[Stillhart]

So I'm working on my first CMoy amp and I'm trying to build one from scratch rather than a kit. I'm following the tutorial here and there's something missing: when putting the board into the metal tin, should there be some kind of insulation or something so the contacts on the bottom don't short against the tin?

I feel like this is a silly question for such an old design but hopefully someone can just help me out quick.

 

[Raketen]

should there be some kind of insulation or something so the contacts on the bottom don't short against the tin?

yes (IIRC JDS recommends using hot glue to make some standoffs on the pcb, but a layer of nonconductive material that won't get pierced/worn by sharp solder spots would also work).

 

[Stillhart]

yes (IIRC JDS recommends using hot glue to make some standoffs on the pcb, but a layer of nonconductive material that won't get pierced/worn by sharp solder spots would also work).

Hm, good point about the sharp solder spots. My first thought was just like a layer of electrical tape or something but that won't hold up. Thanks, I'll try to look more into it!

 

[Stillhart]

Another dumb question: The circuit diagram shows 2x 9V batteries but the tutorial only talks about installing one. Am I good to just throw two in wired in serial (?) or is there more... stuff... required in the power section if I do that?

 

[Stillhart]

Is @tangent still active on here?

 

[whirlwind]

So I'm working on my first CMoy amp and I'm trying to build one from scratch rather than a kit. I'm following the tutorial here and there's something missing: when putting the board into the metal tin, should there be some kind of insulation or something so the contacts on the bottom don't short against the tin?

I feel like this is a silly question for such an old design but hopefully someone can just help me out quick.

Just looked at mine and it looks like hot glue was placed under my board, or clear silicone...I am guessing hot glue.

 

[tangent]

My first thought was just like a layer of electrical tape or something but that won't hold up.

Indeed it will not. Ask me how I know.

The circuit diagram shows 2x 9V batteries but the tutorial only talks about installing one. Am I good to just throw two in wired in serial (?) or is there more... stuff... required in the power section if I do that?

The short answer is that both configurations can work.

The long answer is that it depends on matters discussed in other articles on my site. You may find these most immediately helpful:

• Virtual Ground Circuits
  
• Op-Amp Power Supply Quality Considerations
• Notes on Audio Op-Amps

Is @tangent still active on here?

I scan this sub-forum thread titles occasionally, but must have missed this one when it came up the first time. But speak a demon's name, and ye may summon him.

 

[Stillhart]

Just looked at mine and it looks like hot glue was placed under my board, or clear silicone...I am guessing hot glue.

Thanks, I saw this as an option after looking into it more. I am still not sure what I'l end up doing; the casework is my biggest worry with any DIY project. I'll see how much room I have once everything is assembled and I do a "dry run" with everything in place but not connected.

Indeed it will not. Ask me how I know.

The short answer is that both configurations can work.

The long answer is that it depends on matters discussed in other articles on my site. You may find these most immediately helpful:

• Virtual Ground Circuits
  
• Op-Amp Power Supply Quality Considerations
• Notes on Audio Op-Amps

I scan this sub-forum thread titles occasionally, but must have missed this one when it came up the first time. But speak a demon's name, and ye may summon him.

Hey thanks so much for piping in! I've been following your tutorial on the CMOY and I've looked into a lot of the articles and video tutorials you've made. What an amazing resource, thanks so much for putting it all together!

I've skimmed some of the noted articles but my understanding of electronics is still VERY rudimentary; I am having a hard time understanding some of the more complicated topics. Per your suggestion, I picked up that Art of Electronics book and I've started working through it so I hope to understand a lot more by the time I'm done with this project.

One thing I'm proud of (which isn't really MUCH of an achievement): I ended up getting the SB300 breadboard instead of the Radio Shack one in your tutorial so I had to redo the layout of the components on the board. I used your wiring diagram and the layout pdf to work backwards to an understanding of what was going on, then I worked out a new layout (similar to yours) on the SB300. I felt like it was a great exercise to understand the relationship between the wiring diagram and what's going on on the board.

I haven't finished assembling my amp yet, just started last night, but the power section works so that's something! Of course the LED doesn't turn on so I plan on fixing that first thing tonight. What's the best way to test an LED without blowing it out (I know you said that running it right off the battery without the resistor can blow it out in a matter of seconds)? What's the best way to test the + vs - poles to make sure it's in in the right direction? If I understand diodes correctly, current can only flow in one direction so running it backwards will just do... nothing.

Anyways, I *think* I can just run a 2nd 9V battery in serial with the first without breaking anything. I believe I read on your site that adding a 2nd battery can actually improve the stability of the circuit assuming the Opamp can take it. I'm using an OPA2132PA so I think it should be fine?

Anyways, thanks for all the great info. I still have a lot to learn...

 

[tangent]

What's the best way to test an LED

If it's a plain old red, amber/yellow, or green LED, many DMMs' diode test functions are capable of lighting the LED safely. Chances are decent that you just have it soldered in backwards, and the resistor may have protected it from damage. (Unlike a normal diode, LEDs are not very good at blocking reverse voltage.)

If it's a more modern LED type, many DMMs can't develop enough voltage to light the LED, so they're not very helpful.

