Announcement

Collapse
No announcement yet.

Orange OR200 the definitive maintenance thread

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #31
    Originally posted by boroman View Post
    I have re-capped the beast, changed all the screen resistors to 330 ohm / 2W (original: 270 ohm / 1W). I just had a chance to plug it in and it works, but I haven't hooked it to a bass cab yet. If you want me to make some close up shots, I'll do that. Seems that not many of these are around. Also, near the 16+16uf multicap there were two resistors across it 150k and 100k. Schermatic calls for 2x100k... I left the factory ones.

    By the way, just a free 2 cents... that's a serious desing failure with the OR200 - if you all suggest not to "drive the amp hard" and worry about the screen resistors/fuse pop/tube blow, how this amp was even constructed that way? I'm really disappointed, because I wanted to gig this amp pretty serious way.

    Re-cap:
    Limit the drive as nick suggest. Marshall get the most notorious failure rate in 200w Majors known as 'smoke on the water' effect. The same circuit as you OR. It well depend how hard is drived. You can use it for decades if don't push it over limit.
    "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

    Comment


    • #32
      So, again, a design flaw.

      I don't drive it past 11 o'clock on the volume, but it's not a problem even at this setting to hear a bit overdrive when hit hard.

      What formula is used to bias it, I mean, UL amps? It's other than Class AB?
      Those for trim pots at the top chassis will do the jobn if set to 1.4V? (I can't turn it more than 1.1V - end of the pot).
      I'm speaking more user-wise rather than technical side. Just to set and play. All I need to know.

      Comment


      • #33
        Hi. I saw you did not changed the caps into cathodes. If the measuring points are over cathode resistors think the caps cannot influence the measurement as the current is very low...but they can influence somehow the operation of tubes in case they are broked and not decouple well. Try to change one cap and do the measurement before and after to see if make a difference...just my thought...
        "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

        Comment


        • #34
          Originally posted by boroman View Post
          Also, near the 16+16uf multicap there were two resistors across it 150k and 100k. Schermatic calls for 2x100k... I left the factory ones.


          Re-cap:
          the resistors you.re talking should be balancing resistors if we refer to c10, c11 in to schematic. There are there to balance the voltage over top capacitor and bottom capacitor. They should be equal , and I think there are. Each section of 32+32uF is in series with 16uF caps in top. One have 150k+150k, the other 100+100k. The bottom resistors are visible, the top resistors are in red varnish in you pics.
          Last edited by catalin gramada; 06-25-2020, 07:39 PM.
          "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

          Comment


          • #35
            Originally posted by catalin gramada View Post
            Hi. I saw you did not changed the caps into cathodes. If the measuring points are over cathode resistors think the caps cannot influence the measurement as the current is very low...but they can influence somehow the operation of tubes in case they are broked and not decouple well. Try to change one cap and do the measurement before and after to see if make a difference...just my thought...
            Can you point me where they are/values? I dont see anything connected straight to the trim pots...

            the resistors you.re talking should be balancing resistors if we refer to c10, c11 in to schematic. There are there to balance the voltage over top capacitor and bottom capacitor. They should be equal , and I think there are. Each section of 32+32uF is in series with 16uF caps in top. One have 150k+150k, the other 100+100k. The bottom resistors are visible, the top resistors are in red varnish in you pics.
            This multicap is 16+16, not 32+32, and yes, it's in series with another 16uf's (small, changed ones, on my pic)
            The 2x 64+100uf (changed to 100+100 now) have 100k balancing resistors, the 3rd multicap 16+16 has 150k+100k - so should I go for 150+150 or 100+100 here?

