Head-Fi's Sponsors
(Premier Sponsors bolded) |
|
|
Can Jam '09
(2009
International
Head-Fi Meet)
Impressions,
Reviews, Photos |

Can
Jam '09 graphic
courtesy of Edwood
Click on the links below
for Can Jam '09 photos,
impressions and reviews:
blubliss
1,
2,
3
dallan
1,
2,
3,
4
santacore
1,
2
nhat_thanh
1, 2
vpivinylspinner
1,
2,
3
amb
1
augustwest
1
eaglejo
1
johnsonad
1
shellylh
1
Jon L
1,
2,
3,
4
Germancub
1
zippy2001
1
IPodPJ
1
bhd812
1
Edwood
1,
2
abellaw
1,
2
minidiscs
1
atothex
1
HighLife
1
achristilaw
1
SiBurning
1,
2,
3,
4
SiBurning
5,
6,
7
LFF
1
Iron_Dreamer
1
doping panda
1
morphsci
1
ironbut
1
shaizada
1
jasper994
1,
2
jp11801
1
Uncle Erik
1
drubrew
1
(More impressions/photos
still being added.)
|
|
|
Head-Fi Blogs
and Facebook |
|
|
Head-Fi's Sponsors
(Premier Sponsors bolded) |
|
|
|

09-08-2007, 12:36 AM
|
|
100+ Head-Fi'er
|
|
Join Date: Jun 2007
Posts: 208
|
|
A few PIMETA build questions (New: troubleshooting questions)
i'm putting together a BOM for my first PIMETA with a TREAD power supply, and i want to make sure i order the right parts and connect everything properly, especially since this will be my first experience with building a separate power supply. as such, i have a few questions for those who have experience with these two projects.
1) per other members' recommendations, i've decided to use single LMH6321 buffers for R/L/G in lieu of single/stacked BUF634s. based on all the info i can find in the forum archives, it seems that all i need to do to adapt the 6321 to be BUF634 pin-compatible is to omit R11 and clip pin 1. is this correct? i think i read something about adding an input resistor as well, but i can't remember where i read it.
2) will this adapter work well with the TREAD board, regulated down to 24V to supply the PIMETA? should i use a similar 24V adapter instead? will the heatsink recommended by tangent work with this configuration?
3) i'd like to use a Hammond enclosure rather than the recommended Serpac H-65/67. i want a case that will accomodate both the TREAD and PIMETA boards. any recommendations from people who have done the same thing?
4) tangent suggests placing a diode in-line with one of the power connections, rather than in the crowbar configuration, if the PIMETA will be used exclusively with wall power. he says that this allows the use of a much smaller diode. how much smaller are we talking? i picked out this one from the same line of Vishay Schottkys as the crowbar diode he recommends - is it appropriate, or would a different type/value work better?
5) what is the big advantage of the ALPS RK27 over the 097 or the Pana EVJ? just better tracking? smoother movement? since this will be a non-portable PIMETA, i'm not concerned about the space it occupies. in everyone's opinion,is it worth the extra $12-13?
6) if anyone wants to just take a quick look at my BOM and see if there are any glaring mistakes, i'd be very appreciative.
thanks for any help you can give!
|

