Connect with us

Unknown capacitor.

Discussion in 'Datasheets, Manuals and Component Identification' started by Tha fios agaibh, Nov 9, 2018.

  1. Tha fios agaibh

    Tha fios agaibh

    1,892
    599
    Aug 11, 2014
    Mr. 73's de Edd,( I would just call you Mr.Edd but it reminds me of a horse)
    Firstly, Thank you so much for taking the time to break this down. I always enjoy your witty and well illustrated responses.

    Forgive me if my reponces are out of order to points raised but I will get there.

    As far as operation goes: Yes, The molex connector goes to a microswitch pushed to initiate the punch cycle. The triac turns on a motor that drives a cam to make punch go in and out.
    I have on order several zener diodes to replace shorted one found. Power supply was dead. I wasn't sure I was going to successfully identify the bad zd so I just ordered a variety of voltages. The original zener looks to be 15v.

    This failed zd is likely the heart of the problem but I'm left with a few questions.

    The Ir diode section of the photo interrupter has a 3.3k limiting resistor which only leaves about 4ma for it given a typical Vf of 1.2v according to the data sheet H301-05.

    How much current is necessary to produce enough light to switch on the photo transistor? I'm sure it depends on how the transistor is loaded and supplied, but is there a range should I be looking at?
    I see the graph shows If at 20ma at 1.2v.

    Had I not found the 15v marking on the zener I would think the supply to be much larger than that.

    I know maximums are important but if insufficient Vf or If is supplied, would this cause the Ir led to fail?
     
  2. Tha fios agaibh

    Tha fios agaibh

    1,892
    599
    Aug 11, 2014
    2018-11-16 01.37.02.jpg
    A crude sketch drawn without knowing the itty bitty transistor pinouts, in case anyone's interested.

    Waiting for postman to bring me a zener.
     
  3. 73's de Edd

    73's de Edd

    2,650
    1,076
    Aug 21, 2015
    Sir Tha fios agaibh . . . . .

    Looking at the circuitry of that system, I feel that your friends 'ole Mean Machine will PUR R R R R R R like a kitten again, just as soon as you replace that formerly shorted Zener diode. Thinking of a 1N4744 . . . . . 15V
    unit or possibly, one of even heftier wattage.
    Or take a TO220 cased NPN transistor and a 100 ohm base resistor and make a current amplified POWER Zener.
    So . . . .it does look like the 100 udf main cap ) . . . and that Zener do seem to be sharing side by side pads on the foil, so that would be the sole DC voltage used for the systems supply.

    MACHINE OPERATION . . . . .

    With your reverse sketch, I can now see that the WHITE MOLEX switch starts a paper punch operation and simultaneously starts a timeout function with the two discrete transistors.
    Then, if all is well, the punch action is completed and the punch cycle end was confirmed with the Optical Interrupters beam disruption.
    However, IF the paper sizing/weight or number of sheets stalled the mechanism, the time out function stops the motor unit and the compression spring releases pressure on the indented / unpunched paper..

    To be able to replot your initial sketch as a proper schematic . . . . . confirm for me . . . . . units BRAND and Model number, also wattage rating, if on its label.

    • 1 . . . I think that your connections of the C4 22 ufd @100VDC rated (for having low leakage) timing cap at the bottom of ur sketch are in error, I see its + going to a transistor collector.
    • 2 . . . I see that you caught cap bleeder resistor R2 1 meg, but that you omitted the R1 1 Meg resistor that is located between neutral AC line to the SWITCHED hot AC.
    • 3 . . . I can make out the markings of the wire terminal stakes of MOTor . . . . .C . . . . . LINE . . . .SW2 ( it being the units main power switch) and up above, the top MOTor . . . . N ??? . . . . . PL?? also, at board bottom there is an 8W3 , probably being SW3, what function is it ? and what does it go to, and where is that switches other end connection being ?

    • 4 . . . On the two PL's, I suspect them to be going to an AC line operated NEON pilot light, power presence indicator / power on, with its own internal current limiting resistor.

    You seemed uncertain on your transistors element assignments . . . .if their I.D. markings are visible, see if they all might be the same or a mix. Then do a diode test from a base to each collector-emitter for the beeps identifying good junctions and also for that telling if they are being either an NPN or PNP type of transistors.
    Here is basing, with the Collector tab being alone at the top
    Left lower tab is Base . . . . . . . . . .Right lower tab is Emitter.

    upload_2018-11-16_5-16-1.png
    • 5 . . . What symbolization I.D. is being assigned to the 2 pin WHITE molex plug ?

    FIO . . . . .
    On the XXX rated (Wooo-Wooo !) AC line dropping capacitor . . . . it is being marked as M68 . . . . so comparatively . . . if you were to see a resistor marked as R68, you would relate to it as being .68 ohms.
    Likewise . . . . . that cap would relate as being .68 microfarad.

