ad: KF7PMW-1

Daiwa RS-300A power supply issue

Discussion in 'Radio Circuits, Repair & Performance' started by N2BSS, Jan 28, 2016.

ad: L-HROutlet
ad: l-rl
ad: L-MFJ
ad: Subscribe
ad: Left-3
ad: Left-2
ad: MessiPaoloni-1
  1. N2BSS

    N2BSS Ham Member QRZ Page

    I have an old Daiwa RS-200 12v 30a adjustable power supply that has been on the fritz for a few years. I decided it was time to repair the thing. This uses an op amp voltage regulator. It powers up but only outputs some 0.5v. The voltage regulator is not functioning.

    Voltage measurements at key points are correct (transformer and rectifier output). The vr board has 5v a/c.
    upload_2016-1-27_20-48-50.png

    I already replaced the op amp but the problem remains. I took some additional measurements and found a problem at op amp pins 12/13/14. Pin 12 has the expected 7v but pin 13 measures 0. Pin 14, instead of having a specified 15v, has 27v. (Measurements at other op amp pins are appx normal, so I'm concentrating here).

    The 2sc1815 transistor is another story. Instead of 10v at the base I am measuring 1v. I have 28v at the collector, and also 28v at the emitter rather than the specified 15v. This kind of perplexes me as none of the reading make sense... where did all the base voltage go (there is 28v at pin 14) and why is the emitter voltage so high? Is it possible the transistor is shorted?

    I'll be grateful for any and all debugging advice and assistance!
     
  2. N2BSS

    N2BSS Ham Member QRZ Page

    Come to think of it, I may be picturing Q1 upside down, in which case the emitter has only 1v while the base and collector each read 28v. I'm thinking there is a base-collector short?
     
  3. KB9DT

    KB9DT Platinum Subscriber Platinum Subscriber QRZ Page

    If Q1 has 28 volts on the base and 1 volt on the emitter, the base emitter junction is open. You can check this with an ohmmeter (reverse leads and try both ways). 28 volts on the base is trying to raise the output voltage, and it should be closer to 2 volts (3 base emitter drops) higher than the output voltage when the supply is working.
     
  4. KW4EK

    KW4EK Ham Member QRZ Page

    As it is an Op-amp circuit, if the positive and negative inputs (12 and 13) are wrong then it is a given that the Op-amp's output (14) is necessarily going to be wrong as the inputs define how the Op-amp's output will behave as it tries to bring the inputs into line (it wants to generate whatever output, at pin 14, is required to make the inputs match). As those inputs are defined by the resistors for the voltage selector and voltage adjust potentiometer then you first need to figure out why those inputs do not match the specified voltages. Be aware that the voltage output by the bank of series-pass transistor's will directly influence the voltage beng fed into pin 13 of the Op-amp, so it is important to know their output voltage, too.

    If the Op-amp's output (pin 14) measures 0 volts when measured directly at the Op-Amp, considering it is a low impedance output (and a high impedance input) driving into a 3.3K resistor then your immediate problem should either be the Op-amp, the resistors that form the voltage dividers to the input, or an issue with the rails those resistors draw their power from. Since you have known voltages to troubleshoot with, first rule out the Op-amp, then work backwards from there before working the problem further forward. However, since pin 13 of the Op-amp is influenced by the output of the series-pass transistors, you may also need to work forward some as well since their output will influence the voltages on pin 13 that inputs into the Op-amp.

    The 28V at the transistor's output is not terribly surprising since there is 1V at the base, which is greater than a diode drop, in the absence of a significant load on its output. I assume that the 1V you are measuring is making it to the driver transistor's base via the diode (with the customary diode voltage drop) by way of the pin 7 Op-amp's output driving its output in the absence of proper inputs. That does not mean the transistors are necessarily good, but you are not there yet as you must first get the Op-amp to output a proper output to drive the driver transistor. BTW, transistor's sum current at their output, not voltage, so that may help with some of your confusion. The Op-amp should be driving the driver transistor in an effort to bring the two inputs back into line so that their input voltages match...since it is outputting 0V it is clearly unable to influence the series-pass transistor's because it is not even trying to drive the driver transistor that drives those series-pass transistors.

    There is so much happening in the feedback loops that are the Op-amps that I am having a very hard time trying to break this problem down into bite size pieces so that you can more easily follow, which is how I would normally present such a possible solution. You may find it easier to first verify that all the resistors, capacitors, and diodes are in good order, then move on to verify that the transistors (ideally using a transistor tester, but a mockup simple transistor circuit may also be used coupled with diode measurements to verify proper operation) and Op-amp's are all in good working order, replacing anything that is suspect (replace any suspect Op-amps after ruling out all else, ideally). Do not forget to verify the switches and potentiometers are in good working order with no bad contacts and check that there are no broken wires, links, or bad/broken solder joints on the board. Most of the time the fault will be a singular misbehaving component, but sometimes that be failed component can take out several others. Oftentimes it is something frustratingly simple like a bad switch contact, dirty potentiometer, broken wires or bad solder joints, etc., so do not rule out the simple things too soon). Always verify the internal voltage rails and check for any excess ripple (which may be a sign of failing bulk filter capacitors or defective rectifier diodes), with and without an applied load on the output, as bad internal voltage rails tend to mean a malfunctioning device.