09-08-2007, 06:46 AM
|
 |
1000+ Head-Fi'er
|
|
Join Date: Aug 2006
Posts: 1,017
|
|
1)
Quote:
Originally Posted by ATAT
Warning for those using LMH6321 - The LMH6321 is a great buffer, can output nearly .7A of current and its ridiculously fast.. one problem. Pin 2 is used as current control. Before you solder the LMH6321, cut off Pin #2 (or bend it upwards). Pin 2 is connected to output on the schematic (so that opamps can be used as buffers) but this will prevent sane operation of the LMH6321.
|
It's from the PPAS build thread Good Looking Full size phones
6) Are you sure you want a gain of 11? Of course it depends on what phones you use. But even with high impedance phones, this gain is unnecessary high.
Are you sure you want input caps? I don't think you need them. They will cost you money and degrade the sound.
While you're at it, read the forums here for opamp reviews. Buy some of the opamps (not the very expensive ones) you find promising and do some rolling. The opamps have the major influence on how the amp is going to sound. Nothing comes close - choice of power supply, resistor or capacitor brand etc. LMH6321 is excellent. AD8620 is very good but more a matter of personal preferences.
Examples of good opamps are AD8599 (my new favourite) neutral, detailed, still warm mids, less grain than other Analog opamps. AD8066 (dual, the single AD8065 is not unity gain stable and can't be used in PIMETA) detailed, "fast" sounding", good sound stage, a bit grainy but the main flaw is a bit of a hollow mid just like... LM4562 (= LME49720 dual, LME49710 single) very sweet sounding and detailed, open, absolutely no grain but something is lacking in the mid/lower mid just like... AD8058 clean, no grain, detailed and fast but lacking warmth. AD744 "comp pin out" (single, for ground channel) neutral, detailed, a little bit grainy and harsh treble (like most of the Analog opamps), AD825 (single), haven't heard it yet 'cause I can't find my BrownDog adapters. There's OPA2107/627 if you want a darker/fuller sound.
Like I said, do some searching and reading here on HeadFi. The above is just what I think and shuldn't be taken for the truth.
|

09-08-2007, 08:04 AM
|
|
100+ Head-Fi'er
|
|
Join Date: Jun 2007
Posts: 208
|
|
interesting - this is from a thread entitled "Tweak Output Stage of PIMETA":
Quote:
Originally Posted by quicksilver96
I also got rid of my BUF634s though I didn't use the HA500X series. The National LMH6321 is currently my prefered IC buffer solution and is pin compatible with the BUF634. Just cut off or lift pin 1. I believe this is connected to V- through the BW resistor the PIMETA board, but you want this pin left open or tied to ground when the error flag function is not in use. The LMH only comes in PSOP-8 so you can use a socket on the PIMETA and a browndog SOIC->DIP adapter so you can swap/compare with the BUF or other ouput stages; or you can just solder the LMH to the SOIC pads on the PIMETA.
|
there seems to be some discrepancy as to which pin(s) must be clipped/bent for proper function of the LMH6321. any idea what the deal is with that? it sounds like there may be reasons to clip both pins 1 AND 2.
as far as the gain, i'm glad you mentioned that - i took the stock resistor values from the schematics and completely forgot to tweak the gain. i'll probably build the amp with a gain of 5 or 6, and leave room for tweaking by installing the resistors in a DIP socket.
the input caps i was planning on ordering just to have them around, in case i need them for some particularly high-offset source in the future.
thanks for the advice, and for all of the opamp suggestions. i do plan on rolling a bunch of chips through the PIMETA, but i figure i'll give my wallet a breather in between the purchase of the main parts and the purchase of all the other candidate opamps and tweak parts. the 8620/8610 combo sounds like one of the better performers, so i figure it's a good place to start.
EDIT: if i'm reading the data sheet and other people's posts right, it looks as though pins 1 (error flag), 2 (current limit), and 5 (gnd) can and should all be clipped when using the LMH6321 in place of a BUF634.
|

09-08-2007, 08:59 AM
|
 |
Contributor  Headphoneus Supremus: Top Mall-Fi poster. The "T" in META42. Member of the Trade
|
|
Join Date: Sep 2001
Location: Aztec, NM
Posts: 5,251
|
|
Quote:
Originally Posted by xmokshax
it seems that all i need to do to adapt the 6321 to be BUF634 pin-compatible is to omit R11 and clip pin 1.
|
I haven't tried this chip, but as far as I can tell, you can do just one or the other: you would only clip pin 1 if you had a PIMETA with R11 installed already and you didn't want to desolder it. For a new build, just leave R11 out, and it won't matter that these two chips use pin 1 for different things.
Quote:
|
i think i read something about adding an input resistor as well, but i can't remember where i read it.
|
Don't bother unless you experience problems. It'll be tricky to add.
Quote:
|
how much smaller are we talking? i picked out this one
|
It'll work, but there's no point in using a Schottky here. I'd just use a 1N4001.
__________________
"Give a man a fish and he eats for a day. Teach a man to fish and he knows what tackle to buy."
|

