Announcement

Collapse
No announcement yet.

Help needed with new boost pedal build

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

  • Help needed with new boost pedal build

    Alright, I've beat my head against the wall (figuratively speaking, so far) with this new pedal but so far I've not been able to figure it out. I'm hoping maybe someone here can help me get over this hurdle.

    First of all, this is my first pedal, my first time working with ss amplification. I'm an amp guy, built, repaired and tuned lot's of amps, tube amps. I've got a pretty good understanding of tube amp tech and have quite a few amps under my belt. So when the need for a clean boost pedal came up I figured why buy one, I'll build one for the fun of it and the experience. Well, the fun part was in learning bi-polar transistor theory and application, designing it and assembling it, but now it's not so fun.

    The design is pretty much based on the old LPB1, which is a pedal I always really liked; but I added a diode clipping circuit to be able to add some dirt if/when needed. The LPB1 schematic is here: http://www.montagar.com/~patj/ehbsters.gif. and my pedal uses pretty much the same boost circuit, using a 2n5088 transistor. For the clipper circuit I just tapped into the boosted circuit after the collector, after the coupling cap prior to the volume pot, and ran two diodes in parallel with opposite polarity, then ran those to a pot wired as a variable resistor then thru a switch to ground. I also added a circuit to allow me to switch the value of the coupling caps, to go from full frequency boost to treble boost.

    The problem I'm having is their is no amplification from the transistor. I've ensured I have the collector, base & emitter terminals wired correctly, as per the data sheet and verified with a DMM set to diode check mode. I've made sure the pins on the transistor match the positions indicated on the data sheet, and have them connected as per the schematic. Testing the pedal with a 1k 100mv sine wave I have good signal transfer all the way thru the circuit, from input to the transistor base, but the signal strength drops appreciable across the transistor and is very weak coming out at the collector. I can inject a signal at the collector and it goes right on thru the rest of the pedal just as it should, indicating everything has continuity and the volume pot works as it should. The problem, as near as I've been able to narrow it down, is right at the transistor itself.

    DC voltage checks show there is very little current flow across the transistor, I've got 8.75v at the battery, 8.71v at the collector (just .04v dropped across the collector resistor), .92v at the base and .49v at the emitter. My ground connections appear to be good, I'm getting the correct resistance reading to ground in the all the right places.

    I have replaced the transistor to verify that I wasn't working with a bad transistor, and I find it hard to believe that I would have two bad transistors, though that's not impossible. I even changed my connections between the emitter and collector to eliminate the chance that I installed them backwards. So far I've got nothing.

    I'm new to transistors, so bear with me, but does the fact that I'm getting a signal, weakened though it is, across the transistor, from base to collector, mean that it's trying to conduct, but something is preventing it?

    I've followed the circuits out many times, I'm sure it's all connected correctly. What would cause a transistor to not conduct?

  • #2
    Check the values of your collector and emitter resistors, it looks like one or other of them is wrong. I'd guess the emitter resistor is 390k instead of 390 ohms?
    "Enzo, I see that you replied parasitic oscillations. Is that a hypothesis? Or is that your amazing metal band I should check out?"

    Comment


    • #3
      Hi,
      I'm with Steve on this one.
      If you're positive the BJTs you tried are OK, the thing could be improperly biased ( a 0,04 VDC drop across the collector resistor means the collector current is 4 microAmpere ), so, as Steve said, check that the resistors' values ( colors ) are correct ( I'm not questioning your ability to read the resistors' color code, I'm rather questioning the way some resistors' manufacturers print some colors - It happened to me once that what seemed to be "red" was in fact "orange" - and I'm not colorblind - leading me to a 10x mistaken reading ).

      Also, test them to make yourself sure they haven't drifted on the high side.

      The 0,9 VDC you measured at the base seems to indicate that the two base resistors are probably OK, though I'd expect the base voltage to be somewhat lower ( about 0,8 VDC ).

      Hope this helps

      Best regards

      Bob
      Hoc unum scio: me nihil scire.

      Comment


      • #4
        Thanks for the reply's. I'll recheck all resistor values to make sure they're correct. I'm hoping you're right. I'd feel a little foolish, but perhaps I'd get this thing working! That's a fair trade off.

        Comment


        • #5
          Well kudos to you for at least checking all the pin connections against datasheets. I'm constantly amazed by how many different ways there are to misidentify those 3 simple pins. Easily 40% of the queries I field about nonfunctioning circuits that involve discrete transistors end up being a case of pin mistaken identity. So your first step is clearly the correct one.

