Ad Widget

Collapse

Announcement

Collapse
No announcement yet.

How to protect the OT?

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

  • #46
    thanks steve. i (along with daz i think) thought that the suggestion was to put the fuse right at the CT so that a fastblow fuse could be used without worrying about the filter caps for the screens and preamp blowing the fuse when charging up.

    i take it that the point was to put a upstream of the CT tap so that it protects everything running off the secondary aside from the rectifier.

    thats a good idea steve. only issue is i am running off a single voltage supply and am either going to use a choke input for the screens/preamp or use zeners and a resistor to drop a couple hundred volts to ~400v.
    while we are on it, i remember you saying that modern chokes arent designed for choke input's. what are the probable issues of using one? i understand that it will need a large inductance so that it will still be able to filter the current at idle (having the screens rise to the plate's voltage when idling isnt ideal when the plates are running at 620v).

    one thing i am thinking would be bad about the choke input is if i decide to run fewer output tubes then there may not be enough load for the choke to drop the voltage correctly. would say 450v worth of zeners from the choke's output to ground work so that when the voltage floats up there is extra current draw clamping it down? could probably end with the zeners shorting out and blowing a fuse (imagine that, trying to play the amp hard to keep it working).

    thanks

    Comment


    • #47
      OTs almost never fail due to the power valves pulling too much current. You basically need a direct short from anode to cathode to do any such damage, or some serious positive grid voltage. A simple shared fuse in the cathode circuit of the power valves is quite sufficient to protect against this.

      The speaker jack should always be of the shorting type, to protect against no speaker cable being pligged in. Wth smallish amps then a dummy load resistor can be connected to the 'switch' on the jack socket.

      A largish resistor (100 to 200ohms, 5W) should be permenantly connected across the secondaries, to protect against a cable being plugged in, but no speaker attached, or a blown speaker.

      Flyback diodes connected from anode-to-ground are mainly meant to protect the OT by clamping negative flyback voltages. They prevent either end of the OT primary from going negative. They're not really meant to 'fail' with positive flyback voltages (although if they do then all the better), that's why 2 or 3 are connected in series, to withstand flyack conditions. If they were meant to fail, then only one 1000V, 1N4007 would be necessary.

      The most bullet-proof protection you can use is a gas-discharge surge arrester connected directly across the primary (anode to anode). You'll want one with a breakdown voltage about 2 to 3 times the supply voltage in a class AB amp, or about 3 times the supply voltage in a class A (single ended) amp.
      This sort of thing: LITTELFUSE|CG2-1000L|GAS DISCHARGE SURGE SUPPRESSOR | Farnell United Kingdom
      You can even get industrial bolt-on versions, which look pretty cool!
      Needless to say, the normal precautions can/should be taken in addition to one of these.

      Comment


      • #48
        Hey Merlinb, thanks for the gas discharge tube idea, I'm quite a fan of your Valve Wizard site btw

        Putting the HT fuse right at the CT might be theoretically better, except it's not theoretically better, because it endangers the screens of the power tubes. So that's why it's not placed there. Look at some schematics of Marshalls or whatever to see where they put it. I seem to remember on my old T'n'B 50 (with bridge rectifier) it was placed in series with the HT winding, to protect the transformer if a diode goes short. They don't do that on tube rectifier amps because it would require two fuses.

        As far as I know, current production chokes aren't designed for choke input filters. Possible issues you might encounter are:

        They don't "swing" enough, so the critical current is too high.
        They get hot from iron losses and emit a lot of acoustic and/or magnetic buzz.

        Or then again they might work fine. This is the kind of question that someone at Hammond should actually be able to answer. In any case I imagine there would still be hum problems, because chokes have an airgap that will leak flux, and the choke input filter generates lots of AC flux in there. It could potentially leak far worse than a PT.

        I've used the zener trick to keep things above the critical current, but only in a small high-frequency DC-DC converter, and even then it was a bit of a bodge. It might be worth scrapping the choke idea and using a pass tube running off the main 600V rail to drop the voltage. Or even a second small PT to power the screens and preamp.
        Last edited by Steve Conner; 06-13-2009, 10:09 AM.
        "Enzo, I see that you replied parasitic oscillations. Is that a hypothesis? Or is that your amazing metal band I should check out?"

        Comment


        • #49
          the gas discharge idea looks good. thanks merlin. another fan of the valve wizard here too.

          not a bad idea with the second pt steve. i may use zeners to drop the voltage, but i'm not sure. i already have 2 of the PT's (antek 4t450 400va 450v secondaries). i got 2 so i could go with 4 pairs of 6550/kt88's. i may go with 3 pairs and use one of those pt's and get a 250v or so transformer for the screens. using 2 of those pt's is probably a bit overkill, but the power supply is somewhere that most bass amps need improvement on. i dont know what to do with the other PT either. i dont think i will need any more amp than the one i'm making.

          maybe i'll get 2 of the 250v transformers and make another amp/power amp to sell. i have some free time september-feb...

          Comment


          • #50
            Originally posted by Steve Conner View Post
            I seem to remember on my old T'n'B 50 (with bridge rectifier) it was placed in series with the HT winding, to protect the transformer if a diode goes short. They don't do that on tube rectifier amps because it would require two fuses.
            Or you could put it between HT center tap and ground. Older Selmers (with two way rectifiers) have it this way.

            Cheers,
            Albert

            Comment


            • #51
              Originally posted by Merlinb View Post
              The speaker jack should always be of the shorting type, to protect against no speaker cable being pligged in. Wth smallish amps then a dummy load resistor can be connected to the 'switch' on the jack socket.

              A largish resistor (100 to 200ohms, 5W) should be permenantly connected across the secondaries, to protect against a cable being plugged in, but no speaker attached, or a blown speaker.
              Hi Merlin

              (Can I ask) why a 100R - 200R resistor, and not a resistor rated at the same sort of DC resistance as a speaker would be (i.e. 4R-16R)?

              (PS I am also a frequent user and fan of your site, and I promise to buy a copy of your book when it gets to NZ if you answer my question. Thanks )
              Building a better world (one tube amp at a time)

              "I have never had to invoke a formula to fight oscillation in a guitar amp."- Enzo

              Comment


              • #52
                the reason is because the resistor is permanently in the circuit. it is parrallel with the speaker. it draws some of the power from the speaker at all times, and saves the ot by making the load less than infinite when nothing is plugged in.

                Comment


                • #53
                  Because a 100-200R in parallel with the speaker will hardly change the load at all. If i add 8 ohms and 100 ohms together in parallel in my app called electronic assistant, the result is 7.4 ohms. So that value doesn't change the load to any notable degree yet puts a load there in case the speaker goes open or isn't plugged in.

                  Comment


                  • #54
                    Duh! Thanks guys (I should've figured that one out for myself). I got disoriented with talk about a switching jack, thinking that somehow the unplugged jack must be switch to the resistor-only when the speaker isn't plugged in etc...
                    Last edited by tubeswell; 06-16-2009, 08:19 AM.
                    Building a better world (one tube amp at a time)

                    "I have never had to invoke a formula to fight oscillation in a guitar amp."- Enzo

                    Comment

                    Working...
                    X