09-08-2007, 10:50 AM
|
 |
1000+ Head-Fi'er
|
|
Join Date: Oct 2006
Location: Vancouver, BC
Posts: 1,239
|
|
Quote:
Originally Posted by tangent
It'll work, but there's no point in using a Schottky here. I'd just use a 1N4001.
|
Crowbars are pretty crude indeed, but I'm thinking maybe a Schottky would be beneficial since it would likely last longer at short circuit (lower voltage drop = lower dissipation, no?) and push the fault to something intended as a fuse upstream in the power supply. Not sure if that's really better or not though, a $0.02 1N4001 failing and having to replace it is probably cheaper and almost as easy as replacing a fuse...
Hmm.
|

09-08-2007, 11:21 AM
|
|
100+ Head-Fi'er
|
|
Join Date: Jun 2007
Posts: 208
|
|
thanks for the input. i've switched from Schottky to 1N4001 - i guess in the (fairly unlikely) event of reverse voltage, i'll suck it up and pop in a new diode.
as far as the LMH6321 goes, has anyone had success using it as a buffer with pin 2 (current limiting) connected? it sounds like ATAT recommends against it for his PPAS boards, but i'm not entirely sure what the effect of connecting it would be.
also, any input on an ideal enclosure for PIMETA+TREAD, a good unregulated adapter for the TREAD@24V, or opinions re: RK27?
|

09-08-2007, 11:56 AM
|
 |
1000+ Head-Fi'er
|
|
Join Date: Oct 2006
Location: Vancouver, BC
Posts: 1,239
|
|
Quote:
Originally Posted by xmokshax
as far as the LMH6321 goes, has anyone had success using it as a buffer with pin 2 (current limiting) connected? it sounds like ATAT recommends against it for his PPAS boards, but i'm not entirely sure what the effect of connecting it would be.
|
I haven't done it, but it's a bad idea. The current limit of the chip is controlled by a current sourced through a resistor into that pin. Instead of seeing a static current here as designed, the buffer is going to see its own output - effectively 0R at changing voltage. From a quick look at the datasheet it looks like this would always keep it out of current limit, but there could be nonlinearities or pathological cases that cause the current limit to engage, or worse, oscillate. It's definitely not an intended use for the CL pin, so I would strongly recommend you not connect this pin.
|

09-08-2007, 03:21 PM
|
 |
Contributor  Headphoneus Supremus: Top Mall-Fi poster. The "T" in META42. Member of the Trade
|
|
Join Date: Sep 2001
Location: Aztec, NM
Posts: 5,251
|
|
Quote:
Originally Posted by error401
Crowbars are pretty crude indeed, but I'm thinking maybe a Schottky would be beneficial since it would likely last longer at short circuit
|
I think you're misunderstanding something here. The 1N4001 suggestion is for a series diode in the V+ line, not as a crowbar. Yes, Schottkys make better crowbars, which is why I recommend them, but we're not talking about crowbars here.
Quote:
|
Originally Posted by xmokshax
i guess in the (fairly unlikely) event of reverse voltage, i'll suck it up and pop in a new diode.
|
If you reverse the voltage, nothing will happen....which is the point.
__________________
"Give a man a fish and he eats for a day. Teach a man to fish and he knows what tackle to buy."
|

09-08-2007, 08:45 PM
|
 |
1000+ Head-Fi'er
|
|
Join Date: Dec 2003
Location: Seattle
Posts: 1,321
|
|
You might consider buying your AD8620 and AD8610 from BrownDog as they will be pre-mounted. They are durn easy to work with then: you can just drop them into the DIP sockets. And soldering DIP sockets is farrrrrr easier!
http://cimarrontechnology.com/index....ATS&Category=6
|