          I can strongly recommend the advantages of running an electrolytic cap to ground from the emitter as a means to boost gain. As with triodes, the amount of bost and the point at which the boost kicks in will be set by the cap value and any resistance in series with the cap. Sadly, those linked-to schematics, drawn several decades ago by my buddy RG Keen, do not show such use of caps (they also show an error in the Muff Fuzz, connecting the diodes to V+), but if you poke around for similar schematics you'll see them in action.

          Depending on what diodes you used, the clipping may not necessarily be all that evident in the absence of added gain, even with the variable resistance set to minimum.

          Not to take away from your clearly diligent first effort, but take quick peek at this one and see if it comes close to what you were aiming for: Build The Tweak

          Comment


          • #6
            Thanks Mark, I've seen that schematic and mine is not that much different I suppose, the biggest difference is the clipper circuit. On the Tweako the clipper circuit feeds back to the bas, and is paralleled with what looks like a mid scoop circuit? Plus the split load at the collector resistor, w/ the upper resistor paralleled with the .01uf cap, not sure what going on there, suppressor cap?

            Re the bypass cap (it is a bypass cap, right? I'm associating it with a cathode bypass cap on a tube gain stage) on the emitter resistor, what values? I've got common values I use in amps, .68uf, 4.7uf, 10uf, 25uf. Whats the formula for frequency roll off?

            Re your comments that the clipper circuit may not be that evident, is that due to low signal level? I suppose the added gain from the emitter bypass cap would help with that.

            Comment


            • #7
              Originally posted by hasserl View Post
              Thanks Mark, I've seen that schematic and mine is not that much different I suppose, the biggest difference is the clipper circuit. On the Tweako the clipper circuit feeds back to the bas, and is paralleled with what looks like a mid scoop circuit? Plus the split load at the collector resistor, w/ the upper resistor paralleled with the .01uf cap, not sure what going on there, suppressor cap?
              Well I'm not as nimble with discrete stuff as I am with chips, but my understanding is that a lower impedance path provided by C6 will have an impact on top end. But don't quote me on that.
              Re the bypass cap (it is a bypass cap, right? I'm associating it with a cathode bypass cap on a tube gain stage) on the emitter resistor, what values? I've got common values I use in amps, .68uf, 4.7uf, 10uf, 25uf. Whats the formula for frequency roll off?
              I actually don't know what the formula is. I just do it empirically...by ear. However, values around 2u2 to 47u could work well in the circuit you described. One tweak to consider is using a pot and 2 caps of different values. Consider the following: The wiper of a 5k-10k pot is connected to the collector. Each outside lug of the pot goes to a separate cap, which is then tied to ground. When the pot is dead center, you get no boost. Rotate the pot towards one side and you get gradually increasing boost at one of two "inflection points". So, you could have a 680n-2u2 cap on one side and a 22u-100u cap on the other. One way gets you treble boost, and the other gets you full-range boost.
              Re your comments that the clipper circuit may not be that evident, is that due to low signal level? I suppose the added gain from the emitter bypass cap would help with that.
              Yes. Those diodes are gonna need a half volt in each direction to clip. While a nice Townsend "windmill" strum will get you that, the thing is that such peaks don't last long, and in the absence of any appreciable gain they drop below 500mv (clipping threshold) shortly after pick impact. As such, you'll get some bite on pick attack, but nothing that hangs around long enough to be considered "distortion"...that is unless you inject some boost.

              Comment


              • #8
                Hasserl,
                Why don't you try a couple of Germanium diodes ( 1N34A, 1N270 and the like ) ?
                Their forward voltage is lower ( between 200 and 300 mV ) so they'll clip earlier and more easily. Another "option" would be to put in 2 couples of diodes ( one couple Si, the other couple Ge ) and add an SPDT switch to choose between "softer" and "harder" clipping.
                The resulting waveform will be closer to a square wave, so it will enhance odd harmonics.
                Should the diodes' clipping action result in too fuzzy a sound ( due to the above mentioned effects ), you could try "asymmetric" clipping, by using one diode for one half-wave and two diodes in series for the other half...Asymmetric clipping adds even harmonics and softens somewhat the sound.

                HTH

                Best regards

                Bob
                Hoc unum scio: me nihil scire.

