SMD code nightmare!

[armdnrdy]

I'm in the midst of tracing the MXR Carbon Copy.

I'm working with one that I've owned for about four years to get preliminary component value information.

I have a non working CC on the way (Ebay) so that I may lift some of the components (capacitors) to take accurate readings.

For those of you that aren't familiar with this delay...it is a packed, double sided board consisting of mainly SMD components, save for the BBDs and a few through hole parts on the "hardware" side.

My goal is to work up a project that would include the inner trimpots (Modulation Speed, Width) being brought to the top of the enclosure as controls....and the small push switch that activates the modulation being changed to a foot switch.

With all of the delay circuits out there...why would I go through so much trouble to reverse this circuit?
I've built an AD-80, AD-900, and an AD-3205....the Carbon Copy sounds better than all of the former delays.
It has a very "clean" sounding delay that still retains the analog warmth.

Enough about the hows and the whys! 

I've just been thrown into the world of SMD codes! 

There are two 6 pin devices that measure out to 2.1mm X 1.2mm (SOT-363) that have the code AB2 on the top.

I haven't fully traced the circuit but...it appears that these devices are used as clock buffers. They are located in between the CD4047 and the BBD clock pins.
Since they are six pin devices..I would imagine that they are either dual buffers, dual BJTs, or dual mosfets...but...since the component designations are Q1 and Q3...I am leaning toward transistors of some type.

I've tried the "SMD Codebook" page...and various similar pages where you punch in the code and it gives you options according to package size and code. Nothing!

I have searched all over the net with the information that I have...and have drawn a blank. Does anyone more familiar with SMD know of a good resource for identifying components?

[UKToecutter]

Larry,

Image isn't great.

Are you SURE it's not A82?
And could it be a SOT23-6?

Andy

[cloudscapes]

if I had to guess, single cmos inverters.

[armdnrdy]

Here is a better shot.



It would make sense that cmos buffers or inverters would be used but the reasoning behind the "Q" designation instead of "U" designation has me thinking otherwise.

[armdnrdy]

And could it be a SOT23-6?

This device is smaller than the minimum size for a SOT23-6 package.

[UKToecutter]

This has me a bit 'flummoxed'.
The nearest I can find is an NXP PUMB30 that is marked as *B2 where the * designates its country of manufacture.
Unfortunately none of the countries are designated by the letter A.
Been searching a couple of hours.  I hate SMD!!!!!!!!!!

[armdnrdy]

This isn't the correct part but...it could be something like this:
http://www.datasheetarchive.com/dl/Datasheet-097/DSA00157398.pdf

This part is listed a being used for driver circuits, interface, and logic switching.

Or....it could be something like this:
http://www.nxp.com/documents/data_sheet/NX3008NBKS.pdf

This is listed as having the same uses.

[Transmogrifox]

With surface mount parts, you will be left guessing.  Some don't have enough information in the codes to identify exactly what the part is.  This is where you need to finish mapping out the circuitry around it to see if you can deduce what is going on with this IC based upon how it is connected to the rest of the world.  An oscilloscope would be a useful tool here.

I love working with surface mount components.  Makes prototyping go so much faster, and uses far less space on a perf board.

[akc1973]

I love working with surface mount components. 

I guess you're like those people who love running ultra-marathons, canoeing across the pacific, etc, etc. 

[armdnrdy]

With surface mount parts, you will be left guessing.  Some don't have enough information in the codes to identify exactly what the part is.  

Yes...after extensive research...I still don't know what component this is!
This makes it very difficult to repair or trace circuits.

The component in question is so small (SOT-363)...it has been difficult to hold three test leads in place on the pins while attempting to identify this part with a atmega328 tester. Imagine yourself holding three test leads in place on a 2mm X 1.2mm component under a lighted magnifier...and then having to press the test button!  

I have identified dual diodes with the cathodes connected (NPN transistor) across a few of the pins while using two test leads.

I believe that these components are dual NPN transistors...possibly used in a Darlington configuration...as clock buffers.

My sole purpose of identifying these parts....is to look up the data sheet, read the specs, and find a through hole replacement.
Since the "mystery" components are used as clock buffers...I don't believe that replacing them with some other form of buffer would alter the workings of this circuit.

I have to be honest...I am not very impressed with the SMD code system!
I have seen that parts of different package size can have the same code...so...you have to identify the package as well.
The same code can be a diode in one package, and a transistor in another package!
I have also seen information on components that give direction to refer to the data sheet for the manufacturers code.
Well....in my opinion...it would be difficult to refer to the data sheet when you don't know what part it is due to lack of information!

