[LazyAzN]

*** The problem I'm having is on page 6***

I'm just a complete amateur at building portable amps (and reading schematics while we're on it)

... Anywho, I've been wanting to make a CMOY for a long time after lurking for a while. I've read up on the William Neo blog for making a DIY CMOY along with Tangent's site on making one too. So after reading up on it, I decided I want to just make one from scratch instead of soldering to a PCB board (since I've done it before when I was making a robot from a kit) and stick it in an altoid tin...

Here's my parts list, which I utilized Tangent's parts list to make it (the protoboard will be from Radio Shack, rest from Digikey):

1X IC SOCKET .300 8 DIP GOLD AE10034-ND
1X POT 10K OHM 12MM HORZ MET BUSHIN P2U4103-ND
1X KNOB CLR/MATTE.50"DIA .250"SHAFT 226-3100-ND
1X SWITCH TOGGLE SPDT 6A SLDR 5PCS 360-1788-ND
1X STRAP BATTERY 9V I-STYLE 4"LEAD 2238K-ND
2X 3.5mm Stereo Panel Mount Jack 274-0246 (Parts Express, friend)
1X LED 3MM GREEN DIFFUSED 160-1710-ND
1X Multipurpose PC Board with 417 Holes 276-0150 (Radioshack)
2X (C1)CAP 470UF 25V ELECT VZ RADIAL 493-1304-ND
2X (C2)CAP POLYPROPYLENE .47UF 50V 2% P3933-ND
1X (R1 replacement)IC 1/2 RAIL VIRTUAL GND TO-92 296-1994-ND
5X (R2)RES 100K OHM 1/4W 1% METAL FILM 100KXBK-ND
5X (R3, Gain 6)RES 2.00K OHM 1/4W 1% METAL FILM 2.00KXBK
5X (R4)RES 10.0K OHM 1/4W 1% METAL FILM 10.0KXBK-ND
1X IC DUAL PRECISION OPAMP 8-DIP OPA2228PA-ND

Anywho, here are questions (that might be answered as I research) that I have as a first time attempter/DIYer/etc.

Will this actually work with the list as is, or do I have to make modifications for the OPA2228PA op-amp? I know that it requires at least a gain of 5, which is why I chose 6, since I know 11 is too much.

What is the difference between the PA and P version anyways (besides one costs $3 more)? Is there a better op-amp besides it (which will be very opinionated)? Is the 2228's pinout the same as something like the 2134's?

I've also read up that I may need to solder a feedback resistor to keep the OPA2228 stable, would that be the case here? If so, where would I have to solder it (the schematics of Tangent's to the average joe like me = not understandable one bit).

Is there a layout that utilizies the TLE2426CLP, because I'm not sure if I can just drop it in and replace the resistors for R1 for the virtual ground.? Actually, I'm not even sure how different this will be from Tangent's layout... So if it is MUCH different, if it can be designed for me, thanks...

Do I also need a resistor for R5? If so, what kind of resistor?

I've also read on William Neo's DIY CMOY that he needed a R6 for his power LED, do I also need to do the same? If so, where would it go? The same place he put his??

Another amateur question, how will I be able to tell which leg of the resistor goes in the slot on the layout since each leg has either a negative or positive side? I know how to read positive or negative legs ON the resistor, I just need to know how to tell on a layout.

Two more amateur question... I just need to solder each part to the copper on the bottom of the protoboard right??? (never seen the RS protoboard, so I'm verifying so I don't have to start making traces on the bottom)... Also what is a good gauge to use to hook up the wires?





Thanks for your help...

Sorry my post is soooo LOOONGGGG....

 

[xmokshax]

i haven't used the OPA2228P/A, but it sounds like it might be a bit more ornery than the old standbys, the OPA2132 and 2227. why not start with an op-amp that's easy to work with, then tweak your circuit later if you want to? nothing will be more frustrating than to have to do a bunch of troubleshooting on your first DIY audio project. go for the "PA" version of whatever BB op-amp you get - they're cheaper and will work the same as the "P" versions for audio. this answer can be found on tangent's site.

as far as switching from the resistor divider to the TLE2426, it's pretty trivial, and is covered in the first part of this article on tangent's site.

under ideal circumstances, you won't need a resistor at R5 - you'll just jumper it. if your amp is noisy, you might add a resistor between 10 and 100 ohms (metal film, 1% tolerance, same as the others in the signal path).

you'll need an LED resistor - it's called RLED in tangent's schematics, not R6. it can be a plain old +/-5% carbon film resistor, and its value can be calculated based on the information here (as you can see, tangent really has put together an exhaustive tutorial on the construction of a CMoy, and he has included answers to just about every question a DIY newbie might ask). the LED resistor can go in line with either leg of the LED, but make sure you get the LED polarity right. the side with a bigger metal piece within the body of the LED, or the side with the shorter lead, is the negative side.

in answer to your resistor polarity question, resistors don't have a polarity - there's no negative or positive side of a resistor. they can be soldered in in either orientation.

you only need to make two connections (the power supply stage to the amplifier stage) aside from those specified by the jumpers in tangent's schematic. everything else just gets soldered to the copper pads. 22 or 24AWG hook-up wire will work nicely. stranded wire is a bit easier to work with than solid-core.

