New Clipper problems

[black mariah]

I just built a New Clipper, but I had to make a couple of part substitutions because Radio Shack sucks my left tes... wait, I'm preaching to the choir.

I had to use 100k and 100 resistors in place of the 150k and 150 that it calls for. Also, quite by accident, I ended up using non-polarized eloctrolytics. Also, I had to use a .022 in place of the .033 it calls for. As you may have guessed, the thing doesn't work. I get sound when I hit the strings hard, so I know it's a bias thing (thank you, Geofex!). I'm pretty sure it's the resistors, so no big deal.

BUT, could it be the capacitors? Does polarized vs non-polarized make a difference? I am a COMPLETE IDIOT when it comes to knowing what's going on in a circuit. I have no idea what the changes I made did. All I know is, it doesn't work and I'm too tired to mess with it right now. My multimeter isn't here anyway. :x ANYWAY, I'm rambling because I'm tired. The polarized/non-polarized thing is what's important to me right now.

[Paul Marossy]

Assuming that you have checked for bad solder joints, etc., the first thing that I would do is change the 100K and 100 ohm resistors to a value as close as possible to what is specified on the schematic. When I built mine, I believe that I used a 100K and a 47K in series and a 100 ohm and 47 ohm in series, which gives you 147K and 147 ohms. That may be part of the problem here.

The caps specified on the schematic are polarized. I don't know for sure if non-polarized electrolytic caps would cause you grief or not, but it's possible, I think. I'll let someone more knowledgeable than myself answer that one. I have wondered about it before myself.  
The .022uF cap substitution shouldn't have much of an effect other than lessening the bass a little bit.

[black mariah]

Okay, interesting event. I replaced the TL072 with a different one, and now I get output! Too bad it's so low I have to crank the clean channel on my amp to about 5 to get any sound. :shock:

I reflowed the joints, and that didn't help. I'm completly lost now. The debugging pag says NOTHING about what to do when there's a low output. Found my multimeter at least, so what can I poke to tell me why I have a low output? :lol:

[aron]

>I had to use 100k and 100 resistors in place of the 150k and 150 that it calls fo

That should be ok.

>Also, I had to use a .022 in place of the .033 it calls for

Hmm... it calls for a .003, not .03


>I get sound when I hit the strings hard, so I know it's a bias thing

Check your resistors that connect to the non-inverting input of the op amp. These would be the 22K and 22K junction. In fact, you could have used two 10K resistors. I suspect the problem is here. Using your multimeter, check the voltage at the input pin. It should be roughly half the voltage of the battery. If it's not, there's your problem.

Non-polarized electros WILL WORK. This is not your problem. Your problem is probably the wiring of the two 22K resistors and possibly some other problems after the op amp.

[black mariah]

> Hmm... it calls for a .003, not .03

Wow, you're right. The schematic says .0033, but the LAYOUT says .033. How much of a problem is this?  :shock:


> Using your multimeter, check the voltage at the input pin. It should be roughly half the
> voltage of the battery. If it's not, there's your problem.

Guh, I've never checked voltage before. Let's make sure this is right (I feel like such a stupid noob... oh, I am. :cry:  ). Battery connected, black to ground, red to whatever I want tested, correct? If that's the case, I'm getting about 8.7v at the #3 pin. I'm guessing this is not a good thing.

>Non-polarized electros WILL WORK. This is not your problem. Your problem is probably
>the wiring of the two 22K resistors and possibly some other problems after the op amp.

Thanks for the clarification. I checked the FAQ and noticed that buried in there after I asked. The 22k resistors seem to be working just fine. Each one registers just about 22k, and they measure 43k together.

Thanks a lot for all the help, seriously. There's no way I'd even know where to begin if I had to do all this on my own.

[B Tremblay]

I see that you used a TL072.  If you followed the perfboard layout, that could be the source of your issues, since it is for a TL071.  The TL071 is a single op-amp with a different pinout than the TL072, a dual.

The 741 from RadioShack uses the same pinout as the TL071 and worked fine in my build.

Regarding the discrepancy between the schematic and perf layout, always follow the schematic.  I'll fix the error on the layout, thanks for catching that.

[black mariah]

I am really, really, REALLY electronics stupid. Everything I'd read made it sound like the TL071 and TL072 were compatible chips. But, to my surprise, I just did some checking and ran across the TI page for these things... and I'm even more stupid. There's a 71, 72, and 74, which are single, dual, and quad opamps. *BANGS HEAD ON WALL* Ah, the obviousness... once you know it. :lol:

So, I'm guessing that using the RIGHT opamp will fix my problem. Sometimes living a block from a Radio Shack is a good thing. I didn't even notice the discrepancy between the schematic and layout. I read them both as .033. I will be way, WAY more careful with that from now on.

