Modifying a Labophot for use with Hi CRI LED

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

aidanmoore
Posts: 104
Joined: Thu Sep 24, 2015 8:09 am
Location: Toronto

Modifying a Labophot for use with Hi CRI LED

Post by aidanmoore »

Hi,

I hope this post may be of interest to anyone considering swapping an incandescent lamp out for a LED source on the older Labophot 1 series microscopes.

I recently purchased a number (!) of very inexpensive Labophot microscopes on eBay in the hopes of getting one good unit with all parts in good shape.

After building my keeper microscope, I had a number of sort-of-working scopes left over and one of them just needed a new transmitted light source.

It seems the transistor that drives the incandescent light was blown, and I did not have success with a direct substitute transistor (2N3716)

The 12V DC output of the powersupply was operational, so I used that to power a DC DC converter with an input and output decoupling capacitor.

I used two off the shelf components:

From Aliexpress :

https://www.aliexpress.com/store/produc ... 72484.html

And Yuji LEDs:

https://www.yujiintl.com/packages/displ ... ge=BC-135L

I was able to remove the trimmer pot from the DC DC converter and wire in the 1K Labophot potentiometer in series with approximately 100K resistor (actually another trimpot to fine tune the current)

The LED was mounted on a TO220 heatsink selected to fit the available space in the lamp housing. I used a thermal adhesive to attach the LED to heatsink.

The results are very good. I monitored the LED using a FLIR camera and after 45 minutes the temperature was in the 75C range. The original Labophot Potentiometer adjusts the light intensity from off to a level that is comparable to the original lamp.






Image
Image
Image

Update showing actual light and thermal measurements

Images below show the original incandescent lamp in the labophot in IR and using a lightmeter coupled to the transmitted light port at full aperture:

Image
Image

And the LED is of course much cooler and brighter


Image
Image
Custom Nikon/Thorlabs Microscope Nikon Z7ii ZCAM E4

enricosavazzi
Posts: 1474
Joined: Sat Nov 21, 2009 2:41 pm
Location: Västerås, Sweden
Contact:

Post by enricosavazzi »

If the illuminator works satisfactorily, then the following comment is probably not relevant in practice. If you have a heat problem, on the other hand, it may point to a possible solution.

My main concern is that the light emitting surface in a multi-chip cob is much wider than the filament of the halogen lamp for which the illuminator, its collimator and the condenser were originally designed for. This means that much of the light emitted by the surface of the cob is wasted, and does not enter the condenser. Most of it is probably reflected, diffused and/or absorbed by opaque parts of the chassis surrounding the optical path from illuminator to condenser, and have no other effect than heating up these parts (this stray light might become a flare source, or a source of distraction/discomfort for the user, if it leaks out of the chassis/illuminator).

Since the total amount of power used by a small cob like this one is lower than the power of the original lamp, this is unlikely to cause much heating anyway, unless you turn to 50W or 100 W cobs hoping to get a stronger illumination (which most likely will give you plenty of heat but no significantly stronger illumination, for the reason explained above). I would regard a temperature of 75 C (or anything higher than comfortable to the touch) as quite high for a LED and would try to do something about it, but opinions may differ.
--ES

Pau
Site Admin
Posts: 6051
Joined: Wed Jan 20, 2010 8:57 am
Location: Valencia, Spain

Post by Pau »

Very nice and (relatively simple) adaptation, of course it is of interest!

Does the MT3608-DC-DC-Step-Up-Power-Apply-Module-Booster-Power-Module actually control the intensity or does it through the voltage control? In the first case it could be a good alternative to the costly LED drivers.
enricosavazzi wrote:My main concern is that the light emitting surface in a multi-chip cob is much wider than the filament of the halogen lamp for which the illuminator, its collimator and the condenser were originally designed for. This means that much of the light emitted by the surface of the cob is wasted, and does not enter the condenser....
I've seen this issue with Retrodiode illuminators: from the nine actual LEDs only four (maximum) could be focused at the rear focal plane in Köhler.
This is why I now use a Cree XM-L that has approximately the same size of the lamp filament.
If you can remove any frosted glass of the light path, you would be able to see what part of the COB LED are you actually profiting. Anyway, if you get enough light as you show the extra light will be of little concern IMO.
Pau

aidanmoore
Posts: 104
Joined: Thu Sep 24, 2015 8:09 am
Location: Toronto

Post by aidanmoore »

Hi,

Thanks very much for the comments.

Regarding the thermal concern, I did back off on the maximum voltage to the LED in order to reduce peak current and temperature. I found that I could still get 3X the brightness of the incandescent lamp and keep the LED at about 50C.

