Page 9 of 9 FirstFirst ... 56789
Results 81 to 88 of 88

Thread: Medical Argon Lasers???

  1. #81
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,890

    Default

    Quote Originally Posted by platinum View Post
    To be serious for a moment, let me suggest a flux capacitor.
    duct work eats airflow like there is no tomarrow. It is far better not to to duct, or you have to construct huge plenums. The flexible ductwork on your dryer eats about 50-75 CFM in a bend. two tight bends and your laser is COOKING. 68Bs when used for opthalmics dump into a 12x12 INCH plenum.

    I bolt 3, sometimes 4 patriots to huge holes in the tops of HGM 5s in the past for clients.

    A sixty X dissapates 1750 watts of heat at full current, a HGM5 is easily twice that.

    Steve

  2. #82
    Join Date
    Jul 2008
    Location
    Miami
    Posts
    506

    Default ar/kr

    This is my ar/kr I90ma system I once had . I built the head from spare parts. The tube was processed for white light. The supply is a alc 909 from Nitelite. But ran out of money for the white light optics so I sold it to Australia. System is still in service.
    Attached Thumbnails Attached Thumbnails IMG_0339.JPG  


  3. #83
    Join Date
    Apr 2008
    Location
    Albuquerque, NM, USA
    Posts
    19

    Default Problems repairing HGM5 power supply

    It has been 2 years since I blew my HGM5 power supply (publicly here on Photon Lexicon) trying to run an ALC-905 laser head. I thought the magnet wires on the laser head were for a temperature sensor and so I was trying to run it without any magnet. It did work but after running for about 45 minutes the power supply blew. I took a detour for a while building a CNC mill but I am now back to trying to repair the HGM5 power supply. All of the linear pass bank transistors and all of the switching transistors were blown. Several resistors on the power board and a transistor on the control board were bad. All of the capacitors on the power board were OK by capacitance but I can’t test ESR so I replaced them all. Several of the diodes on the power board were bad and replaced. As far as I can tell everything else is OK. I am not so sure about the high voltage blocking diode. It tests OK at low voltages. I have not found a suitable replacement for it.

    I now have 230 vac (rectified) across the laser head magnet so that compass north points to the anode end of the tube. I have made a light sensor that picks off part of the waste beam at the back end of the tube but I had to guess as to how much light should relate to what voltage back to the control board. I have wired a pot to the control line on the control board. I did not know if it should be positive or negative. For now it is set up to adjust from 0 to +10 volts.

    When I try to power up the laser the main capacitors charge up to about 300 volts, the control board powers up at + and – 12 volts so all this is as it should be I think. A white card held behind the HR mirror reflects a little red light from the filament. After about 40 seconds the igniter starts ticking and there are bright flashes of blue laser output. The beam is also easily visible coming out of the high reflector end. The light sensor output pulses up from 0 to about 2.5 volts with these flashes – I forget but I think the pulses lasted around 10 msec. When the laser flashes the main capacitors discharge to about 100 volts (on a slow voltmeter – so maybe it is really lower than that) then quickly charge back up to 300 volts after the flash.

    Here is what I am thinking.
    1) The bright flashes of laser output mean the tube is not the problem - unless the pressure is high in which case it might behave this way???
    2) The fact that the voltage on the main capacitors drops so much after the tube fires means that the problem is in the power supply.

    Is my interpretation right?

    Since I have not been able to isolate the problem in the power supply here is what I have in mind. Disconnect the control board. Exclude the linear pass transistors from limiting the tube current by saturating their base current. Drive the switching transistors at a fixed 20kHz, ~40% duty cycle. I have the schematics so here is how I plan to do this: The control board has only two outputs that go to the rest of the power supply. One goes from the pulse width modulator IC (AM6301) to an opto-isolator on the drive board which then drives the switching transistors. The other goes from the op amps on the control board that take as input the control signal and optical feedback signal. The output of those op amps controls the linear pass transistors. I plan to break the two output wires from the control board and apply: -29 mA to the line that goes to the bases of the linear pass transistors (MJ2500 Darlingtons) to saturate them and a square wave (0 to +12 volts, 20kHz, ~40% duty cycle via 680 ohms) to the line that goes to the opto-isolator that drives the switching transistors. I may have to post part of the schematic to make this more clear.