 

[infinitesymphony]

OPA2228 will oscillate in a stock CMoy because the gain bandwidth product is too high (too fast). I found this out first-hand (

). My solution was to add two decoupling capacitors connecting the op-amp's V+ to ground and V- to ground. This worked for me, and allowed the use of cranky op-amps like the OPA2228 and LM4562.

Here's a photo (sorry about the blur):

 

[LazyAzN]

You used some cheap 0.1 uF ceramic capacitors to stop the oscillation(google shows all

), right infintesymphony? Would that count as the feedback resistor?

As for still understanding the schematics & language of what Tangent is saying (even though it should be for the average DIYer/techie), I still don't get it... Ahhh... Well I somewhat do... just don't know where to place it on the board... or anything ELSE besides what's on the pretty layout (since pictures means 1000 words)...

So I'm going to see if I can customize Tangent's layout because I have no clue what the heck I'm doing in terms of placing new parts... Woo for learning ya?...

As for the RLED, I chose a 4.75 k OHM 1/4w metal film resistor, since 6.9 / 4.75 and 5.9 / 4.75 would still be above 1 ma(if 9V's die @ 8V)...

Is there anything else I should change besides the op-amp (I'd rather work on a failure than a success which would make me want to upgrade more... It makes the whole success thing oh so much better, which may not be good thinking...).

 

[balrog]

I followed Tangent's tutorial to build a stock Cmoy amp with a gain of 11. That was way too high. Now it is at a gain of 3. It was not hard at all and the forum members are great at helping you through troubleshooting. Go with the stock version then start doing mods so you can hear the differences in the various upgrades.

 

[MisterX]

Quote:

So I'm going to see if I can customize Tangent's layout because I have no clue what the heck I'm doing in terms of placing new parts... Woo for learning ya?...

Let me save you a little effort

http://i23.photobucket.com/albums/b3...evisedcm0y.png

 

[LazyAzN]

Quote:

Originally Posted by MisterX /img/forum/go_quote.gif Let me save you a little effort http://i23.photobucket.com/albums/b3...evisedcm0y.png

Whoa...

that's awesomee... thanks!... Although I need a double check to be sure...

Do the blue dots verify the r.out and l.out (since the jumper seems to be modified slightly)? Plus... pink dots are... *draws blanks...*...?

Also one last question on layouts... For the resistors where they mark the hole, does that mean one leg goes in one hole and the other in the other? So the resistor jumps over a few holes (as in spanned across)?... Along with how does one cut a protoboard? Exacto knife?


But yes... you didn't save me a little effort... you saved me a big amount of effort.... When I mean big... I mean I'd be bonking my head on my desk wondering where it goes because I don't know squat about it... Which would equate to a couple hours trying to figure it out... Something trivial to the experienced will be a puzzle to the amateur... Thank you!!...

Of course I need to start buying the parts...
Is there any preference to the ceramic capacitors besides the 0.1 uF?

 

[balrog]

I scored it on both sides using a straightedge and razor knife then snapped it on the edge of the desk.

 

[MisterX]

Quote:

Do the blue dots verify the r.out and l.out (since the jumper seems to be modified slightly)? Plus... pink dots are... *draws blanks...*...?

Blue dots are the outputs when R5 is placed outside of the feedback loop.
Pink dots are the outputs when R5 is placed inside the feedback loop.
The problem with this setup is ideally R5 should be like 10 times smaller when it is inside the feedback loop (somewhere around 6 ohms) so you can't just change the wire and get.......

The modded jumpers are just my way of doing it, it never made much sense to me to cut the leads off the resistors and then solder them back to the board when you could just bend the leads over, feed them through the hole, solder them and then clip the extra off to get the same thing. I also have a different way of connecting "the power rails" so maybe a photo is prudent?

http://i23.photobucket.com/albums/b3...e/HPIM1417.jpg

Quote:

For the resistors where they mark the hole, does that mean one leg goes in one hole and the other in the other?

Yes, the black dots on the image represent the solder joints.

Quote:

Along with how does one cut a protoboard?

A Dremel rotary tool with an abrasive cutoff disk works for me. It's kind of messy though.

Quote:

Is there any preference to the ceramic capacitors besides the 0.1 uF?

No, I would avoid disk type ceramics but other then that even 1206 smt caps will fit there so prolly whatever is cheapest?

 

[infinitesymphony]

Quote:

Originally Posted by LazyAzN /img/forum/go_quote.gif You used some cheap 0.1 uF ceramic capacitors to stop the oscillation(google shows all ), right infintesymphony? Would that count as the feedback resistor?

