Modifying a Labophot for use with Hi CRI LED

[mawyatt]

Chris mentioned putting the electrolytics on the 60 Hz input side (frequency doubled to 120Hz because of full wave rectification), this is a good idea. The circuit will attempt to keep the LED current constant, but the input voltage will be following a 120Hz ripple with highly varying amplitude, this may actually fall out of regulation during the ripple minimums. When this happens the LED current will be "modulated" at 120Hz and the light output will also be modulated.The input filter caps will help create a somewhat average input voltage and reduce the ripple.

The input cap for ripple attenuation seemed a necessity as at the low brightness end of the pot setting I could clearly see flicker, and the cap fixed this. I agree that an electrolytic at the output probably isn't doing too much good.



Also, the MT3608 board that I used from AliExpress for $0.54 per board matches the picture Rik references in his post, so they are all most likely built from the same schematic.

So your MT3608 schematic matches the top portion (printed) of the above schematic with the pot used as a proper potentiometer (3 terminal) rather than having the wiper shorted to a pin and used as a variable resistor (2 terminal)?

[glmory]

The boards I bought were these. They definitely don’t match the schematic in that two pins are shorted together. That seems to have been the primary reason I failed boards although bad wiring may have also failed some.

WINGONEER 5PCS MT3608 DC-DC Step Up Power Apply Module Booster 2A Power Module for Arduino https://www.amazon.com/dp/B06XHJCHX6/re ... rAbT02NVGA

[rjlittlefield]

WINGONEER 5PCS MT3608 DC-DC Step Up Power Apply Module Booster 2A Power Module for Arduino https://www.amazon.com/dp/B06XHJCHX6/re ... rAbT02NVGA

Interesting -- and perhaps the most bemusing product description I have ever read:

Product Description
Matters needing attention:
...
2. The peak current output current no more than TV university.

--Rik

[mawyatt]

Seems these boards based upon the MT3608 are everywhere on eBay, different versions, and the sellers have no idea what they are or how they work!

Since these boards are all (one's I've found) configured as boost switch mode regulators the output voltage must always be greater than the input, but I didn't find that spec anywhere either.

Another interesting thing is the schematic shown allow the output voltage to rise to ~27.9 volts, which means the chip "sees" a voltage at pin 1 of Voutput + a forward diode drop or about ~28.7 volts. This exceeds the maximum voltage specification for the chip of 28 volts! If the 100K pot has a 10% tolerance the voltage could rise to ~ 31.4 volts, a simple worst case analysis with 5% resistor tolerance and 10% pot tolerance yields ~32.8 volts

https://www.olimex.com/Products/Breadbo ... MT3608.pdf

Good news is they are dirt cheap, so smoking a couple doesn't cost much...other than hurt ego, no harm

Best,

Mike

[ChrisR]

Also note the boards evidently do not have the 22uF capacitors

That's what I was wondering, though having looked things up capacitors these days are a lot smaller than I thought. 1206 package is 3.2mm x 1.6mm x 0.55mm. They could be?

[mawyatt]

Also note the boards evidently do not have the 22uF capacitors

That's what I was wondering, though having looked things up capacitors these days are a lot smaller than I thought. 1206 package is 3.2mm x 1.6mm x 0.55mm. They could be?

Chris,

Don't think they can squeeze a 22uF cap with any reasonable breakdown voltage (would need to be >28 volts) into a 1206 package. My guess would be these are just decoupling for the ~1MHz switching frequency, so 0.1 or 0.22uF.

Best,

Mike

[aidanmoore]

So your MT3608 schematic matches the top portion (printed) of the above schematic with the pot used as a proper potentiometer (3 terminal) rather than having the wiper shorted to a pin and used as a variable resistor (2 terminal)?

I photographed a new board from AliExpress below, and it appears to match the reference schematic, with a direct connection of the pot wiper to pin 3


Here is the schematic traced from the board and resistor values referencing the schematic



This is the connection to external R1 and R2 resistors:



To get the vin/vout ratio below I measured R1 = 98K and R2 =2.4K + 2.2K = 4.6K




I still have not tried the series 5 ohm resistor as I am looking for a suitable power rated resistor.

[glmory]

I still have not tried the series 5 ohm resistor as I am looking for a suitable power rated resistor.

The resistors I am using are:
https://www.amazon.com/gp/product/B0087 ... UTF8&psc=1

They are 50W resistors which are obviously over-kill. There is a lot of room in the Labophot though so it doesn't really matter that they are too large.

Interesting that the board I bought on Amazon is somewhat different. If I remember right Amazon delivered much quicker so I bought that one. Eventually it worked out, but was more trouble than I expected.

I increased the 5 ohm resistor to 10 ohms, reduced the resistance to ground and added some additional capacitance. Here is what I have now:



Overall this runs quite well. It now runs from about 0.06-0.5 amperes and is cooler than before the LED upgrade while being a lot brighter. There are three issues, but they are relatively minor and probably can be fixed if I decide to put in the effort:

1. It is still a bit too bright at the lowest setting and its color is a bit off from what I would prefer, a blue filter should be able to fix both problems.