[Paul Marossy]

Well, at least I have an answer on the non-polarized electros now.  :wink:

As you have figured out, a TL071 and a TL072 are not compatible/interchangeable. One is a single opamp and the other is a dual opamp. This sounds like the bulk of your problem. Like Aron said, that 100K/100 ohm probably will work fine, but it will change the sound a little bit. I didn't catch that it is a .003 cap on the output - that could also swamp your output.

[aron]

Code Select Expand

So, I'm guessing that using the RIGHT opamp will fix my problem.

It probably will fix the problems  

You are probably approx $1 away from the fix!

Good luck!

[black mariah]

You are probably approx $1 away from the fix!

Pfft... more like $1.07. :lol:

But, sadly, no. I swapped the IC out with a 741, and now I get even less output. The voltage reading at the IC input is 4.7, which should be just fine. I'm totally lost now. Any more ideas? All the components check out as working.  :cry: I hate electronics. I'm not patient enough for this crap. I'm a guitar tech. I'm used to hitting things with a hammer to make them work. None of this sissy multimeter stuff. :wink:

[black mariah]

Okay, I just noticed something I forgot to do and if this fixes it, I will punch myself in the face. I checked the wiring on this freaking thing like 30 times and I SWEAR I never noticed this. :evil:

[Paul Marossy]

"Okay, I just noticed something I forgot to do and if this fixes it, I will punch myself in the face. I checked the wiring on this freaking thing like 30 times and I SWEAR I never noticed this."

Been there, done that. Sometimes it helps to take a break from it for a day or two and then come back to it. Then I usually find my error(s) and then I also kick myself in the butt, literally! :wink:

[black mariah]

I am a MORON. I didn't have the 4th IC pin grounded. So I have sound now, but the gain is low. The sound clips (http://www.runoffgroove.com/salvo.html#newclipper) have about twice as much gain as I'm getting. Crapspackle... more problems. On the plus side, now I know where to connect everything on an opamp. :lol: If anyone has ideas on the gain issue, I'm open to ideas. :lol:

[Mark Hammer]

It always helps to know what you're doing.

1) The 150k and 150R resistors.  The gain has to be high enough to produce clipping at the diodes.  The gain of the circuit is given by the ratio of (150k+150R)/150R, which is 1001 in this case.  That just doesn't seem quite right.  Unless you are feeding the circuit an input from single nerve cells, a gain of 100 to 200 ought to be more than sufficient to produce serious clipping with a pair of diodes.  Working backwards, if you stick with your 100k feedback resistor, and replace the 150R (100R in your case) with something between 200R and 1k, that should give you gain somewhere between 500 and 100, which is plenty.

Keep in mind that the 4.7uf cap to ground after the low-value resistor needs to be oriented proiperly, and that it will roll off at the low end depending on the value of the resistor before it.  If you use a 1k resistor, it rolls off around 34hz.  If you use a 470R resistor, it rolls off around 72hz.  (Use of a 150R resistor rolls off around 225hz, yet one more reason why I think the ROG schem is incorrect.)  If you use a smaller value resistor for more gain, you can offset this by simply increasing the value of the cap.  So, with a 470R resistor, increasing the cap to 10uf will lower the rolloff down from 72hz and hang onto more bass.

2) The .0033uf cap (mistakenly treated as .033) rolls off high end at a frequency determined by the combo of the cap and the 10k resistor ahead of it.  As it, it rolls off around 4.8khz.  Increasing the cap to .033 - whether on purpose or not - drops that to 480hz, a bit too muffled for a fuzz.  Decreasing it to .0022 raises that rolloff to 7.2khz, which may be a bit too fizzy for many, though still within reason.  If you stick a cap in parallel with the 100k (150k) feedback resistor, you will also roll off high end, only *before* clipping, rather than after.  Use of pre and post-clipping treble smoothing will get you a more vocal, less ragged and fizzy tone.  If it were me, I'd stick a 270-390pf cap in parallel with the 100k resistor for a rolloff in the 5khz zone or thereabouts.  Once the output of the op-amp goes past the diodes, there will be harmonic content added above that, but the .0022uf cap later on won't have as much fizz to clean up.

