optiphot transformer replacement

[DavyC]

Shame about the 100/120 v instead of 120/240 v selection. The voltage 'jump' could be down to a worn/dirty pot. The thermal security device, might actually be a power transistor or voltage regulator. It's providing the muscle for the control element of the board circuit. Transformer rewind not always straight forward, but I would spec grain orientated m6 steel laminations here if there is an option in the rebuild.

[mawyatt]

Mike, that is what I did for the cold-light source a while ago. Only there I could easily replace the existing knob and burnt pot-meter with the power switcher.
Now I have an excellect power wheel with off switch, well placed and easy to use. I would like to keep it.

As I see it, I can use such a power supply indeed. Only the physical space in the stand limits the choice. They produce a reasonably stable 12V that I can connect just after where now is the rectifier and leave most of the rest as it is. The power off-switch will of course go before the PS. The volt-meter I will replace by a new digital one. I could do without it, but why having a dead meter or a hole if I can have a replacement for next to nothing.

Then the auxiliary 12V lines can be branched off immediately after the PS, and also a 5V step-down module too to feed the arduino and possibly the new volt meter.

I'll make a new sketch tomorrow.

The biggest question I have: There are constant voltage and constant current sources, and it's not always clear on ebay what it is. Does that matter here?

Wim

Highly non-linear devices like diodes & LEDs are better operated with constant current supplies, since they are fundamentally current devices. Regular light bulbs have a nice non-linear relationship that allows both current and voltage modes to be effective, so I would think either mode would be effective.

Best,

[iconoclastica]

The voltage 'jump' could be down to a worn/dirty pot. The thermal security device, might actually be a power transistor or voltage regulator.

You might be right there. It looked familiar, but I couldn't place it. After I read this, I went up to the attic where I keep a bag of electronic left-overs and found similar things right away. It looks vry much like this power transistor only with all the markings underneath a layer of white paint. Thanks.

I don't think though that the volt-meter jumping is caused by the pot. For the lamp is just one resistor away and volts up smoothly. BTW, 'jump' is not the best description: it goes down to 5 or 6V and then, while still reducing V, jumps up to 9 and comes down again. I have no good contacts there to meter it myself in parallel.

[iconoclastica]

Here's the corrected schematic (absolutely not guaranteed free of mistakes):


What I intend, is:


And here's for the records what's on the pcb:

[DavyC]

You mentioned 5 capacitors, but only 3 in your schematic. Ok, with a new transformer in place you are going to have 2 additional regulators. Assuming the extra loading is fine, to avoid possible interaction, maybe prudent to fit additional bridge rectifier in the supply line to regulators. A smoothing cap then, at the dc side of the rectifier. I would go with a low voltage dropout Schottky bridge. I only suggest what I would try. It's up to you how you proceed.

[iconoclastica]

Ay, I hoped you wouldn't notice...

Yes, the board is rather too small for me. I wasn't until I looked at it under the stereo micoscope that I saw that one of them was actually a diode and the last one wasn't at all.

[iconoclastica]

The transformer in the Optiphot is not suitable to convert to 240V. It had to go. I have unsoldered it, as well as the rectifier. The 220/240V transformer in my Labophot is quite a different thing. I first thought it was a home made adaptation (it still may be a replacement), but it corresponds with the specs on the silver sticker on the outside of the stand.

Now I have a 12V PS from a laptop connected to see if that will work. So far, it doesn't. It powers the lamp, but the front panel's pot-meter is ignored, so it tries to output full power (which is just too much for the PS, so it switches off if the lamp is in place).

The resistance I measure over the front panel pot equals that of the trimpot on the board, no matter what I turn it to. Before, it ran from 0 to 35K ohms. Disconnected it goes from 0 to 100k.

The volt meter is also a big puzzler. Why does it want 4 wires? The only reason the transformer outputs 4V too is to feed it to the volt meter, so I took care to do the same after I had removed the original transformer. However, I did a bad soldering job and it got disconnected again. But then it still worked...

On the brighter side, chimerizing my Labophot into a Labtiphot went without any problem. The columns of both types are bound to their feet with four hex screws and despite the differences, they are equally spaced. When detached, the know of the condenser lift is easliy accessible and could be moved to the right (so the left) side. Stage and head are identical, so now I have a Labophot with detachable turrets. Which was the primary reason to buy the Optiphot. Only the brighter lamp would be a nice extra, shining through leaves while primary lighting is reflective. And it must come to live again or I won't be able to sell it on. Which was a rather essential step in the financing plan.

Have any practical solutions been thought of to store the extra nose piece while not in use?

[iconoclastica]

I have come out of darkness and seen the light. We decided that the pcb probably had had it when we simultaniously blew the fuse and the lamp. So I stripped the bottom completely bare except for the fuse (not decided yet) and the potti. Since I have a couple of future plans that require some form of intelligence in the microscope, it was the easy choice to put in an Arduino (uno, for the moment) and have that PWMing a mosfet module to dim the lamp.

Not being well-informed about the Arduino's PWM, I went for the difficult option, and to my surprise it worked immediately. A smooth transition from no light to max with no dead zones. Then, reading up the PWM instructions, I tried the easy option and the light went out and stayed out. The voltage after the mosfets was way too low and stayed low, whatever I tried. It took me half a day to get back to a working setup again.

It's all temporarily in place, not fixed, for the real estate planning is still in progress. There's only so much space in the foot of the stand. I want to make two power outlets, one regulated 12V and one continuous 12V for the epi-lamp. And I need to find a place to let the usb out, to update and communicate with the Arduino.

If I decide to keep the power supply externally, then there's plenty of space both inside and as port holes in the frame. But will it be an internal AC/DC converter, then I am to apply some more creativity.





-------
edit: Now it's ready, this phase at least, and I am relaxed enough to do some experimenting, the simplest Arduino way (analogWrite) works too. I am ##### sure it wasn't this morning. Sometimes the world confuses me...

[mawyatt]

Great solution!!

I'm using the same 12 to 5 volt converter you show, be careful when you first power it up as mine have come up with the output at 12v initially until I adjusted the voltage to 5 volts

12 volts on the 5 volt rail can easily take out most electronics like the Arduino or Raspberry Pi, so I always measure and set the output voltage first then connect the rest of the electronics.

BTW if you run into any weird Arduino behavior, this might be caused by not having enough decoupling capacitors on the converter input and output. A few hundred microfarads will usually do.

Best,

[iconoclastica]

Mike,

I understood the voltage regulator of the (this) Arduino could handle 5-12(-20)V. Anyway, I took the precaution you advise. Only afterwards I upped it to 5.9V for apparently during power up, too much power drained to other components and the board didn't boot up well. Since I changed it to nearly 6V, it always starts well.

Wim

[mawyatt]

Wim,

Just checked the specs and the Arduino you use (uno) has a on board regulator so it can handle the larger voltage, the Raspberry Pi I have and use doesn't have a regulator on board, so I have to supply the regulated 5 volts externally.

You can probably operate the Arduino uno directly from the 12 volts without the converter, at least the specs and schematic indicate so. If you want to keep using the converter then the minimum voltage according to the spec sheet is 6 volts and the recommended minimum is 8 volts. The schematic show a series reverse voltage protection diode into the 5 volt regulator input, so using 8 volts is probably a better voltage.

Best,

[iconoclastica]

I am going to reconsider the raspbery pi instead of arduino very seriously. The arduinos have unresolved issues with their serial communication causing many unexpected program restarts. And I am surprised that c++ still hasn't been banned by the Geneva convention. I have been a software engineer for some 30 years, but it still surprises me what bad language it is.

Today I wanted to screw back my microscope and get it fully operational again. During the summer season I want to use it for what it was made for instead of tinkering with it. I choose the opportunity to burn the program onto the much smaller Arduino mini. From that moment I was back in the dark. As soon as I connected the lamp, the power supply switched itself of to prevent overdraft.
For quite some time I suspected the lamp. When cold, the resistance is rather low, allowing a strong inrush when switching on. I am surprised that that hasn't given problems I am aware of.

Only after several hours I found the mosfet module defect (always fully open). Apparently they are not entirely reliable. Luckily I had bought a good number of them.

[Saul]

Why do not use this:
https://www.ebay.com/itm/Switch-Dimmer- ... 2772987180 ?

All my leds are controlled by these cheap dimmers ...

[mawyatt]

I am going to reconsider the raspbery pi instead of arduino very seriously. The arduinos have unresolved issues with their serial communication causing many unexpected program restarts. And I am surprised that c++ still hasn't been banned by the Geneva convention. I have been a software engineer for some 30 years, but it still surprises me what bad language it is.

Today I wanted to screw back my microscope and get it fully operational again. During the summer season I want to use it for what it was made for instead of tinkering with it. I choose the opportunity to burn the program onto the much smaller Arduino mini. From that moment I was back in the dark. As soon as I connected the lamp, the power supply switched itself of to prevent overdraft.
For quite some time I suspected the lamp. When cold, the resistance is rather low, allowing a strong inrush when switching on. I am surprised that that hasn't given problems I am aware of.

Only after several hours I found the mosfet module defect (always fully open). Apparently they are not entirely reliable. Luckily I had bought a good number of them.

I'm certainly not a programmer, nor software engineer, but c++ has always been hard for me to follow, Python is much easier, and what I'm using. I can't complain about the Raspberry Pi, it's handling everything I'm doing which involves a Complex S&S System with 2 THK KR15 and KR20 linear rails with a magnetic homing sensor on the Z axis, camera interface and strobe delay & interface. It's a completely hands off S&S System once the XYZ position parameters are established or loaded. Amazingly little computer IMO.


Yes incandescent bulbs have a very low cold resistance and only reach rated resistance when at full temperature. What you experienced with an initial over-current shutdown due to the startup cold bulb resistance is not uncommon. If you want to stay with a voltage mode for your bulb you could slowly ramp up the voltage, that should keep the initial current lower.

Best,

[iconoclastica]

Why do not use this:
https://www.ebay.com/itm/Switch-Dimmer- ... 2772987180 ?

All my leds are controlled by these cheap dimmers ...

For an entirely external light such a dimmer is excellent. But in this case, the on/off switch and power control potmeter are integrated in the stand and I want to continue using them.