            Comment


            • #36
              Originally posted by boroman View Post

              Can you point me where they are/values? I dont see anything connected straight to the trim pots...
              there is a small board on back, between impedance selector and slave output jack. There are 4x 30 ohm cathode resistors over there which are bypassed with 50uf electrolytic blue caps (c12-15 in you schematic). Normal this caps should not influence the dc measurements but is easy to try and is recommended to change it. Nothing physical related to trim pots. The trim pots are in fixed bias circuit. The capacitors are in cathodes circuit and the wiring from those board should be related to the check bias points instead. If you said the recommended value in check points is 1.4v, and the bias is measured over cathodes you have 1.4v / 30 ohm = 46.6 mA. Multiply with the plate voltage at iddle - said 650v - and get 30 w dissipation. For kt88 the max dissipation is 42 w, at 30w the tubes will be biased 72%.
              This is just an example. You have to measure you voltages in you amp to calculate yourself. If wanna draw 46mA to the tube is not equal if the plate is supplied at 650 or 600v, the dissipation will be different (not too much, 2,3w, but still). For this reason you should measure the voltage over plates at iddling.
              With 1.1v over cathode res at 650v the tube will draw 37mA arround determine 24w dissipation, meant biased 57%. As result I can suppose you get little bit more crossover distorsion as the circuit did not benefit by global nfb network. That is not necessary a bad thing in MI world, and the tubes will be more happy with that...
              Last edited by catalin gramada; 06-27-2020, 09:52 AM.
              "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

              Comment


              • #37
                This is how my output looks like up to clip, with a unmatched quad with tubes biased around 50%. 8db global nfb was applied.
                Clean 230W rms into 7.8 ohm resistive load.

                Click image for larger version  Name:	20200627_022800.jpg Views:	0 Size:	1.77 MB ID:	908306
                Last edited by catalin gramada; 06-27-2020, 10:01 AM.
                "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

                Comment


                • #38
                  catalin gramada That's awesome! Thatnks for the formula! That makes sense to me, and I found the little board there! Did you make those measurements with the 1K 5W screen resistors? So you haven't lost any power, don't you? Seems Okay for the rating fo the amp?

                  I have 604V on the plates and the same 604V on the screens. Bias set to 35mA per tube, corresponds well with the readings from the test points (about 1.1V).
                  Isn't it a bit too high to have over 600V screens?

                  Comment


                  • #39
                    Originally posted by boroman View Post
                    catalin gramada That's awesome! Thatnks for the formula! That makes sense to me, and I found the little board there! Did you make those measurements with the 1K 5W screen resistors? So you haven't lost any power, don't you? Seems Okay for the rating fo the amp?

                    I have 604V on the plates and the same 604V on the screens. Bias set to 35mA per tube, corresponds well with the readings from the test points (about 1.1V).
                    Isn't it a bit too high to have over 600V screens?
                    Is something was discussed. You have 600v at iddle (no signal applied). You power supply is not regulated, when you put signal on you amp the tubes start to draw more current. That put more strain on power transformer and in consequence the voltage will drop little bit, so when you power tubes works at nominal power the supply voltage will slide more than 560v arround which is within tube specs. You have understand the screens will not blow (arcing) at 600v, the danger is to not exceed the max power dissipation of the screens. I used with success over years dc rms calculation (this is something I can handle) and found no more than 12w is acceptable. To measure the max screen dissipation at nominal power and clipping you need a power load resistor, signal gen and scope. At iddle the screens will draw insignificant current, no relevant for our context.
                    The output power is related to the screen voltage. Even with 1k screen limiter will not be huge voltage dropped by resistor and anyhow the screen supply voltage is huge due to UL connection. To be safe with 600v supply the limiter screen resistor may be excessive big causing unwanted effects or limiting the tube to achieve maximum power .Otherwise for mild clipping 470-1k may be enough.
                    l.m occassional builder, the pros guidance may vary...
                    Last edited by catalin gramada; 06-27-2020, 04:58 PM.
                    "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

                    Comment


                    • #40
                      One kt88 blowed during my previous hard clipping tests with 470ohm grid resistors. At 5w power rating they didn't broked and the tube passed away. Small power resistors could be the right approach to save the tubes
                      Attached Files
                      Last edited by catalin gramada; 06-28-2020, 08:07 AM.
                      "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