2. I still get ripple when using an electronic shutter at very high shutter speeds and high LED power. I can get about 1/1600 second with bright field and about 1/60 seconds with phase contrast on a 10X 0.25 NA objective though. This is really good enough I might just leave it as is. Adding some more capacitance may be able to fix the problem though so I will likely try.

3. It runs backwards. 1 is high and 6 is low. This isn't as annoying as I anticipated so I may not bother fixing it. Rewiring the pot should be able to fix it, but the one time I tried it seemed like the rewiring burned up my board. The pot is connected to some other circuitry which I haven't investigated, some adjustments may need to be made to that circuitry to make it run normally.

[rjlittlefield]

Also note the boards evidently do not have the 22uF capacitors

That's what I was wondering, though having looked things up capacitors these days are a lot smaller than I thought. 1206 package is 3.2mm x 1.6mm x 0.55mm. They could be?

Don't think they can squeeze a 22uF cap with any reasonable breakdown voltage (would need to be >28 volts) into a 1206 package. My guess would be these are just decoupling for the ~1MHz switching frequency, so 0.1 or 0.22uF.

I had a few spare minutes tonight to disassemble one of these boards to get an unambiguous out-of-circuit measurement. The two caps measured 11.3 and 11.8 microfarads shortly after unsoldering. Google search on 1206 capacitor and some quick scanning brought me to HERE which specifies a Murata GRM31CC8YA106KA12L as 10 microfarads, 35 VDC, 3.2mm x 1.6mm x 1.6mm, multilayer ceramic, $0.178 each in quantity 50. All those specs are quite close to the parts I have in hand, except the manufacturer of these boards is obviously getting them a lot cheaper than $0.17 each, considering I got 5 boards for $1.40 each, shipping included.

The Wikipedia article on ceramic capacitors says that "The minimum thickness of the dielectric layer, which today (2013) for low voltage capacitors is in the size range of 0.5 micrometers..." Things have gotten tiny!

Gerd posted an image of some capacitor (not these) at http://www.photomacrography.net/forum/v ... hp?t=34255 .

--Rik

Edited to add: see http://owenduffy.net/blog/?p=12250 for discussion of performance of various voltage booster cards.

[mawyatt]

Also note the boards evidently do not have the 22uF capacitors

That's what I was wondering, though having looked things up capacitors these days are a lot smaller than I thought. 1206 package is 3.2mm x 1.6mm x 0.55mm. They could be?

Don't think they can squeeze a 22uF cap with any reasonable breakdown voltage (would need to be >28 volts) into a 1206 package. My guess would be these are just decoupling for the ~1MHz switching frequency, so 0.1 or 0.22uF.

I had a few spare minutes tonight to disassemble one of these boards to get an unambiguous out-of-circuit measurement. The two caps measured 11.3 and 11.8 microfarads shortly after unsoldering. Google search on 1206 capacitor and some quick scanning brought me to HERE which specifies a Murata GRM31CC8YA106KA12L as 10 microfarads, 35 VDC, 3.2mm x 1.6mm x 1.6mm, multilayer ceramic, $0.178 each in quantity 50. All those specs are quite close to the parts I have in hand, except the manufacturer of these boards is obviously getting them a lot cheaper than $0.17 each, considering I got 5 boards for $1.40 each, shipping included.

The Wikipedia article on ceramic capacitors says that "The minimum thickness of the dielectric layer, which today (2013) for low voltage capacitors is in the size range of 0.5 micrometers..." Things have gotten tiny!

Gerd posted an image of some capacitor (not these) at http://www.photomacrography.net/forum/v ... hp?t=34255 .

--Rik

Rik,

I'm surprised they can get 10uF @ 35V in a 1206 package, would have thought that 2.2~4.7uF @ 35V would be the largest cap in that size!!

However, this a Class 2 capacitor (we never use Class 2 caps for obvious reasons). It's a really poor capacitor as shown, but for this cheap DC-DC converter, probably OK to use and keep the cost as low as possible.

Best,

Class 2 Capacitors :

Class 2 capacitors have following characteristics

Non-Linear temperature coefficient : Their capacitance varies non linearly with temperature
High Voltage dependency : Their capacitance highly depends on the applied voltage
Suffer Aging : They suffer significant aging process due to ferro electric materials used in manufacturing
High electrical losses : Have significantly high electrical losses compared to Class 1 capacitors
High Permittvity : Due to high Permittivity they have high volumetric efficiency
High Capacitance values : Due to high Permittivity large capacitance values can be achieved in small device casings like 0201, 0402, 0603
Used in smoothing applications : Due to the above properties they are used in applications like bypassing, coupling, decoupling.
Have low stability and accuracy