09-09-2007, 12:54 AM
|
|
100+ Head-Fi'er
|
|
Join Date: Jun 2007
Posts: 208
|
|
Quote:
Originally Posted by bhjazz
You might consider buying your AD8620 and AD8610 from BrownDog as they will be pre-mounted. They are durn easy to work with then: you can just drop them into the DIP sockets. And soldering DIP sockets is farrrrrr easier!
http://cimarrontechnology.com/index....ATS&Category=6
|
hmm, good to know... i didn't realize they sold pre-mounted 8620s and 8610s. however, i think i might circumvent the brown dogs entirely and just use Sijosae's SO-8 to DIP-8 technique (bending the end pins outward and soldering directly into a DIP socket). $0.20 for a DIP socket sure beats $2.39 for a brown dog adapter, and i don't see any reason why the two would behave differently from one another, assuming i can manage to bend the pins to the appropriate angles without snapping them off.
|

09-13-2007, 10:11 PM
|
|
100+ Head-Fi'er
|
|
Join Date: Jun 2007
Posts: 208
|
|
ok, another noobish question:
i've just realized that i don't know precisely how to mount my PCB(s) inside the case that i've ordered for my PIMETA. all the DIY projects i've done to this point haven't required me to mount the PCB to the bottom of the case. what's the best way to do it?
i see that hammond sells plastic, self-adhesive standoffs that can just be stuck to the bottom of the case. do those work well?
EDIT: another random one - anyone know of a nice panel-mount LED holder that will "frame" it on the front panel and hold it in place?
|

09-14-2007, 07:08 AM
|
 |
1000+ Head-Fi'er
|
|
Join Date: Dec 2003
Location: Seattle
Posts: 1,321
|
|
You can mount it any way you like! The right sized Hammond case will allow you to slide the board in, but if you prefer to get the power supply in the same box, you might want something larger. Mounting it on the bottom is fine. Just make sure there is enough space between the bottom of the board and the case (if it is metal) so you don't short out your new project! I had four screws and an handful of teflon tube washers. Voila!
|

09-14-2007, 07:36 AM
|
|
100+ Head-Fi'er
|
|
Join Date: Jun 2007
Posts: 208
|
|
Quote:
Originally Posted by bhjazz
You can mount it any way you like! The right sized Hammond case will allow you to slide the board in, but if you prefer to get the power supply in the same box, you might want something larger. Mounting it on the bottom is fine. Just make sure there is enough space between the bottom of the board and the case (if it is metal) so you don't short out your new project! I had four screws and an handful of teflon tube washers. Voila!
|
so you just used the tube washers as standoffs, drilled holes in the bottom of the case, and threw some nuts on the outside ends? these specifics are what i was hoping to get at. i guess i'd like to avoid drilling through the case if possible, but it sounds like that's probably the best and most secure option, right?
unfortunately, i believe the PIMETA board was designed with the Serpac H6x cases in mind, and there doesn't appear to be a Hammond case that fits the width of the board exactly. of course, i'm also mounting a TREAD within the same case, so that's kind of a moot point.
|

09-21-2007, 06:25 AM
|
|
100+ Head-Fi'er
|
|
Join Date: Jun 2007
Posts: 208
|
|
another quick question: i've received my LMH6321s from Digi-Key, and they're packaged in a bag that says the contents are both static- AND moisture sensitive. anyone know why these buffers are moisture sensitive, and whether this should be a worry during/after assembly of my PIMETA?
|

09-21-2007, 07:25 AM
|
|
Headphoneus Supremus
|
|
Join Date: Dec 2002
Location: 49009 (Kalamazoo, MI)
Posts: 3,362
|
|
Quote:
|
and whether this should be a worry during/after assembly of my PIMETA?
|
Static yes, moisture not so much unless you plan on reflow soldering them.
(just dont go crazy with the heat while you are soldering them)
http://www.jedec.org/download/search/jstd020c.pdf
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
All times are GMT. The time now is 08:22 PM.
|