Some of these have been in professional use for five years.

lorez

Can you show it to us?

I am using this lead because I had it already. Gives a very strong light. I haven't try it in the Micros as I am replacing broken screws that are hard to find here.

Regards

PeteM wrote:Your approach - once you have a proper power supply and heat sink looks good. Hard to tell from the photo if your LED is a multi-die unit? Ideally you'd want something (the light emitting diode part) about the size of the original filament -- and something like 2-3 watts (actual) should be plenty of light.

I've adapted one of those Series 10's by removing the back part (socket) and finding a LED flashlight that fits. My recollection is the front end had to be removed to get the LED die in the same position as the original filament. Don't recall if I had to turn down the barrel or bore out the holder to get a snug fit -- think I found something around 100 lumens that fit just right. Don't have that scope any more to check.

Pro was that the scope became portable. Con was that the rechargeable batteries needed recharging every hour or so. You could also wire it to a power supply.

Yes, is multi-die unit 20 wats 2000 lums, 32 vilts

I am using 2 old laptop power suppliers in serial to get 34 volts with a 2 ohms resistance.

Can you send pictures of your adaptation?

I am trying to fit it using an old bulb....

Regards

MicroBob wrote:The emitter surface of your LED is much bigger that that of the bulb filament. Does your microscope have a Köhler illumination? LEDs have a low efficiency and need really good cooling. It is usually best to use a massive Aluminium heat sink. This should always stay cold enough so you can touch it.

Good advice, when I 'converted' my stereo-zoom to LED (actually all I did was replace the Halogen bulb with an LED-disc without any alteration to power supply etc and it worked perfectly, including dimming) I stuck such a 'chunk' of Aluminium heat-sink to the back of the disc using the heat-transfer glue made for the job. The LED (which is a multi-LED array of course) has worked perfectly ever since, without a single LED drop-out occurring.

John B.

mrsonchus wrote:

MicroBob wrote:The emitter surface of your LED is much bigger that that of the bulb filament. Does your microscope have a Köhler illumination? LEDs have a low efficiency and need really good cooling. It is usually best to use a massive Aluminium heat sink. This should always stay cold enough so you can touch it.

Good advice, when I 'converted' my stereo-zoom to LED (actually all I did was replace the Halogen bulb with an LED-disc without any alteration to power supply etc and it worked perfectly, including dimming) I stuck such a 'chunk' of Aluminium heat-sink to the back of the disc using the heat-transfer glue made for the job. The LED (which is a multi-LED array of course) has worked perfectly ever since, without a single LED drop-out occurring.

John B.

I think this is what John B. used.... I also bought one .... Works well...

https://www.ebay.com/itm/New-G4-DC-12V- ... 239c5a39a0

BillT

. . .

I think this is what John B. used.... I also bought one .... Works well...

https://www.ebay.com/itm/New-G4-DC-12V- ... 239c5a39a0

BillT[/quote]

Bill, are you able to get Kohler (and a sharp image of the field diaphragm) with a multi-LED array?

mrsonchus , Can you show us how you did it?


I burnt my fingers...

Still experimenting how to dissipate heat.

Regards

stjepo wrote:I am trying to adapt this led 20W 32V to replace the 20V 20W bulb without altering the original housing .

As others have already mentioned; the chosen LED may be too powerful, and have too much emitting area to do the job well ... But the mechanical adaptation should be simple enough:
Find a scrap [burned out] bulb of the original type and safely remove the glass envelope. Then make a metal spacer that fits over the tubular section of the base and has a flat end to support the LED.
... If you don't have the tools to do this from scratch, then start looking for suitable low-cost items to butcher [something with fins would be ideal].

MichaelG.

stjepo wrote:mrsonchus , Can you show us how you did it?


I burnt my fingers...

Still experimenting how to dissipate heat.

Regards

Hi, I just bought a small tube of heat-conducting glue and a chunky Aluminium heat-sink, which I cut to size with a 'junior hacksaw'....
I'll take a picture of it and post it here a little later today. In the meantime I will echo MichaelG's suggestion that your LED is very-much overkill! You really don't need a 20W LED, in fact you don't really need 10W unless you're using high-mag darkfield a lot....

Back soon with a picture of my LED/heatsink.

John B.

mrsonchus wrote:
Hi, I just bought a small tube of heat-conducting glue and a chunky Aluminium heat-sink, which I cut to size with a 'junior hacksaw'....
I'll take a picture of it and post it here a little later today. In the meantime I will echo MichaelG's suggestion that your LED is very-much overkill! You really don't need a 20W LED, in fact you don't really need 10W unless you're using high-mag darkfield a lot....