No, I don't think so... But that was a separate suggestion made to me as well--incorporating a resistor in the feedback loop--though something tells me that adding one wouldn't have made a difference. Bypassing/decoupling seems like the best solution to me.

Quote:

Originally Posted by LazyAzN /img/forum/go_quote.gif Along with how does one cut a protoboard? Exacto knife?

Score with an exacto knife, suspend half over the edge of a table, and press down. *snap*

Quote:

Originally Posted by LazyAzN Is there any preference to the ceramic capacitors besides the 0.1 uF?

Yes... I did a lot of reading about decoupling capacitors. Everyone has different suggestions. The op-amp datasheets often recommended 0.1 uF ceramics, but others have suggested that polyester film capacitors would be a better choice. As with the input capacitors, values higher than 0.1 uF are recommended, but higher values increase size.

I took all of those factors into account, and wound up with the flat ceramics due to one main factor: space. They were the only type of capacitor that made it possible to mount the amp in a typical tin. If you do go with a ceramic, you can probably do better than the Radio Shack ones, which aren't well-matched or high quality--but they work! You might check out the Vishay/BC X7R radial ceramics (Digi-Key part number BC1084CT-ND may be a nice match).

 

[LazyAzN]

Quote:

Originally Posted by MisterX /img/forum/go_quote.gif Blue dots are the outputs when R5 is placed outside of the feedback loop. Pink dots are the outputs when R5 is placed inside the feedback loop. The problem with this setup is ideally R5 should be like 10 times smaller when it is inside the feedback loop (somewhere around 6 ohms) so you can't just change the wire and get....... The modded jumpers are just my way of doing it, it never made much sense to me to cut the leads off the resistors and then solder them back to the board when you could just bend the leads over, feed them through the hole, solder them and then clip the extra off to get the same thing. I also have a different way of connecting "the power rails" so maybe a photo is prudent? http://i23.photobucket.com/albums/b3...e/HPIM1417.jpg

So you soldered all over the place under the board huh... I might need more solder, but hey, it would look cleaner

So let me make sure I get this correct... When I don't place R5, the blue is where I should have my outputs... When I do, then the pink?...

6 ohms is kind of small isn't it

?

 

[MisterX]

Nevermind... I just edited my other post instead.

Quote:

When I don't place R5, the blue is where I should have my outputs... When I do, then the pink?...

It doesn't matter when there is no R5.
(you install a zero ohm jumper in that position when there is no R5 so the pink and blue dots represent the exact same position in the circuit)

 

[LazyAzN]

Quote:

Originally Posted by MisterX /img/forum/go_quote.gif Nevermind... I just edited my other post instead. It doesn't matter when there is no R5. (you install a zero ohm jumper in that position when there is no R5 so the pink and blue dots represent the exact same position in the circuit)

Alright, thanks for clearing up most of my questions

...
Now to order parts and hope the assembly goes well!
I'll post pics if it is

 

[error401]

Quote:

Originally Posted by infinitesymphony /img/forum/go_quote.gif Yes... I did a lot of reading about decoupling capacitors. Everyone has different suggestions. The op-amp datasheets often recommended 0.1 uF ceramics, but others have suggested that polyester film capacitors would be a better choice. As with the input capacitors, values higher than 0.1 uF are recommended, but higher values increase size. I took all of those factors into account, and wound up with the flat ceramics due to one main factor: space. They were the only type of capacitor that made it possible to mount the amp in a typical tin. If you do go with a ceramic, you can probably do better than the Radio Shack ones, which aren't well-matched or high quality--but they work! You might check out the Vishay/BC X7R radial ceramics (Digi-Key part number BC1084CT-ND may be a nice match).

From a physics perspective, usually recommended is a 10nF to 0.1uF ceramic depending on who you talk to. If anyone recommends larger values, they are almost invariably in addition to the standard decoupling ceramic. Film caps should generally not be used as they have much higher ESL, and ESR is usually not much better than ceramics.

Your recommendation is a good part for this purpose

.

 

[infinitesymphony]

Quote:

Originally Posted by error401 /img/forum/go_quote.gif From a physics perspective, usually recommended is a 10nF to 0.1uF ceramic depending on who you talk to. If anyone recommends larger values, they are almost invariably in addition to the standard decoupling ceramic. Film caps should generally not be used as they have much higher ESL, and ESR is usually not much better than ceramics.

Very interesting--thanks for clearing me up! I hadn't heard that 0.1 uF was the typical maximum limit for decoupling, or maybe I had my values confused with those of the input capacitors (0.1 uF minimum, higher is better to avoid low bass roll-off).

I'd heard that film capacitors were better than ceramics for decoupling due to higher linearity in the audio range, whereas ceramic capacitors had better high-frequency linearity, but I might be misremembering what I've read. Either way, it's nice to know that ceramic isn't a bad choice.

Just curious... When would it be recommended to use additional decoupling capacitors?