The LED intensity is voltage controlled from below 13.5V to about 14.5V.

The MT3608 is a very low cost DC voltage in to DC voltage out boost device with an output voltage determined by the equation below. (Copied from MT3608 datasheet)
Image

It just so happens that the Labophot DC voltage at the output of the rectifier is about 12V no load, so I was able to use the 1K labophot potentiometer in series with a small trimpot (5K as I recall) as R2 and a fixed 100K resistor as R1 to control the output voltage between about 14.5V max and some voltage below the turn on threshold of the LED.

The fact that the Labophot native DC voltage is so close to the LED voltages means that the DC DC converter is dissipating only a small amount of excess power, and stays very cool boosting approx 12V to between 13.5 and 14.5V maximum.
Custom Nikon/Thorlabs Microscope Nikon Z7ii ZCAM E4

glmory
Posts: 25
Joined: Fri Dec 30, 2016 12:45 pm

Parts list

Post by glmory »

This seems like the best adaptation of the Labophot to LED I have seen as it is relatively low cost and uses the existing potentiometer to adjust the lighting level. Since my lighting is slowly dying, I am going to give it a shot.

Based on the comments, I looked a little for a smaller LED which would light up between 13.5 and 14.5 volts and give similar light levels. For some reason I failed. Either I am bad at searching, or that is not a common enough voltage for LEDs to have a lot of choices. Therefore I will try using the same LED as you.

Just to confirm, the parts required are:
1. A MT3608 DC-DC Step Up Converter
2. A BCL 135L LED, Maybe the 4000 K one?
3. A 100k resistor
4. A trimpot, perhaps https://www.amazon.com/Doradus-Trimpot- ... 01DM4HC1U/

Is there anything I am missing? As I understand it, you use the existing potentiometer to set the voltage output on the DC-DC converter and therefore control the LED output.

Pau
Site Admin
Posts: 6051
Joined: Wed Jan 20, 2010 8:57 am
Location: Valencia, Spain

Post by Pau »

Based on the comments, I looked a little for a smaller LED which would light up between 13.5 and 14.5 volts and give similar light levels. For some reason I failed. Either I am bad at searching, or that is not a common enough voltage for LEDs to have a lot of choices.
This is because LEDs actually operate at low voltage. COB LEDs are made of several LEDs packed together and connected in series so the voltage to operate them is higher. In most cases you can see that higher power COBs require higher voltage.
Pau

aidanmoore
Posts: 104
Joined: Thu Sep 24, 2015 8:09 am
Location: Toronto

Post by aidanmoore »

Hi,

Thanks for your comments.

Here is how I recall the assembly was performed (unit is sold so I cannot confirm):

1. Locate the 12V DC output of the bridge rectifier

2. Remove the existing transistor and locate the wires from the incandescent bulb

3. LED assembly is two solid jumper wires spaced the same as the bulb footprint and soldered to the LED. This allows the LED to plug into the incandescent socket.

4. LED is mounted to the heatsink (fins backwards)with thermal adhesive to a TO-220 heatsink such as:
5. https://www.sparkfun.com/products/121

6. The tricky part is removing the trimpot from the DC DC converter and wiring in the labophot pot in series with a trimpot.

7. Those DC DC converters have a Vout/Vin relationship that makes it easy to blow up the part if you accidentally set R2 near 0 (denominator of equation). You also have to get R2 fine tuned with the Labophot pot and 5K trimmer in series to give you a range that goes from LED dark to about 13.5 V . You can go higher but the LED gets quite warm at full voltage.

8. Trimmer pot is 5 or 10K with at least 10 turns. Mine was 25 turns like:
9. http://www.taydaelectronics.com/5k-ohm- ... 3296w.html

10. I’d suggest getting a few DC DC modules as they are cheap and handy to have around. You might accidentally blow one up as well.

As far as color temperature is concerned, I think the original bulb is around 3200K, but I always color correct that with a blue filter, so I went with 3200K and used a blue filter in the condenser. You could try a bluer (4K/5.6K) if you do not want to use a CC filter.

The purchase cost of the LED is high as you have to buy qty 10. I purchased 10 for $80.00 plus 25 shipping, so I could sell 1 or 2 for 10.00 each + shipping if you like. On the other hand I found these LED’s really useful for other lighting projects, so you might want to order a quantity for yourself.

Good Luck with your project!

Aidan
Custom Nikon/Thorlabs Microscope Nikon Z7ii ZCAM E4

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

Aidan,

Nice work!!