    Also, you can see the " trimming " resistor used, that I referred to, as being R3.
    Since available higher value capacitor values, are being all spread apart . . .in this case, with them being either .47 or .68 or 1 ufd.
    AND, say that your design parameter of max power ever required, actually computed out as needing .64371and 11/64th ufd , you then use the the next highest available capacitance . . . . . .68 ufd . . . . .and then the easily available, proper series trimming resistor corrects excess voltage downwards for you.
    HOWEVER . . . .I am still ULTRA- BARELY able to see the top edge of R3, as being a definite Green (5) , but isn't the next band being a BLUE (6) vice a VIOLET (7) and then BLACK, thus making it an EIA standard value of 56 ohms ?

    Final thought . . . .
    Methinks that units main power switch is being a rocker switch type and probably almost totally hidden away behind the unit near the AC power cord input.
    Considering then, that if this unit is being used in an office / work / school application
    that might see this unit having that power switch being left always on . This unit will then be seeing itself taxed into 61 min hours - 25 hr days - 8 day weeks and 366 day years of use. And since that emitter LED within the OPTICAL interrupter will then be lit all of that time, it might benefit from that potentially life extending reduced current consumption provided by its 3.3k supply resistor.


    Thasssssssit . . . . .


    73's de Edd
    .....................


    Viagara . . . . . . ( The secret ingredients that make up its formulation )
    3% Vitamin E
    2% Aspirin
    2% Ibuprofen
    1% Vitamin C
    5% Spray Starch
    87% Fix-A-Flat

     
    Last edited: Nov 16, 2018
    Richard9025 and Tha fios agaibh like this.
  4. kellys_eye

    kellys_eye

    4,284
    1,145
    Jun 25, 2010
    I don't know why you're worried over the 3k3 resistor issue. Presumably the board worked prior to the power supply failure? Presumably the 3k3 resistor is still 3k3Ω (i.e. hasn't changed)?

    Given the manufacturers would have designed the board to actually 'work' with the 3k3 resistor - and it obviously did (does) then why the concern? Your problem is the PSU section!

    Of course, curiosity is said to have killed the cat (and satisfaction brought it back) so out of pure inquisitiveness I can see where you're going but as far as fixing the board is concerned.......?

    Personally I see no problem with the value.
     
  5. hevans1944

    hevans1944 Hop - AC8NS

    4,203
    1,990
    Jun 21, 2012
    As @Tha fios agaibh mentioned in his post #62...
    Methinks the problem was solved much earlier in this thread with the discovery that the zener diode had shorted, the remaining questions being (1) why? (2) what voltage was it? (3) was the wattage rating too small?

    Once the postman cometh, and a new zener diode is selected and soldered in place, the paper punch thingy will probably work as well as it ever did. Questions (1), (2), and (3) then become moot and hopefully a new thread can be started, since this one is getting a little long in the tooth.

    Of course, if there is a lot of time available, the entire paper punch mechanism can be reverse-engineered and "improved" for the sake of... well, because we can. I can see where some sort of Microchip PIC processor (or two or three) would lend itself to timing functions, jam detection, bell ringing and whistle blowing, perhaps to the amusement of those needing their paper or tickets punched. Maybe a torque sensor could be added to the motor drive gear-train to detect and abort operation when attempts are made to punch too many layers of paper, or when the punches need to be sharpened or replaced or lubricated. There might even be the possibility of replacing the existing power supply with a circuit less prone to catastrophic failure...
     
    Tha fios agaibh likes this.
  6. kellys_eye

    kellys_eye

    4,284
    1,145
    Jun 25, 2010
    Any time you find some half-way decent equipment that looks capable of lasting many years in operation I agree, it is a wise move to investigate any areas where failure may let the device down and see what you could do to keep it working.

    The zener failure is quite typical of where problems occur as I don't see zeners as being anything other than 'old' technology and a 'cheap and nasty' way of regulating a voltage. They are my first port of call in a faulty circuit (i.e. if there's a zener in it, the zener will be the fault).

    Whilst the KISS principle is to be lauded it can be taken to extremes and I'd much prefer a linear regulator to a zener in any situation where it can physically fit - sod the expense! But that 'me' speaking as an engineer and not a bean counter - the bane of engineers......
     
  7. Tha fios agaibh

    Tha fios agaibh

    1,892
    599
    Aug 11, 2014
    I'm trying to understand the datasheet, not just fix it.

    The main question I'm asking is;
    "what is required to sufficiently turn on the led to the point where the photo transistor sees it?"

    Even current in the microamp range will likely produce some photons of light but probably not enough effect the photo transistor.
    How does one interpreted the Vf vs If graphic chart? It starts at 0ma and shows a curve but I'm not clear on how to interpret where the "on" range is.
     
  8. Tha fios agaibh

    Tha fios agaibh

    1,892
    599
    Aug 11, 2014
    @73's de Edd: Excellent observations all spot on except the 22uf + does go to transistor side with only one leg (what your illustration shows as collector) I think its the base.
    I'm not worried about plotting out the exact circuit at this point but its fun to know how stuff works. (....Including Viagra. lol)
     
  9. Tha fios agaibh

    Tha fios agaibh

    1,892
    599
    Aug 11, 2014
    @kellys_eye: I agree that 3.3k is not a problem and your advice that 4ma is sufficient. I am just trying to understand the datasheet minimum requirements for the PI.
    Particularly the ir led input.