Lacking a DMM in your situation, or at least one capable of doing the test, I think I'd just desolder the LED's hookup wires from the PCB and touch it to a 9V battery both directions until I found out which side was positive. If neither direction lit it up, I'd call it dead and move on.

Going forward, be sure to pay attention to which leg is the short one by consulting the LED's datasheet. There's a convention, but it isn't binding; some LEDs are "backwards." You can also learn to study the LED's inner construction through the lens to determine which terminal is the anode.

I'm using an OPA2132PA so I think it should be fine?

That's the point of the voltage tests in the op-amp notes article. Notice how close to 9V you get before you start clipping with that op-amp with headphones requiring higher voltage. That means you get very little use out of the battery, as pointed out by another of the articles I pointed you to. A third article I pointed you to brought up the issue of virtual ground shift, which eats into this still further.

More voltage means you drain both batteries more fully, making for a more efficient whole.

I still have a lot to learn...

That's the thing: the circuit is simple, at one level, but fully understanding it requires that you grasp a large swath of the field of electronics. It's a good teaching tool because you aren't required to grasp the whole swath at once, however.

 

[Stillhart]

If it's a plain old red, amber/yellow, or green LED, many DMMs' diode test functions are capable of lighting the LED safely. Chances are decent that you just have it soldered in backwards, and the resistor may have protected it from damage. (Unlike a normal diode, LEDs are not very good at blocking reverse voltage.)

If it's a more modern LED type, many DMMs can't develop enough voltage to light the LED, so they're not very helpful.

Lacking a DMM in your situation, or at least one capable of doing the test, I think I'd just desolder the LED's hookup wires from the PCB and touch it to a 9V battery both directions until I found out which side was positive. If neither direction lit it up, I'd call it dead and move on.

Going forward, be sure to pay attention to which leg is the short one by consulting the LED's datasheet. There's a convention, but it isn't binding; some LEDs are "backwards." You can also learn to study the LED's inner construction through the lens to determine which terminal is the anode.

Thanks. The diode setting on the DMM did the trick. It turns out I had it in just fine. I'm using a amber/yellow LED and it's not super bright and the resistor makes it even less bright so I literally couldn't see it under my bright work bench fluorescent. /embarassed but learning

That's the point of the voltage tests in the op-amp notes article. Notice how close to 9V you get before you start clipping with that op-amp with headphones requiring higher voltage. That means you get very little use out of the battery, as pointed out by another of the articles I pointed you to. A third article I pointed you to brought up the issue of virtual ground shift, which eats into this still further.

More voltage means you drain both batteries more fully, making for a more efficient whole.

Thanks again. I started reading that stuff and hadn't put it all together. I'll go back and study it again with this summary in mind. I'll also go ahead and do that voltage test for some headphones I'll be using with it and see how it maths.

That's the thing: the circuit is simple, at one level, but fully understanding it requires that you grasp a large swath of the field of electronics. It's a good teaching tool because you aren't required to grasp the whole swath at once, however.

Slowly but surely!

 

[tangent]

the resistor makes it even less bright

The implication is that if it wasn't for the resistor, the LED would be brighter. That's true, but only very briefly.

LEDs have very little inherent current-limiting ability. Contrast an incandescent lamp, which is a type of resistor, which is a current-limiting device. Specifically, a modern incandescent lamp's filament is a positive temperature coefficient resistor, meaning that it starts out low in resistance but increases in resistance as the filament's temperature goes up, so that the system quickly achieves a steady equillibrium as long as the supply voltage is steady.

LEDs have a nearly-fixed forward voltage. It varies slightly on temperature, current, and such, but it's nearly constant. This means that if you take a yellow LED with a typical forward voltage drop of 1.9 V and put it across a 9 V power supply, the main thing that tries to take up the excess 7.1 V is the wiring, and it takes a *lot* of current to drop 7.1 V across the negligible resistances involved. Mathematically, I=V/R, so if V=7.1 and R is nearly 0, I is very large. Since a typical LED is rated to withstand only about 20-40 mA, it dies pretty quickly when given that load.

That brings us to the advice in the tutorial to start with 10 kΩ and go down as far as maybe 1 kΩ. If you work the math, you'll see that with a 9V supply, the absolute minimum safe value is 355 Ω for a 1.9 V LED.

I'll go back and study it again

I've added advice on insulating the bottom of a mint tin, and I've added a similar set of links to power supply info to the tutorial.

 

[Stillhart]

Thanks for adding that stuff! I finished with my first swipe at a CMoy using no special fancy anything, just the basic circuit and one battery. I was going to go with 2, but I wasn't able to fit everything in the case. I don't see how I can possibly get the board in there under the IO jacks without notching the lid (or board-mounting the connectors).

I bought enough parts to make a second one, including upgraded resistors. I'll probably try to get fancy with the power supply and caps on the second one and see how the sound compares. I might also look into alternative IO jacks to see if there are smaller ones since I don't use the disconnect sensor thingys.

Oh and I'm going to lower the gain. With my MrSpeakers AEON (easy to drive), it's barely out of channel imbalance area when using line out voltages.

Anyways, thanks again for the great tutorial! I'm studying everything you've posted and trying to wrap my head around all of it. Fun times!