gar2520 build thread

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

gar381

Well-known member
Joined
Jan 27, 2005
Messages
1,226
Location
Retired in (Amherst) Cleveland, Ohio
NOTICE>>>

NOTICE OF PART CHANGE 10/5/23

Had to happen sometime... The pnp BC560C has been out of production
for some time now and my current supply of NOS Fairchild BC560Cs
that I have been using for last 3 years is running very low. I will be using
the remainder of my supply for only for the matched pnp input pairs for the
garOA10 and gar918.


I knew this day was coming so I have auditioned many possibilities.
From now on I will only be using..

BC557C mouser # 637-BC557C in the following positions...

gar2520 Q6, Q9.
gar1731 Q3, Q4
garAM10 Q1,Q5,Q6,Q7
garOA10 Q2, Q3, Q5, Q7, Q9
gar918 Q3, Q4

Note that these are all on tape and will just drop in place to the silk
just like the original BC560C

Gary


NOTE TO FOLKS SOURCING THEIR OWN PARTS..1/27/2020

The BC560C is no longer being manufactured. If you can
Not source these, with lots of listening time with lots of different
PNP transistors, we have come up with suitable replacements.

gar2520 Q6 and Q9
gar1731 Q3 and Q4
garAM10 Q1, Q5, Q6, and Q7

(gar918 and the gar OA10 will have other recommended replacements)


FIRST CHOICE 10/5/23 Mouser part # 637-BC557C

First choice BC32740TA Mouser Part# 512-BC32740TA
Second choice BC557BF Mouser Part# 512-BC557BTF

Note that both of these transistors have the same pinout as the BC560C
So you should be able to just drop them in place using the silk on the pcb.

If you want to experiment with other PNPs, Make shure the Vceo max
Is at least 45v. hfe is not all that important for these PNPs but above 250
Will work fine. Just be aware of pinout.

All of my kits until future notice, BC560Cs will be included in the kits.


NOTE: DUE TO SUPPLY CHAIN, SOME KITS MAY HAVE THE DIODES LOOSE
PACKED WITH THE RESISTORS INSTEAD OF ON TAPE


There are several versions of both the gar2520 and the gar1731 PCBs

All the build manuals are available for download HERE:

http://capi-gear.com/catalog/support_docs.php#opamps
Make Shure the manual you are downloading is for the PCB version you are using.

Look on the PCB and it will say which version to down load.
ALSO... the BOMs in all the manuals reflect the new murata parts




The Build PDF manual is good but here a few notes that might make the kit
A little less confusing. ( I really need to add this to the Manual :-\)..

1) The little bag with the Millmax pins also contains Q1 and Q2.
2) The tape strip with 3 transistors are Q3, Q4, Q5.
OR in later kits it’s a small bag of 3 transistors labeled Q3, Q4, Q5
3) The tape strip with 2 transistors are Q6, Q9 (the BC560c)
4) The little bag of caps..
The tan ones are CD1, CD2.
The blue ones.. see page 7 of the assembly manual
for value and location.
5) Its a good idea to solder the millmax pins both top and bottom.
6) Please meter the resistors to determine value.
7) After assembly PLEASE double check for unwanted solder bridges
on the Bottom of the PCB. and make SHURE THAT on the top of the
PCB that no component lead wires are touching each other.
8) When inserting your new DOA into the millmax sockets of
a module, Please insert straight down and not at an angle.
DO NOT push down on the component parts but use the edges
of the DOAs PCB. Remove DOAs the same way. Pull straight up
from the edges of the PCB and not at an angle.
9) Never insert a DOA or a module with the power on.

10) HAVE FUN !! :) :) :)

GARY

Note.. The gar2520 PCB is manufactured on a v-scored panel. If you find any sharp edges or loose fibers on the edges
Please use a fine grit sandpaper on the edges to clean them up before assemblying the kit.

Note.. Murata has changed the marking on the Blue 10pf cap from 10D to 10J
These two parts are exaxctly the same. (D = +/- .5pf and J = 5%)
You may see either one of these in my kits.

Note: Dec. 7 2013

MURATA IS DISCONTUNING THE RPC SERIES OF CAPS USED IN THE gar2520 and gar1731!!