                      Comment


                      • #41
                        Originally posted by catalin gramada View Post

                        Is something was discussed. You have 600v at iddle (no signal applied). You power supply is not regulated, when you put signal on you amp the tubes start to draw more current. That put more strain on power transformer and in consequence the voltage will drop little bit, so when you power tubes works at nominal power the supply voltage will slide more than 560v arround which is within tube specs. You have understand the screens will not blow (arcing) at 600v, the danger is to not exceed the max power dissipation of the screens. I used with success over years dc rms calculation (this is something I can handle) and found no more than 12w is acceptable. To measure the max screen dissipation at nominal power and clipping you need a power load resistor, signal gen and scope. At iddle the screens will draw insignificant current, no relevant for our context.
                        The output power is related to the screen voltage. Even with 1k screen limiter will not be huge voltage dropped by resistor and anyhow the screen supply voltage is huge due to UL connection. To be safe with 600v supply the limiter screen resistor may be excessive big causing unwanted effects or limiting the tube to achieve maximum power .Otherwise for mild clipping 470-1k may be enough.
                        l.m occassional builder, the pros guidance may vary...
                        Just about when you posted it i started to get into UL stuff... Thank you. Now I know more.
                        The whole discussion about "what will blow first" - ressitor or tube - is, you know, just odd. Still think the amp is not properly designed or we don't have good KT88s right now. Amp should be designed that way if you put everything on 10, it will do the job just right. And most are. And if we're talking anow about mild clipping, and we still worry about breaking a tube or blowing a resistor... really?

                        I'll put 1k/5W tomorrow. We'll see.

                        How about those mismatched resistors on top of the 16uf+16uf cap? Should I got for 2x100k or 2x150k there?

                        Comment


                        • #42
                          Originally posted by boroman View Post

                          Still think the amp is not properly designed or we don't have good KT88s right now. Amp should be designed that way if you put everything on 10, it will do the job just right. And most are. And if we're talking anow about mild clipping, and we still worry about breaking a tube or blowing a resistor... really?

                          I'll put 1k/5W tomorrow. We'll see.

                          How about those mismatched resistors on top of the 16uf+16uf cap? Should I got for 2x100k or 2x150k there?
                          This is another problem and think is comming mostly from vendors politics to put on sale any craps without to do any selection.
                          I live in Europe .Just as example I bought a supposed matched kt88 quad from TAD Germany. They claimed they do a very rigorous test on their tubes they sell but found it was just relative acceptable matched just for plate currents, the screens currents was rubbish when starts to draw.Think they didn't bother to measure. More, I didn't received any test bulletin to know in what condition they measured the tubes and with what results. I received just a brand sticker instead...



                          On the other side, there is still a chance. as there are serious vendors on the market which know how to do a proper job.
                          Also JJ tubes provided by Tube Town Germany this time.

                          Click image for larger version  Name:	20200629_151331.jpg Views:	0 Size:	1.11 MB ID:	908480

                          And something from the past, when control quality really meant some...

                          Attached Files
                          Last edited by catalin gramada; 06-29-2020, 12:25 PM.
                          "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

                          Comment


                          • #43
                            Originally posted by boroman View Post


                            How about those mismatched resistors on top of the 16uf+16uf cap? Should I got for 2x100k or 2x150k there?
                            As time you measure equal voltage over the each caps from series pair there is no reason to change it.
                            I don't think there are mismatched (really I don't know what values are under red varnish) but think one series pair are 100k bottom/ 100k top, the other series pair 150k bottom/ 150k top
                            "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

                            Comment


                            • #44
                              Originally posted by boroman View Post
                              Amp should be designed that way if you put everything on 10, it will do the job just right. And most are.
                              I don't think designers thought that way back then. Power amp clipping was not 'normal', even more so for a 200W amp. But I agree that the tubes made back then would have been much more likely able to handle the abuse.

                              Originally posted by Enzo
                              I have a sign in my shop that says, "Never think up reasons not to check something."


                              Comment


                              • #45
                                ..and judging by the diminutive size of transformers nowadays it's clear they are still not designed to stand that kind of use, even less so.

                                The solution for this amp is either reduce the drive or lower the screen voltage. The first is easiest IMHO. Catalin mentioned a PPIMV way back which, if you use a dual control, is a good way to do it and you can tweak the safe limit and then just leave it set.
                                Experience is something you get, just after you really needed it.

                                Comment

                                Working...
                                X