    Does this sound like a reasonable way to try to isolate the problem? With 40% duty cycle on the switching transistors and saturated linear pass transistors am I likely to vaporize silicon (again)?

    I think I am just a short way from getting this laser to work again. I would really appreciate any ideas that you folks might have.

    Roger

  4. #84
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,890

    Default

    Um, the HGM 5 magnets are a series parallel combination in series with the filter cap, they do NOT run across the dc line like a normal "wet" argon magnet. I dont understand if you have this configuration or not. If misswired they would pull things down to 100V or so. in other words magnet current IS the tube current..

    Something sounds wrong here.. Are you sure you have a demand signal to the control card? With no analog demand signal it will either shoot up to the max limit or set there and flash. But not both.

    Have you replaced the 14 or so volt zener that switches the supply from linear to digital mode?

    This is NOT a true PWM supply, HGMs switch in mid stream from linear to pwm by sensing collector voltage on the passbank, even on a model 5. Not respecting that Zener's decision making function WILL result in making silicon monoxide smoke with a touch of metal vapor thrown in.

    There ARE more then 2 wires to the passbank, there are considerable feedback signals in the system.


    Steve
    Qui habet Christos, habet Vitam!
    I should have rented the space under my name for advertising.
    When I still could have...

  5. #85
    Join Date
    Apr 2008
    Location
    Albuquerque, NM, USA
    Posts
    19

    Default Problems repairing HGM Power Supply

    Steve

    I am using the HGM power supply with an ALC-905 water cooled tube. They are not a matched set but they did work together two years ago.

    As you said, for an HGM-5 supply with a matched HGM tube, the magnet current IS the tube current. The magnet in my tube has a resistance of 230 ohms so it only draws 1.3 amps at 230 vac rectified. Therefore I could not use the HGM power supply to power the magnet in my tube. I made a separate power supply just for the magnet. It is 230 vac, rectified, filtered and with a small inrush limiting resistor. It seems to work fine.

    The wires that normally go from the power supply to the magnets are just connected together otherwise the power would never get past the first filter capacitor. If the tube magnet was supposed to function not just as a magnet but also as the "L" part of a C-L-C or “Pi” filter then maybe I have to put an inductor in its place??

    I know my power supply is an HGM. I am told it is an HGM-5 but there are no markings on it for me to be sure. Some time ago Adam (Buffo) sent me the schematics for his HGM-5. Mine is almost identical to his schematics.

    I will describe some the circuit in my power supply below. I have also included my interpretation of what some of the parts do so that if I am way off base perhaps you can set me straight.

    The schematics show the control board with a total of 14 lines coming onto and off of the board. I have divided them into INPUTS (going to the control board) and OUTPUTS (coming from the control board) and listed them below. I don’t suppose I can post the schematic online but I will PM the main page to you (Steve) so at least the next section will make sense to you. Here are all 14 lines and my interpretation of the function of each.

    Inputs to the control board:
    J2-19
    Power (positive) to the 7812 voltage regulator
    J2-20
    Power (negative) to the 7912 voltage regulator
    J2-21&22&Y&Z
    Power (ground) going to the control board
    J2-E
    “Control” – has to be an input because it goes straight to one of the inputs of an op amp and no where else.
    J2-F
    “Feedback” – also has to be an input for the same reason as the “control” line.
    J2-5
    I think this is an INPUT to the control board. It comes from the collectors of the series pass transistors and goes to the inputs of two op amps on the control board. The outputs of those op amps go to the input (pin 16) of the PWM regulator IC (AM6301).

    Outputs from the control board:
    J2-6
    Two op amps process the “control” and “feedback” input signals then by way of two 2N2905 transistors the signal leaves the control board by J2-6 and goes straight to the bases of the series pass transistors.
    J2-7
    Control board ground going to the 0.2 ohm resistors attached to the emitters of the series pass transistors. Provides a reference point for J2-6.
    J2-15
    Power on delay – comes from a 555 timer on the control board and is the ground end of the interlock string which controls the solid state relay that turns on the power to the main power board.
    J2-17&W
    +12 volts leaving the control board to the other end of the interlock string.
    J2-V
    -12 volts leaving the control board.
    J2-11
    Output of the PWM regulator IC on the control board to the optoisolator on the drive board that drives the switching transistors.
    J2-12
    Return to control board ground for the optoisolator on the drive board that drives the switching transistors.
    J2-16
    Output from the power on delay circuit that turns on the relay that provides power to the trigger card.