LEDs are Light Emitting Diodes, and as a semiconductor diode have a very nonlinear voltage to current relationship, exponential in fact. This causes a small voltage change to have a large current change, which is what you don't want since LED current is what produces the light output and what you want to control.

I think you will find that if you include a small power resistor in series with the LED the sensitivity of the LED light output to control voltage will be much less. This will make your Trimmer Pot setting much less sensitive and also help protect the LED and converter from possible damage. The resistor isn't critical in value.

If you know the LED operating current then select a resistor (R) that will drop about 2~3 volts (V) at this current (I) based on R=V/I. R will dissipate power as I*I*R, or V*V/R, so select a resistor with a power rating greater than this value.

Say I is 0.6 amps, then select R as ~5 Ohms at 2W or greater (two 10 Ohm 1W resistors in parallel will work too).

Best,

Mike

glmory
Posts: 25
Joined: Fri Dec 30, 2016 12:45 pm

Post by glmory »

I certainly would be interested in buying 2 LEDs off of you if you are willing to sell. Not sure the best way to do the transaction though, ebay?

"I think you will find that if you include a small power resistor in series with the LED the sensitivity of the LED light output to control voltage will be much less. This will make your Trimmer Pot setting much less sensitive and also help protect the LED and converter from possible damage. The resistor isn't critical in value. "

If I am not mistaken he solved this problem already in an elegant fashion with the DC DC converter. Once the trimmer pot is set correctly the DC DC converter is connected such that it alters the value from the built in potentiometer into about a 13-14.5V signal. That way the full range of the potentiometer can be used to fine tune the light output of the LED. Since the output of the DC DC converter is set so it won't over-voltage the LED, a resistor isn't really needed and just adds heat.

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

glmory wrote:I certainly would be interested in buying 2 LEDs off of you if you are willing to sell. Not sure the best way to do the transaction though, ebay?

"I think you will find that if you include a small power resistor in series with the LED the sensitivity of the LED light output to control voltage will be much less. This will make your Trimmer Pot setting much less sensitive and also help protect the LED and converter from possible damage. The resistor isn't critical in value. "

If I am not mistaken he solved this problem already in an elegant fashion with the DC DC converter. Once the trimmer pot is set correctly the DC DC converter is connected such that it alters the value from the built in potentiometer into about a 13-14.5V signal. That way the full range of the potentiometer can be used to fine tune the light output of the LED. Since the output of the DC DC converter is set so it won't over-voltage the LED, a resistor isn't really needed and just adds heat.
Here's a link to LED operation.

https://en.wikipedia.org/wiki/LED_circuit

Note that LEDs have a exponential relationship between voltage and current, so a small voltage change causes a very large current change. This is common among semiconductor diodes.

Because of this relationship the trimmer setting will be very sensitive (reason for multi turn trimmer), and may require periodic adjustment due to temperature, aging and so on. With the inclusion of the series resistor with the LED, the relationship between voltage and current is altered in favor of a more "linear" relationship and reducing the voltage sensitivity, and thus trimmer sensitivity.

Yes some power is wasted, about 2 watts with the values shown, but this isn't much and considering the improvement IMO.

A resistor is pretty cheap and easy to place in series with the LED, and you will be rewarded with a much less sensitive LED control.

Beyond the scope of this thread is a concept that forces the LED current which is under control, this gives precise LED current setting and doesn't depend on the LED characteristics nor input voltage to a 1st order. This is the mode and how dedicated LED controllers work.

Interestingly you can use the same DC to DC converter voltage control mode shown with the same components and the series LED resistor I mentioned to operate with current mode control. A single wire is changed and rearranged and gives a 10X (or more) current control range which is highly linear and not sensitive to settings.

Anyway, I suggest you try the series resistor and see for yourself.

Best,

Mike

aidanmoore
Posts: 104
Joined: Thu Sep 24, 2015 8:09 am
Location: Toronto

Post by aidanmoore »

Hi Mike,

Thanks for the suggestion.

The resistors look like a good trade-off between resistor sensitivity, and increased power output from the DC to DC.

There's lots of space in the base of the unit with bare metal, and the same thermal adhesive I used for the LED could hold those warm resistors to the base as a heatsink.

I might actually try that on a working Labophot base because the higher LED light output (about 3X) is useful for darkfield illumination.

Aidan
Custom Nikon/Thorlabs Microscope Nikon Z7ii ZCAM E4

glmory
Posts: 25
Joined: Fri Dec 30, 2016 12:45 pm

Rewiring hasn't been working

Post by glmory »

I am now up to 7 burned up boards (well 5 burned up boards, 2 which output the input voltage minus 0.3 volts whatever I do). Another batch is in the mail and I imagine I can't burn up so many more before something works, but I have tried every configuration that made sense to me to me and none seemed to work.