Back soon with a picture of my LED/heatsink.

John B.

John... you are right, I don't even know how to use high-mag darkfield the reason I am using it is because I had this led at home that I was using for other electronic experiment.

Some how, I ended refurbishing a old AO SPENCER 1036 that was in a very bad shape.

Look at my heat-sink.. (an exaggeration)

Regards

Not sure if this is any help, but ... Decent photos of the GE 1460 bulb and its holder, here:
https://softsolder.com/2016/04/20/ameri ... -new-bulb/

MichaelG.

Hi again, crikey - that's what I CALL a heatsink!

Here's the one I used - cut to size - cost about £5 for the glue & heatsink... Works perfectly - including perfectly smooth dimming....

John B.

Very nice... for the binocular I got chinese led replacement as you can see in the picture (flickers at the begging):


Regards

Although I do not have an actual example of the lamp that John B is using, I think it worth noting a couple of features that are typical of this style:

• Although listed as 12Volts DC, these G4 lamps are reversible [either pin can be connected to positive or negative]
  
  They have a 'constant current' regulator on board [the current, and therefore the brightness, remains constant over a range of voltages]

https://www.ebay.com/itm/New-G4-DC-12V- ... 239c5a39a0

Presumably the dimmer being used by John B is a PWM [Pulse Width Modulation] device, not a simple voltage adjustment.

Someone please correct me, If I am wrong !!

MichaelG.

MichaelG. wrote:Although I do not have an actual example of the lamp that John B is using, I think it worth noting a couple of features that are typical of this style:

• Although listed as 12Volts DC, these G4 lamps are reversible [either pin can be connected to positive or negative]
  
  They have a 'constant current' regulator on board [the current, and therefore the brightness, remains constant over a range of voltages]

https://www.ebay.com/itm/New-G4-DC-12V- ... 239c5a39a0

Presumably the dimmer being used by John B is a PWM [Pulse Width Modulation] device, not a simple voltage adjustment.

Someone please correct me, If I am wrong !!

MichaelG.

Hi Michael, I'm afraid I don't actually know the type of dimmer - I just plugged the LED into my stereo-zoom and it worked - a lucky break I think!

John B.

Now working...

Hello

1. Is this a 20W LED?

2. I would suggest to test the efficiency of the heat sink. As it looks now, the wiring are in contact with it and are prone to be warmed and the isolation sleeves are under potential risk of damage.
Moreover, when the microscope is in use, standing upright, and since the heat sink has no fins, it will output heat to the bench below and, more importantly, to the microscope chassis itself - possible risk to optics, apertures etc.
3. If heating appears to be excessive, one possible remedy is to place the microscope on a flat aluminum plate, somewhat larger than the base, and thickness of 4-5 mm (my guess), to conduct the heat away. Must have good contact with the heat sink, and this is tricky because of the round cross section of the heat sink.
Maybe make a cradle-like spacer, of aluminum foil, taking care not to shortcut the LED wires. If the heat sink is coupled to the aluminum plate, heat conduction away from the microscope will be improved.

Maybe my suggestion is an overkill, perhaps the microscope was designed to withstand a hot light source in the base without damage.

P.S. did you use heat conducting glue to attach the LED to the heat sink?

Hobbyst46 wrote:Hello

1. Is this a 20W LED?

2. I would suggest to test the efficiency of the heat sink. As it looks now, the wiring are in contact with it and are prone to be warmed and the isolation sleeves are under potential risk of damage.
Moreover, when the microscope is in use, standing upright, and since the heat sink has no fins, it will output heat to the bench below and, more importantly, to the microscope chassis itself - possible risk to optics, apertures etc.
3. If heating appears to be excessive, one possible remedy is to place the microscope on a flat aluminum plate, somewhat larger than the base, and thickness of 4-5 mm (my guess), to conduct the heat away. Must have good contact with the heat sink, and this is tricky because of the round cross section of the heat sink.
Maybe make a cradle-like spacer, of aluminum foil, taking care not to shortcut the LED wires. If the heat sink is coupled to the aluminum plate, heat conduction away from the microscope will be improved.

Maybe my suggestion is an overkill, perhaps the microscope was designed to withstand a hot light source in the base without damage.

P.S. did you use heat conducting glue to attach the LED to the heat sink?

1. Yes...

2. Temperature 38ºC, when variable resistor at it maximum, 60ºC when resistor at its minimun... after 30 minutes in use.

3. Very good suggestion.

All your suggestion are very good for me and I appreciate all of them!!! TAHNK YOU!!

P.S. The LED is attached with 3 screws and a paste to transmit the heat.

Regards

stjepo wrote:

Hobbyst46 wrote:Temperature 38ºC, when variable resistor at it maximum, 60ºC when resistor at its minimun... after 30 minutes in use.

The 38C is too high, let alone the 60C!!. And, after an hour or so in use, will probably reach even higher levels. These levels are because of the huge 20W power and the lack of cooling fins on the heat sink.
In my opinion, improving the heat sink is a must. Your target temperature should be below 27-30C.
I believe that you can keep the resistor at its minimum and do not need to reach the maximum. Because in all probability the LED is an overkill.

Another point: I would suggest having a small stock of neutral density filters or an adjustable neutral density filter.
Please protect your eyesight - with the 4x and 10x, the light through the eyepieces can be blindingly intense.

P.S. Is the LED located at the same point as the original light bulb of the microscope? this is very important.

Hobbyst46 wrote:

stjepo wrote:

Hobbyst46 wrote:Temperature 38ºC, when variable resistor at it maximum, 60ºC when resistor at its minimun... after 30 minutes in use.

The 38C is too high, let alone the 60C!!. And, after an hour or so in use, will probably reach even higher levels. These levels are because of the huge 20W power and the lack of cooling fins on the heat sink.
In my opinion, improving the heat sink is a must. Your target temperature should be below 27-30C.
I believe that you can keep the resistor at its minimum and do not need to reach the maximum. Because in all probability the LED is an overkill.

Another point: I would suggest having a small stock of neutral density filters or an adjustable neutral density filter.
Please protect your eyesight - with the 4x and 10x, the light through the eyepieces can be blindingly intense.

I agree, this system as presented is definitely too high-intensity at 20W! I use a variable ND filter as used on the front of a camera lens - I simply place it over the field-illumination and turn the filter to get the level of light needed - allowing the LED (or in my case a 50W halogen) to be left at nearly it's maximum (which here should perhaps be nearer to 10W for an LED I suspect).
I've posted before re this system and it works perfectly. I took comparison images using the variable ND filter and the 'lollipop' ND filters as used with my Leitz Orthoplan from new - the results were identical - the variable ND filter works perfectly as a light-controller as it were....

Here's a link to the filter I use - variable ND filter

Here's a link to my post re this filter in use - variable ND filter post

Here are a few images of mine, variable ND filter in use over the light-source cover-glass of my Orthoplan 'lollipop' ND filter in use 'Lollipop' ND filters and variable ND filter side-by-side This system is really very good, and allows your bulb, LED or halogen-bulb to be left at the optimal almost-full-on setting..... no white-balance changes and no undue stress upon the bulb's filament. I use a blue filter at the 'other-end' of the light-train - at the output of the lamp-housing at the rear of the Orthoplan.

John B.

Hobbyst46 wrote:

stjepo wrote:

Hobbyst46 wrote:Temperature 38ºC, when variable resistor at it maximum, 60ºC when resistor at its minimun... after 30 minutes in use.

The 38C is too high, let alone the 60C!!. And, after an hour or so in use, will probably reach even higher levels. These levels are because of the huge 20W power and the lack of cooling fins on the heat sink.
In my opinion, improving the heat sink is a must. Your target temperature should be below 27-30C.
I believe that you can keep the resistor at its minimum and do not need to reach the maximum. Because in all probability the LED is an overkill.

Another point: I would suggest having a small stock of neutral density filters or an adjustable neutral density filter.
Please protect your eyesight - with the 4x and 10x, the light through the eyepieces can be blindingly intense.

P.S. Is the LED located at the same point as the original light bulb of the microscope? this is very important.

You are right, remember I told you that I used what I had at home. Do not reach much higher levels than 38ºC at maximum after 2 hours (still too high), it does if it is at minimum.

Agree "improving the heat sink is a must".

"I would suggest having a small stock of neutral density filters or an adjustable neutral density filter." very good advice.

Yes "the LED is an overkill" with the minimum light (maximum resistance) works perfect...

Regards

75RR sent this interesting link:


http://www.microscopy-uk.org.uk/mag/art ... ersion.pdf

Regards