The RPE series is being replaced with the RCE series. I will be phasing these in as my RPE
Stock diminishes. I will be using either 50v or 100v based on availability. For you guys buying
Raw PCBs and sourcing your on parts here are some part numbers:


10p 50v.. RCE5C1H100J0A2H03B
10p 100v. . RCE5C2A100J0A2H03B

39p 50v.. RCE5C1H390J0A2H03B
39p 100v.. RCE5C2A390J0A2H03B

47p 50v.. RCE5C1H470J0A2H03B
47p 100v.. RCE5C2A470J0A2H03B

680p 50v.. RCE5C1H681J0A2H03B
680p 100v.. RCE5C2A681J0A2H03B

1000p 50v.. RCE5C1H102J0A2H03B
1000p 50v.. RCE5C2A102J0A2H03B

In a pinch the AVX SR15 series...
50-100v 5% with a lead spacing of 2.5mm (.1”)
Should work fine as I have tested these


Check out chung's excellent build pics on his web site...
http://studio939.blogspot.com/2012/08/gar2520-opamp-build.html
 
Last edited:
Soooo, first question: q1 and q2 should be matched. What kind of hfe should I look for? does it matter? why not? :D
 
Any matched pair of BC550C should work fine.  Just make SHURE that they are
BC550C NOT BC550B or BC550A.  hfe of 400+ will do it.  The min. hfe for the BC550C
is 420.  The last batch of 500 transistors THAT I checked only 4 of them were below hfe500 and they
were in the 490s.  hfe520-550 is where most of them fell.

GARY
 
It's 6:30AM here on the East coast. I just finished assembling one gar2520 kit. It looks fantastic. The build manual is great. Gary has figured out an easy sequence to stuff the board to avoid mechanical interference when soldering. Holy mackerel. These things are great. I have nine more kits sitting right here but I have to get to work now. I was going to post all sorts of dumb-ass questions to get Gary's blood pressure to rise (What's a diode? Why does it have a black band?). But I decided against that. I don't have time to test it right now. That will happen soon. All I can say is....... FANTASTIC. Thanks Gary and Jeff. DW.
 
PCBs arrived to Poland yesterday. That was the fastest thing I've ever get form US. Starting assembly!
 
Has anyone finished assembling their DOA yet?

I'm having a weird interference like background noise and bad audio
when using the last two gain settings (highest two) on my R312.
It's actually pretty loud and makes the signal unusable.
All the other lower gain settings sound fantastic however!


Is there a solution for this?

cheers,

mike


 
Hi Mike,

I did a huge measuring session with DOAs last week, including the gar2520.
didn't realize any noise / oscillation / whatever problem at high gain settings.
do you have any other DOA to check if the high gain setting on your R312 is working right?
 
hey silentarts,

I had a couple Fabio DOAs in the two R312s that I built and I just checked.
They do not have that problem. They just keep on rocking hard on highest gain setting.

Is it possible that my R312 is somewhat faulty?

cheers

mike
 
The highest gain setting is 66db including transformers.
I also noticed that depending on the mic that's hooked up,
the second last position will sound horrible as well as the last position.


 
I'm using a 3 Cinemag CM75101APC. I believe it's a 1:8 ratio.
I have the suggested 150K resistor across the secondary,
and no Zobel network as it is supposed to do fine without it.
 
If your 312 boards are working with your other opamps, I don't see how it can be anything but the 2520 that is faulty. And since Silent Arts has one working, I would consider it a proven design, so you MAY have put something in the wrong place.

I once had a similar issue with an opamp where I had switched the double diode with a single diode by accident. Tried swapping everything on the 312 board, but nothing. Then tried another opamp and it was fine so I concentrated on the opamp but it took me some time to see the darn diodes. I'm probably the only idiot that's done that, but it may be a good place to check if these boards have them (haven't gotten any yet, so no idea on the schemo/board design).   

Back to the gain question though, is everyone shooting for 60dB of gain with these(and other 2520s)? I mean, my input trafos are 1:8 and 1:10, and outputs are usually 1:2 or 1:3.  So if we take the least gains and add them up we get

input 1:8 = 18.1 dB
output 1:2 = 6 dB
opamp      = 40dB

This still gets 64.5 dB. 

with the input trafo primaries in series (1:4 ratio), you get 58dB total. 

Are you guys putting your input trafo primaries in series? Does this result in lower noise? What RL do you use in this case?


sorry for derailing, but hopefully this will help people decide how best to use with their new GAR381 opamps!
 
I've checked and rechecked all components on the DOA and can't find anything wrong with it.
Has no one else experienced this with their opamps ?
Silent arts says he only checked up to 60db.

I have a feeling it is an input impedance/load issue.
 
mikefatom said:
I have a feeling it is an input impedance/load issue.
me too

mikefatom said:
I'm using a 3 Cinemag CM75101APC. I believe it's a 1:8 ratio.
I have the suggested 150K resistor across the secondary,
and no Zobel network as it is supposed to do fine without it.
do you really mean 150K ???
try 15K if yes
 
Back
Top