Image

The first issue has been, where should R1 and R2 connect? I have been assuming R1 connects the pins I labeled 2 and 3 and R2 connects between pins 1 and 2. I tried a few other ideas for how I was supposed to wire it but none seemed to work.

The second issue is what value should R1 and R2 be?

I arbitrarily have a 100k resistor for R1.

So using the equation:
Vout=Vref(1+R1/R2)

To calculate for R2 needed is then

13.5=0.6*(1+(100,000)/R2), R2=4,650 ohms.

Therefore I started with R2 being the trimpot at 5k, in series with the 1k potentiometer on the front. This didn't seem to work. The last time I tried, I even added a 47k resistor in series to keep R2 from getting too low and it still burned up as soon as plugged in!

Second, no grounding is required of either negative terminal, correct? After it wasn't working I tried this in some of my versions but it didn't seem to make a difference.

Third, the capacitors are only needed for controlling ripple, and therefore are not needed for debugging what values of resistors to use, correct?

aidanmoore
Posts: 104
Joined: Thu Sep 24, 2015 8:09 am
Location: Toronto

Post by aidanmoore »

Hi,

Sorry to hear about the troubles with the DC DC converter.

I recall that it is very easy to accidentally overshoot the value of R2 as you reduce the resistance to increase the output voltage.

Perhaps you can use the trimpot while it is still installed in the board and adjust the output voltage until you get the desired 14.0V or so. This way at least you have a working board.

Then you can remove the trimmer and confirm the wiper terminal and values of the two resistors.

Regarding the capacitors, yes they are just to reduce possible flicker at the low end of the brightness, and do not need to be installed until everything else is working.

I'll be back from vacation on the 28th, and I am planning to use one of these DC-DC modules for a flash modeling light, so I can confirm where the resistors connect at that time.

Good luck with the next steps, and again sorry to hear of the difficulty. These DC-DC modules were not apparently designed with any thought to over-voltage protection.
Custom Nikon/Thorlabs Microscope Nikon Z7ii ZCAM E4

mawyatt
Posts: 2497
Joined: Thu Aug 22, 2013 6:54 pm
Location: Clearwater, Florida

Post by mawyatt »

Hello,

I've attached a schematic that uses the MT3608 as an LED "CURRENT MODE" controller. This requires just 2 resistors, one a current sense resistor (Rx) and another to scale the feedback potentiometer resistance to control the maximum allowed LED current (Ry).

The example shows the MT3608 with a BC-135 LED, a 5ohm 4watt resistor for Rx (can be a couple 10ohm 2watt resistors in parallel), and a 16Kohm resistor for Ry. Maximum LED current is limited to specification of 0.9amps, and minimum LED current is 0.12amps, with nominal at 0.6amps.

I have not built nor tested this circuit, but expect it should work just fine and hopefully keep folks from burning up their circuits :D

It should also work much better than the typical voltage control mode use with significantly less sensitivity to control parameters, components and LED characteristics, since LEDs are fundamentally diodes and thus current mode devices. The exponential relationship between applied voltage and current, and highly temperature dependency, makes voltage control difficult and very sensitive.

Hopefully some folks will try this circuit and provide some feedback.

Best & Happy Holidays,

Mike

Image

glmory
Posts: 25
Joined: Fri Dec 30, 2016 12:45 pm

Post by glmory »

aidanmoore wrote:Hi,

Perhaps you can use the trimpot while it is still installed in the board and adjust the output voltage until you get the desired 14.0V or so. This way at least you have a working board.
I need to try this again. This was actually what I did with the second to last board. Somehow I managed to break the board in the process, it was rather strange I turned the trimpot trying to get to 15.5 volts (I bought the 5 ohm series resistor for the LED so I believe I need an extra 2 volts at the terminal).

At some point the board just started reading about 12.5 volts and would not change with the turning of the trimpot. I removed the trimpot and wired in resistors, still stuck. I guess I somehow broke the sensing on the board. Got five more coming Wednesday though so I think this is the first thing I will do.

I am pretty sure my understanding of the equation which sets the output is wrong. When I first turn on a new board it gives 13.5 volts output with about 12.5 volts input. When I remove and measure the trimpot resistances I get 50k, 50k, and 100k between the three sets of wires. If an R1 and R2 of 50k gives 13.5 volts than that equation must work differently from my understanding. Maybe it is the voltage rise rather than the total voltage?

Post Reply Previous topicNext topic