    While testing, never use anything you value as a test load until you are 99.9% certain the supply is back in good working order and have thoroughly verify the output voltages. During testing, use power resistors (of suitable wattage for the power they will dissipate) or light bulbs and other cheap, but forgiving, loads so that nothing of value can be damaged if the voltage suddenly swings to the full unregulated rail voltage (which would appear to be as much as 28-30V for your supply).

    Most of all, I wish you all the very best. Please keep us apprised of your progress.
     
  5. AI3V

    AI3V Ham Member QRZ Page

    Op, do you know how to test transistors with a vom/dmm ?

    I would measure the transistors. Since you have power at the big filter cap it probably is a regulator problem.

    Pass transistors (really any bipolar transistor) can be checked by setting your meter (i'll assume a dmm) to it's diode test position.

    You should see a reading of approximately .7 when either the base/emitter or base/collector junction is forward biased, and a blank reading (the same reading you get with the dmm probes not connected to any thing) when you reverse the leads and test both junctions again. You should also see a blank reading between emitter and collector.


    Typical failures

    EB, EC ,or BC short= low reading measured with probes in either position

    EB or BC open high reading either way with probes.

    Practice readings with ANY silicone diode, and simply test each xistor as 2 back to Back diodes.

    Disconnect transistor from circuit for final confirmation.

    Also check underside of PCB for broken connections.

    I have fixed hundreds of PS, I find the dmm "shotgun" approach to work 99% of the time faster then I can dig up a schematic.

    Good luck and let us know what you find.

    Rege
     
  6. N2BSS

    N2BSS Ham Member QRZ Page

    Greetings all. Thanks for the feedback. I hadn't had much time to work on this and I just got back to it. Q1 appeared to have a BC short. I was getting low readings with the DMM across the BC junction no matter what direction I probed. I didn't have a spare 2SC1858 so on the advice of a local ham I replaced it with a 2n3904. Outside of the pin out differences the specs between the two looked similar. I'm guessing this may not have been the best substitution as now I only get 9v out of the supply instead of 13.8v. As soon as I get a chance I plan to take a few more voltage readings at Q1 and the op amps. I'm hoping the shorted transistor didn't damage the op amp chip.
     
  7. N2BSS

    N2BSS Ham Member QRZ Page

    Here are the voltage readings:

    Emitter: 1.6v
    Base: 2.2v
    Collector: 32v


    OpAmp 8: 10v
    OpAmp 9: 3v
    OpAmp 10: 3v

    OpAmp 12: 7v
    OpAmp 13: 4.5v
    OpAmp 14: 30v

    The voltage at the base of Q1 seems awfully low considering that pin 14 has 30v. With r17 at 3.3k I either have a huge voltage drop across it or something else has gone wrong.

    Values for the Op Amp at 8/9/10 all look normal.
     
  8. KA9JLM

    KA9JLM Ham Member QRZ Page

    If r17 is good, My guess would be pin 7 of the IC is pulling the voltage low.

    Check pins 5 and 6 to see if that is the case.

    Good Luck.
     
  9. KB0MNM

    KB0MNM Ham Member QRZ Page

    You guessed that the 2N3904 may not have been the best substitution, but did you:
    1. Check the cross-reference guides yourself to see if the gain ( usually abbreviated hfe ) matched?
    2. Use a heat-sink ( eg. hemostat or alligator clip ) to ensure the new transistor was not overheated during soldering?
    3. Give your friend the right part number from the old part??? I see an 1815, not 2SC1858 on my screen.
    It seems as if you are making progress. I would not replace an op amp ( no matter how inexpensive ) without checking the surrounding circuit first. If the supply is old, the electrolytic capacitor(s) may be partially shorted. There are a lot of good texts on how the ESR ( effective series resistance ) of a not-fully-shorted capacitor can make a full repair difficult. The subject was covered in a homebuilt tester ( commercial ones are available ) in QST magazine about 2 years ago- April, I believe.
     
  10. KA9JLM

    KA9JLM Ham Member QRZ Page

    You are correct, it is a 2sc1815. The other number was for a diode I think.

    It does look like a 2N3904 will work, But as mentioned the pin out is different.
     
    KB0MNM likes this.

Share This Page