                Comment


                • #9
                  Originally posted by Mark Hammer View Post
                  I can strongly recommend the advantages of running an electrolytic cap to ground from the emitter as a means to boost gain.
                  This is absolutely true, but the negative feedback across the unbypassed emitter resistor is what makes this circuit the "Linear Power Booster", and not a Dallas Rangemaster or whatever other single-transistor boost.
                  "Enzo, I see that you replied parasitic oscillations. Is that a hypothesis? Or is that your amazing metal band I should check out?"

                  Comment


                  • #10
                    OK, you guys were right, I had a 390k resistor on the emitter. My bad. I'm embarrassed about the newbie mistake, but I'm happy it's now working, and I thank you guys for helping out.

                    I am getting some clipping out of the diodes, and in fact the clipping will even cause a drop in volume. I'm assuming this is due to a loss in voltage as part of the signal is clipped away to ground, though I didn't take any voltage measurements yet (I was too busy playing!). Is this normal for diode clipping in a pedal? I've done quite a bit of clipping diode removal in Carvin tube amps, and that does tend to raise the gain of the amp while reducing the distortion, something a lot of folks have a hard time grasping.

                    The emitter bypass cap adds a good amount of gain, but it's a bit too much now as there isn't much of a soft boost. In other words, if the level control is turned up enough to just boost the volume above the unboosted level, the tone is rough and unpleasant. Turning it up all the way provides a great overdriven tone, and that's cool, but I'm looking for something a bit less gritty at lower settings. So. I've got some tweaking and tuning to do with it yet. I might just connect the bypass cap to the footswitch for the diodes and only engage it when the diodes are engaged, that would overcome the loss of volume effect, would actually give it a boost as the diodes are engaged.

                    Anyway, thanks for the help and the suggestions. Robert I might play around with some other diodes as you suggested, see what effect I can get from them.

                    Thansk again,
                    Hasse

                    Comment


                    • #11
                      The topic of diodes comes up regularly, largely because there are so many ill-informed beliefs out there. The rule of thumb about diodes is that they set a ceiling on maximum signal amplitude. If there is any additional gain stage after the clipping diodes then the final output level is not solely governed by the diodes. If, however, the signal goes directly to your output (or via a pot), then the maximum possible output is dictated by the diodes.

                      If the diodes have a high forward voltage, or if there are more diodes used in series, then the likelihood of clipping is reduced, but the output is pretty much unconstrained by the diodes, therefore louder. If the diodes have a low forward voltage, then you'll get much more clipping, but the ceiling is set pretty low so you'll lose output level. Clearly, the diodes you are using are lowering max output level.

                      Max output would come from something like a pair of red LEDs, or a 3+3 configuration of silicon diodes (1N914, 1N4001, etc.). Low output level (but lotsa clipping) would come from a pair of germanium or schottky diodes. A balance between clipping and output level would come from somethng like a 2+2 configuration of germaniums, or a pair of 1N914s/1N4001s, or even a 2+1 configuration of silicon diodes where one half-wave clips more readily/often than the other.

                      People describe such asymmetrical diode arrangements as providing asymmetrical clipping, and equating it with tube tone. Ummmm......not so much. The asymmetrical diode arrangement does provide for a different sort of clipping sound but the asymmetry really only kicks in for certain signal levels. A 2+1 configuration, however, does permit a little more dynamic range (or rather, less dynamic restriction) than a 1+1 setup.

                      You'll find loads of useful and well-written articles on diodes and diode clipping here: AMZ Lab Notebook & Guitar Effects

                      Comment


                      • #12
                        Originally posted by hasserl View Post
                        OK, you guys were right, I had a 390k resistor on the emitter. My bad. I'm embarrassed about the newbie mistake, but I'm happy it's now working, and I thank you guys for helping out.


                        Anyway, thanks for the help and the suggestions. Robert I might play around with some other diodes as you suggested, see what effect I can get from them.

                        Thansk again,
                        Hasse
                        Hi Hasserl,
                        I'm really glad you sorted that out, don't feel embarrassed, to make you feel better, I will tell you about the time I was trying a Marshall JTM30 out and I was unable to get any sound out of it....I played around with ALL the knobs for some 5 minutes before realizing I had plugged the guitar into the footswitch socket....the owner of the JTM30 ( luckily a dear friend ) is still laughing about it!
                        Cheers
                        Bob
                        Hoc unum scio: me nihil scire.

                        Comment

                        Working...
                        X