Messages

I don't know why I've never done a Small Clone project. I probably should.

Anyway, that is one awesome looking build, marauder!

PM is best for now. The email thing will be fixed ASAP. This weekend is when I should finally be able to switch to the new host (which I've said every weekend for the past month, but then something else comes up).

Both the NE5532 and JRC4558 have BJT inputs. So, why one would cause a problem and not the other, I'm not sure. Unless there's something with input offsets, maybe? I know I have used the 4558 in this circuit in the past.

On the other hand, perhaps the problem lies in the clipping portion. Depending on how involved you want to get in this: you could try temporarily shorting the two mosfets and see if the problem persists with only the BAT41s working. This might be a good guess since the crackling is layered on top of the signal.

That could be a bad IC, actually. I would swap it for another. If that doesn't work, check your voltages and post them here.

It's a PT2399 chorus. Still working on it, but it's mostly there.

This topic has been moved to Cadsoft Eagle Resource.

http://www.madbeanpedals.com/forum/index.php?topic=1327.0

I don't use pads with supply pins anymore...just vias in layout.

So do you just crank up the drill size once you've dropped the via? And then I assume you copy the enlarged pad and rename for the next use?

The only vias I usually copy are grounds. The other ones, IN, OUT, 9v, etc are just used once. So, in layout I set my via parameters (47mil drill, 86mil diameter), drop them in and then name them. Eagle then connects the nets for you and you can route the trace however you need.

For layer vias, I do something similar but it's even easier. I set the parameters (.35mil drill, auto diameter) and start routing the nets. When I want to drop a via, I first route the trace up to the point where I want to insert it, single click once, then use the middle mouse button to change to the other layer. Once you do this, Eagle drops the via for you automatically

Ah that explains as usual, I'll go for the obvious question: why is it much better?

The main reason is that using flags to name nets means you can have as many unique supply pins as you want. For instance, I've had a few layouts where I've had as many as six supply lines that need to be hooked up. Using pins means you have to have a unique supply pin for each in your Eagle library. Anytime you want more, you have to create more pins in the library. Doing it the other way way gets the same result without having to use any supply pins. You just draw a net, name it, then use the flag to identify it. If you name another net the same thing, Eagle will automatically ask you if you want to connect them together. This is really convenient!

I don't use pads with supply pins anymore...just vias in layout. Actually, the only supply pin I use now is ground. Everything else I'm just naming nets and using flags to label them. This is a MUCH better way to do it than supply pins.

If you want to just shift the Rate further down the register, increase R35. Try 1M5 or 1M8.

Have you tried a pullup resistor connected to the gate? Try a 1M @ C1/R2 connected to 9v (keep R2). This will set a constant DC on C1 which might solve the problem.

I wish I had an oscope, but they are like $300 USD for the low end ones.  There are cheaper ones out there, but I don't think the sampling rates are good enough, they are just toys.

This would get you started, at least.

http://www.seeedstudio.com/depot/dso-nano-v3-p-1358.html?cPath=174

Is your regulator putting out 5v as it should?

The pinout should be EBC. You don't have to use BC550, though. You can try 2n3904, or probably even 2n5088/89. Those are CBE, though, so make sure you flip them the right way.

In a single rail supply, ground is 0v. Bias voltage (VB) is created by a voltage divider consisting of two fixed resistors which are usually the same values (but not always). One in series with the supply, and another right after it to ground. This divides the voltage into a bias supply. When the resistors are of equal value, the result is 1/2 of the supply voltage. When the values are mixed, it is directly proportional to the ratio of the the two values.

The bias voltage pulls up our input DC voltage to whatever is required for a particular IC or transistor we are using as an amplifier. Sometimes there is a range of voltages that this will work over, and sometimes we might under/oversupply the bias voltage for a particular effect. In the case of a non-inverted input on our ICs and transistors we also use another resistor between the bias voltage and the input. This resistor is current limiting. It also sets the input impedance into the amplifier. Inverted inputs to ICs work a little differently, and JFETS don't always require a connection to bias voltage (although it can be a benefit in some cases).

In a split rail supply, ground is still 0v, but it can also act as "bias". This is a handy trick to simplify designs, and the exact thing I used in the Kingslayer, which is a split rail design.

Just to further clarify on the reverb schem above: common is the bias voltage. In a single supply design, it will be a voltage divider (VB). If that circuit were used in a split rail supply, common would be ground. So, really what they are doing is describing the circuit without dictating a particular power supply scheme.

To get a better sense of this, take a look at some examples. For instance, look at the Green Bean schematic to see how the bias voltage is created from the supply and then run to the inputs/bases of the ICs and BJT transistors. Now look at the Kinglsayer schematic. See how bias and ground are the same thing in that design.

That looks like a good solution right there. Maybe the best thing to do is chassis ground it? So, rather than connecting it to common ground with the two PCBs, run the ground wire from the copper plate to an o-ring which fits over the 3PDT shaft.