3) The '+' input of the op-amp (now that you know the difference between a dual and single op-amp) just seems plain wrong.  I've never seen that sort of biasing scheme or values used in that way.  What I would recommend is this.  The +9v line goes to a pair of 22k resistors in series, but there is a 330k to 470k resistor in between the '+' pin and the junction of those two resistors.  That ought to work just as well as it does on the many hundreds of comparable devices I've seen it on.  It would also help to stop clicking and popping if there was a 1 to 2.2meg resistor between "IN" and ground.

Taking all of that into account, you should come to the conclusion that you had erroneous information from the outset, and hold no blame for following it sincerely and producing something that wouldn't work properly.  The folks who provided it probably did what I'd do at first - look at the picture and decide it looks "normal" enough.

The New Clipper and Blue Clipper were intended to be ultrasimple "one-knobbers", but that doesn't mean they can't be more.  Indeed, the schematic comes VERY close to being a pared down version of an MXR Distortion+ and DOD 280/YGM308.  Adding one more pot will allow you to produce variable amounts of fuzz/distortion in addition to different volume levels of the same distorted sound.  Playing with the value of the 10k resistor will let you get different tones, and tinkering with the diodes will let you get different distortion qualities and dynamic functions.

So, when you rebuild this (and you will), make sure you leave a bit of space on the board for further mods, because we will lead you to them, my friend.  Oh yes.  Make no mistake.  We will.

[David]

This is the dreaded WTHIIDT syndrome.  I have fallen victim to it many times!  As Mark Hammer so eloquently put it,  you can stare at the same connection for months thinking that's it's right because you've looked at it so many times.  Suddenly, you realize it's not right.  This is the WTHIIDT (or "What the Heck Is It Doing There!") syndrome.

[B Tremblay]

For what it's worth, both Gary and I have built the circuit with satisfactory results according to that schematic.

I believe Paul Marossy and a few others have as well.

[black mariah]

Seriously. Check the sound clips. That's why I decided to go for it.

[Mark Hammer]

Yeah, maybe I was a little too "orthodox" there.

Obviously a pair of resistors can be used to "float" the '+' pin in the absence of a third current-limiting resistor, otherwise the Orange Squeezer wouldn't work at all, right?  And this IS a Dan Armstrong (bless his late soul) design anyways, so it oughta work.

It's the confusing values that throw me.  If the two resistors were, say, 220k each, it would make sense.  Even asymmetrical biasing makes sense, assuming one isn't feeding too much current into the non-inverting pin.  But 22k sitting between a 9V battery and a noninverting input seems a little risky to me.  

Personally, I'd feel much more comfortable with the biasing scheme I suggested.  If you want to stick a 50k pot  in with the pair of 22k resistors to monkey around with Vref, go for it, by all means.  I also wonder if the biasing current makes it a requirement to use specific op-amps.  That's just a guess, rather than any sort of rule.  Alternatively, if one is going to tinker with biasing for some sort of asymmetrical clipping, why not bias low rather than high?  I.E. why not make the existing resistor between V+ and the noninverting input the one you make bigger.  That way you decrease the input current at the same time as varying the Vref.

As for gain, asking a gain of 1000 from a measley little op-amp with a 9v battery behind it is a little overly "optimistic", if you ask me.  Not only that, asking for that much gain in the absence of any sort of HF rolloff is most assuredly going to make one's next project a noise gate!  Far better, from a noise standpoint, to set for less gain, add some HF rolloff, use GE diodes for a lower clipping threshold (hence less need for gain), and add a recovery stage to get the volume boost.

I don't doubt that the soundclips sound fine, but I also know I've built stuff that sounded great when tested out with a practice amp and 3" speaker but revealed serious design issues when I managed to listen with full bandwidth.  I'm not saying that happened in this instance, but it DOES happen.

[Paul Marossy]

Yes, I did in fact build the New Clipper. I built it according to the schematic on R.O.G., and it has a heck of a lot of drive. It can sound a little harsh when you are playing your amp quite loud and it has a helluva boost to it so I added a volume control to quiet it down a little bit. It seems to have plenty of bandwidth as far as I can tell.

In certainly not an expert like Mr. Hammer or RG, but I'm thinking that in this case maybe the specific opamp used has a bearing on the final outcome of how the circuit performs in the real world. I'm trying real hard to remember what opamp I used, but I am drawing a blank...

[black mariah]

So... anyone have any ideas on the gain issue? I'm getting little distortion and what I am getting is quite rough. More like a Fuzz Face than the smoothness of the sound clips. Would the additional resistor before the opamp input help, or am I just completely screwed?