    Data sheet states maximum values, but how do you know minimum values?
     
  10. Harald Kapp

    Harald Kapp Moderator Moderator

    9,137
    1,846
    Nov 17, 2011
    The output of a transistor-based photocoupler is 'on' if the current through the collector creates sufficient voltage drop across the load resistor for the following circuit to detect it as logic low or high, depending on how it is connected.
    The VF vs. IF chart is no help here as it only describes the characteristic of the LED.
    What you need is the current transfer ratio (ctr) from the datasheet. It gives you Ic=CTR*Iled (under the test conditions given in the datasheet). Typically photocouplers are characterized above a certain minimum LED current only. The above equation is only valid in the range characterized. For LED currents below that range the behavior of the photocoupler is not (at least not well) defined. Typical minimum LED currents are 5 mA or 10 mA, also 1 mA and rather seldom less than 1 mA can be found.

    Look for example at the datasheet of a CNY17. Figure 5 suggest you can go down to as little as 0.1 mA LED current, but figure 6 shows characteristics only down to 5 mA and figure 7 down to 1 mA. In figure 5 you can see that it takes considerably more Vce for the transistor to become conducting and the transistor current is accordingly small.
    The switching charactzeristics on page 3 are for If = 5 mA, 10 mA and 20 mA. I wouldn't operate this photocoupler under 5 mA without intensive inspection of the circuit's behavior and simulating worst case components (e.g. by reducing to even lower values).
     
    Last edited: Nov 16, 2018
    Tha fios agaibh likes this.
  11. Tha fios agaibh

    Tha fios agaibh

    1,892
    599
    Aug 11, 2014
  12. kellys_eye

    kellys_eye

    4,284
    1,145
    Jun 25, 2010
    A 'sort of' CTR value is shown in fig 3 on page 4 of the data sheet.

    It shows collector-emitter voltage and collector current (in the transistor) for values of If in the LED.

    At 4mA If you should see 0.4mA (give or take - I can't read the graph that accurately) in the collector of the transistor.

    e.g. on a transistor 5V supply using a 10k resistor the collector voltage will be 1V (4V dropped across the 10k resistor).

    Basically, the lower the If the higher the collector resistor value to see any appreciable voltage drop and the collector resistor will depend on the load drawn from it.
     
    Tha fios agaibh likes this.
  13. 73's de Edd

    73's de Edd

    2,650
    1,076
    Aug 21, 2015
    Sir Tha fios agaibh . . . . .

    WHUT YOU TALKIN' 'BOUT . . .WILLIS . . . . .


    With ohmmeter in hand and having it placed on sub zero ohms range scale ( unless autoranging), check out my surmised foil paths . . . . .


    [​IMG]

    ADDITIONALLY . . . . .
    If that was my specific photo interrupter to evaluate . . . I would have a DC supply and load resistor on the receiver side and use scope monitoring of that load resistor in DC mode.
    Then slowly bring the power feed to the emitter sides LED up . . . to see the point of very initial detection of any change.
    Then go on upwards to evaluate, up to saturation point. Then set operating design parameters accordingly.

    73's de Edd
    . . . . . . . . . . .

    Two hunters are out in the woods when one of them collapses. PLUS . . .he doesn't seem to be breathing and his eyes are all glazed over.
    The other guy whips out his phone and calls the emergency services. He gasps,
    "I think my friend is dead! What can I do?".
    The operator says " Now calm down. . . . . I can help you . But, first, let's make sure that he's dead."
    There is a silence . . . . . . then a shot is being heard.
    Back on the phone . . . . . the guy then says . . . . . . . . . . "OK . . . . . . . now what?"
     
    Richard9025 and Tha fios agaibh like this.
  14. Tha fios agaibh

    Tha fios agaibh

    1,892
    599
    Aug 11, 2014
    Thank you @kellys_eye, that does makes sense.
    Datasheets usually make my eyes glaze over like trying to read the fine print of a financial prospectus.
    Now that you've pointed my nose in the right direction I have a much better understanding of what (at least that chart) it means.
     
  15. Tha fios agaibh

    Tha fios agaibh

    1,892
    599
    Aug 11, 2014
    Yes Sir, That's the path.

    Ok, I popped in a new zener a few minutes after getting my hands on it and the Punch machine is working fine.

    Yes, I thought about bench testing to find out the current required but I was curious how it's calculated form using the datasheet. Alas, I have a lot to learn there.

    In the event that a component can't be identified like that shorted zener, I would think it more practical to inject a separate DC supply like that in order to find the correct value, rather than try to figure the loading/currents and doing complicated math.
    Had we just took a educated guess at where the voltage should be I imagine that it could have gone badly.
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-