[Transmogrifox]

...it has been difficult to hold three test leads in place on the pins while attempting to identify this part with a atmega328 tester. Imagine yourself holding three test leads in place on a 2mm X 1.2mm component under a lighted magnifier...and then having to press the test button! 

I don't have to imagine it -- I've been there many times.  Usually I solder small wires to the SMT leads that I can wrap/twist around the test leads.  On the smaller chips this also can get tricky because you have to be careful not to solder bridge stuff.

As for AKC comment about marathon running, I would try to defend myself but I'm stuck -- I run marathons and mountain trail runs too, so maybe I am just sick self-inflictor of pain

[Ice-9]

With surface mount parts, you will be left guessing.  Some don't have enough information in the codes to identify exactly what the part is.  

Yes...after extensive research...I still don't know what component this is!
This makes it very difficult to repair or trace circuits.

The component in question is so small (SOT-363)...it has been difficult to hold three test leads in place on the pins while attempting to identify this part with a atmega328 tester. Imagine yourself holding three test leads in place on a 2mm X 1.2mm component under a lighted magnifier...and then having to press the test button!  

I have identified dual diodes with the cathodes connected (NPN transistor) across a few of the pins while using two test leads.

I believe that these components are dual NPN transistors...possibly used in a Darlington configuration...as clock buffers.

My sole purpose of identifying these parts....is to look up the data sheet, read the specs, and find a through hole replacement.
Since the "mystery" components are used as clock buffers...I don't believe that replacing them with some other form of buffer would alter the workings of this circuit.

I have to be honest...I am not very impressed with the SMD code system!
I have seen that parts of different package size can have the same code...so...you have to identify the package as well.
The same code can be a diode in one package, and a transistor in another package!
I have also seen information on components that give direction to refer to the data sheet for the manufacturers code.
Well....in my opinion...it would be difficult to refer to the data sheet when you don't know what part it is due to lack of information!

I had a PDF book for SMD which always helped me find 99% of SMD parts, I have spend quite a while searching my old HD's for it looks like after a few different HD formats/installs that I have lost it. I still have the SMD cookbook but that only finds a low % of components.  

Anyway you could try etching a small pcb to hold the component so that you can connect it to your Atmega328 tester or solder some knyar wire to each of the terminals for the same purpose, I will still try to locate this PDF I had which was the dogs at finding SMD parts.

[armdnrdy]

Thanks for the input guys!

Mick,

If you can find the PDF that would be great!

When the "non-op" Carbon Copy arrives...it is going to be sacrificed to the trace gods...so...I'll "rip" into that one. The CC that I'm doing the preliminary work on is one of my personal working delays.

[Processaurus]

Merely because they are labeled Q* makes me guess they are transistors or FET's, like 2 of the same in a package, or a complementary pair.  Traditionally IC's (even little guys, like a single inverter) are labeled with a U*.

[jk-fm]

This came up a few years ago, at that time they might have read AB7?

http://www.diystompboxes.com/smfforum/index.php?topic=77958.0

AB2 is a marking for a mid-range 2SD2425, but it doesn't look like it comes in a 6-lead package.

http://datasheet.eeworld.com.cn/pdf/NEC/34453_2SD2425.pdf

Anyways, it looks like it's either an NPN/PNP push-pull or a dual NPN totem-pole configuration. Maybe a resistor-equipped transistor.

[armdnrdy]

Mystery solved!

I received the "broken" Carbon Copy today. When I popped it open, I noticed that it's a slightly newer version than mine.

On the earlier versions, there was an issue with 3 switching transistors (bypass circuitry) getting destroyed by static.

MXR warrantied the problem and fixed the issue by tacking a diode across each transistor. The board of the "new" CC I just received has a designated place for those diodes.




In the newer version there is a different SMD code for the "AB2" component that I have not been able to identify.

The new code is K46. I was able to identify this code immediately!

All this is .....is a 3904/3906 complementary dual transistor! There are several manufactures...with MMDT3946, and CMKT3946 being a few. http://www.s-manuals.com/pdf/datasheet/m/m/mmdt3946_galaxy.pdf

So jk-fm...your assumption of a NPN/PNP push-pull was correct.

[UKToecutter]

Hallelujah!!!

[Ice-9]

Good result

[baklavametal]

woohoo
im relly looking forward to your results as carbon copy remains my favorite delay ever

[jubal81]

Wow, this is a really ambitious project. I'll be rooting for you.