    Of all these lines J2-6 appears to control the series pass transistors and J2-11 appears to control the switching transistors. All the other output lines seem to serve more housekeeping functions.

    You mentioned a “demand signal to the control card” – I’m not sure what this is. The J2-E “Control” signal on the schematic perhaps? I have set up a pot to adjust this from 0 to +10 volts but it makes no difference in the behavior of the laser so far.

    You also asked “Have you replaced the 14 or so volt zener that switches the supply from linear to digital mode?” I think you are talking about VR1 on the control board schematic (a 15 volt zener). One of the first things I did was to de-solder one of its leads and test it – it was fine. I have not replaced it. Perhaps I should though since I am not sure that confirming the correct zener voltage is enough to be sure it is OK.

    I recognize two of the op amp circuits on the control board – an integrator and an inverting amplifier but the other two op amps and how they all work together are a puzzle to me.

    I thought the linear transistors operate continuously for fine control and the switching transistors switch at about 20 kHz (varying duty cycle) for course control and to maintain reasonable power dissipation in the linear transistors by preventing too high a voltage drop across them. If this power supply changes back and forth between switching and linear then I am seriously confused. Can you explain a little more about how this works? When in linear mode do the linear pass transistors dissipate all the excess power?

    I really appreciate any suggestions and help in understanding this circuit.

    Roger

  6. #86
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,890

    Default

    The PWM is regulated by that zener, which acts to keep the the voltage across the two linears at 15V or less. The PWM is in pure conduction until Vz is hit.

    The input comparators run opposite to how you might think they do, you need a light loop to close the loop and make things stable, otherwise the power supply ramps up to Imax or just sets there and flashes.
    That is why your potentiometer was inactive. The light comparator went full on and saturated the drive signal. It got me the first time I worked on them, too. The card is almost the same as those used in the ALC60X gold boxes, HGM and ALC were the medical and scientific divisions of the former Laser Corporation.

    Imax is not in the primary feedback loop, its more or less the level of passbank drive and pwm, so you need to add either a light sensor or say a Hall Effect current sensor (20-30$) to close the primary loop. Imax sets the max PWM duty cycle more then anything, so as line voltage changes, the tube current at Imax changes. If you turn Imax up too high when running the hack, things blow up.

    In other words the common hack without adding feedback is the linears in almost full conduction and PWM duty cycle * line voltage = tube current. The control inputs drop out of the picture, and things can oscillate.

    The model5 magnets serve as big inductors in a low pass filter to prevent some oscillation that can occur at really low or really high currents AND Smooth out the draw to the power line. They do need cooling air when operating, and are usually square copper wire or flat aluminum tape inside.

    For the most part, the plasma in those shorter tubes is has a negative I/V characteristic, depending on where you are in the pressure curve when the tube is filled. Those inductors provide both reactance and a little bit of series resistance to stabilize things, and keep the start inrush reasonable.

    Sounds like you have a 5L.

    I have a friend that is factory trained service on these. He still services them. He may be able to sell you a chip or card if the AM6XX is blown.

    Steve
    Last edited by mixedgas; 04-27-2010 at 05:58.
    Qui habet Christos, habet Vitam!
    I should have rented the space under my name for advertising.
    When I still could have...

  7. #87
    Join Date
    Apr 2008
    Location
    Albuquerque, NM, USA
    Posts
    19

    Default HGM-5 Power Supply troubleshooting

    I see how the Zener limits the voltage drop across the linear pass transistors to 15 volts. That part makes sense. What is the purpose of having linear pass transistors if there are switching transistors in the circuit?

    Any idea roughly what inductance the HGM-5 power supply expects to see for the magnets in the laser head?

    By “Imax” do you just mean the maximum current or some adjustment on the control board that controls the maximum current (R1 or R13 maybe)?

    I powered up the control board (separate from the rest of the power supply). The circuit produces a nice variable duty cycle square wave output so I think the AM6301 is fine.

    I actually have a light sensor installed on the HR end of the laser head and connected to the power supply. I could not find any specs on the IR50505 sensor shown on the schematic so I used an SGD160 PIN photodiode. The beam exiting the HR end of the ALC-905 tube goes through a glass plate at 45 degrees to the beam. The plate reflects part of the beam (about 20% at that angle??) through a ground glass diffuser in front of the photodiode.

    When I point a 1mW HeNe laser straight through the glass plate onto the diffuser/photodiode arrangement the circuit produces an output of +5.2 volts. When I try to power up the HGM-5 / ALC-905 laser the flashes of laser light produce a pulse of about +2.5 volts from my light card.

    Below is the circuit. Does this make sense? Is the sensitivity (light in to volts out) of my light card all wrong?

    Roger

    Click image for larger version. 

Name:	Light Sensor.JPG 
Views:	13 
Size:	36.2 KB 
ID:	16353

  8. #88
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    9,890

    Default

    perseus says, and I quote:


    I see how the Zener limits the voltage drop across the linear pass transistors to 15 volts. That part makes sense. What is the purpose of having linear pass transistors if there are switching transistors in the circuit?

    The switching portion keeps the voltage across the linear transistors to 15V or less, meaning you can use less expensive transistors and a small heat sink. Tube voltage + passbank voltage must equal the rectified line if there is no PWM . PWM effectively lowers the rectified line voltage, thus reducing the passbank stress and making the PSU smaller and cooler.

    Any idea roughly what inductance the HGM-5 power supply expects to see for the magnets in the laser head?

    No... But its huge. But not that huge. I can ask my friend to perhaps loan you a magnet for measurement? They are potted in hard epoxy and the last time I smashed one to recycle the copper, I did not count the turns, but there is a lot of them because you want 700-1000 Gauss around the Argon tube. (less if KR)

    By “Imax” do you just mean the maximum current or some adjustment on the control board that controls the maximum current (R1 or R13 maybe)?

    Yes!

    I powered up the control board (separate from the rest of the power supply). The circuit produces a nice variable duty cycle square wave output so I think the AM6301 is fine.

    I actually have a light sensor installed on the HR end of the laser head and connected to the power supply. I could not find any specs on the IR50505 sensor shown on the schematic so I used an SGD160 PIN photodiode. The beam exiting the HR end of the ALC-905 tube goes through a glass plate at 45 degrees to the beam. The plate reflects part of the beam (about 20% at that angle??) through a ground glass diffuser in front of the photodiode.

    A PHOTODIODE will not cut it, its designed for a silicon solar cell in photovoltaic mode, producing 2 to 4 volts at full laser after the signal is amplified.. A series of ceramic disks and pieces of blue lighting gel are used to attenuate the light and color correct it.
    The large area of the solar cell is used to integrate speckle noise in the laser beam when it its the solar cell, and to reduce the effective bandwidth of the sensor. A photodiode can be too fast.

    When I point a 1mW HeNe laser straight through the glass plate onto the diffuser/photodiode arrangement the circuit produces an output of +5.2 volts. When I try to power up the HGM-5 / ALC-905 laser the flashes of laser light produce a pulse of about +2.5 volts from my light card.

    Below is the circuit. Does this make sense? Is the sensitivity (light in to volts out) of my light card all wrong?

    Yes, you are perhaps missing the attenuator disk or have the wrong pickoff? The light pickoff is a piece of AR coated glass, it only reflects about .1% of the light from the argon in a medical system.

    That is not the normal HGM light circuit. That is the ALC low noise light circuit. It could be perhaps adapted, but the HGM circuit usualy produces a minimum of .6V even at no light and ranges up to 3-4 volts. It consists of a solar cell, a single 741 opamp and a 2n2222 transistor , a few resistors and a few filter caps to keep noise out of the light card..

    Roger

    Click image for larger version. 

Name:	Light Sensor.JPG 
Views:	13 
Size:	36.2 KB 
ID:	16353[/QUOTE]
    Last edited by mixedgas; 04-30-2010 at 07:09.
    Qui habet Christos, habet Vitam!
    I should have rented the space under my name for advertising.
    When I still could have...

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •