i would be interested if someone's making a run on the pcb production. do PM me if you're planning to do this, earfanatic
indeed he is, but hes also a very busy guy. theres a bit of interest over at headfi too, something to do with my spruking perhaps haha. i think there would be even more interest this time around. afaik though Owen was looking at doing a fully balanced build (output too) if there was another run.
New Run - The Wire in two Variants
Hi Guys,
I've been offline for the last little while working on a separate project, but I've got a little free time, and seeing as I've been more or less inundated with requests for more PCBs and kits, I think it's time for another run of "The Wire"
On that note, I'm looking to see how much interest there is for this so I can figure out how many to order. Please post up the following:
1. Do you prefer a board only, or a full kit?
2. Would you rather have a PSU included or not?
3. Would you be interested in a fully balanced version (bridged output with two LME49990) or the existing version?
4. How many units are you interested in?
Since I'm re-spinning the boards, I can make certain changes like omitting the PSU altogether. I'll make the decision based on what the majority of people are interested in. If there is a lot of interest for the fully balanced version, then I might do a run of both.
Please spread the word, and let me know who is interested in what!
Cheers,
Owen
Hi Guys,
I've been offline for the last little while working on a separate project, but I've got a little free time, and seeing as I've been more or less inundated with requests for more PCBs and kits, I think it's time for another run of "The Wire"
On that note, I'm looking to see how much interest there is for this so I can figure out how many to order. Please post up the following:
1. Do you prefer a board only, or a full kit?
2. Would you rather have a PSU included or not?
3. Would you be interested in a fully balanced version (bridged output with two LME49990) or the existing version?
4. How many units are you interested in?
Since I'm re-spinning the boards, I can make certain changes like omitting the PSU altogether. I'll make the decision based on what the majority of people are interested in. If there is a lot of interest for the fully balanced version, then I might do a run of both.
Please spread the word, and let me know who is interested in what!
Cheers,
Owen
Hello opc:
Wonderful news - very exciting!
1. Prefer full kit.
2. Prefer PSU included.
3. Existing version is fine (re-run of "The Wire" board).
4. Two units (2x amp + 2x PSU).
Thanks so much for all your efforts,
mike-sp
Wonderful news - very exciting!
1. Prefer full kit.
2. Prefer PSU included.
3. Existing version is fine (re-run of "The Wire" board).
4. Two units (2x amp + 2x PSU).
Thanks so much for all your efforts,
mike-sp
That is great news, opc!
1. Just the board
2. No psu
3. Prefer the existing SE out, but I would buy it either way
4. One board
And speaking of changes to consider... one thing that came out of the LME49600 amp thread in this section was that the BUF634 and LME49600 have similar packages available and the same pinout. I'm a fan of multi-use circuit boards - maybe the design could allow the use of either? In fact, it looks like the BUF634 might just drop in the board the way it sits. I'm a LME49600 fan, but from that thread it sounds like some folks prefer the BUF634.
From that thread I will also go ahead and concede defeat on having the option of two LME49600's in parallel on each output. 🙂 Doesn't sound like many folks would have interest in that. Plus I found an article about paralleling multiple BUF634's that calculates some considerable "balancing" currents that start moving between the multiple buffers.
1. Just the board
2. No psu
3. Prefer the existing SE out, but I would buy it either way
4. One board
And speaking of changes to consider... one thing that came out of the LME49600 amp thread in this section was that the BUF634 and LME49600 have similar packages available and the same pinout. I'm a fan of multi-use circuit boards - maybe the design could allow the use of either? In fact, it looks like the BUF634 might just drop in the board the way it sits. I'm a LME49600 fan, but from that thread it sounds like some folks prefer the BUF634.
From that thread I will also go ahead and concede defeat on having the option of two LME49600's in parallel on each output. 🙂 Doesn't sound like many folks would have interest in that. Plus I found an article about paralleling multiple BUF634's that calculates some considerable "balancing" currents that start moving between the multiple buffers.
Last edited:
Hello, this is my first post. I'm a member over at Head-Fi and a thread there about "The Wire" got me interested. I would love to try this, it's been a while since I last DIY'd and it's about time I got back into it! 😀 I'm interested in...
1. Full Kit
2. PSU Included
3. Existing Version
4. 1 Unit
1. Full Kit
2. PSU Included
3. Existing Version
4. 1 Unit
1. Full kit or board only, either way.
2. PSU in or out, doesn't matter.
3. Fully balanced version.
4. Two units.
Hi Owen,
For fully balanced, are you thinking about OPA1632(THS4131) + 2x LME49600 ?
Thanks,
2. PSU in or out, doesn't matter.
3. Fully balanced version.
4. Two units.
Hi Owen,
For fully balanced, are you thinking about OPA1632(THS4131) + 2x LME49600 ?
Thanks,
1. If you have the importunity to offer a kit, you should, but also while keeping the possibility to buy the boards separate. The reasons for creating a kit form are self explanatory. And for the sake of freedom (or stockpiling), just the boards should be offered separately too.
2. Is there any interest in improving the PSU? Anyone could build a simple LM3xx supply on a perfboard. For that reason, exclude the power supply unless it's improved to a super regulator or a shut reg (or just something special).
3. Having both balanced and single ended output could be possible on the same board, just one is option over the other. That way both party's would be happy
4. If I had money left over from my last project, I'd probably be interested in a couple amp boards with a upgraded supply, or just the amp boards if there wasn't a PSU. It's a shame I didn't get in on this when I saw this thread after it just stared, I'd probably be listening to "the wire" right now 😛. Oh well I ain't.
Cheers,
Kurt
2. Is there any interest in improving the PSU? Anyone could build a simple LM3xx supply on a perfboard. For that reason, exclude the power supply unless it's improved to a super regulator or a shut reg (or just something special).
3. Having both balanced and single ended output could be possible on the same board, just one is option over the other. That way both party's would be happy
4. If I had money left over from my last project, I'd probably be interested in a couple amp boards with a upgraded supply, or just the amp boards if there wasn't a PSU. It's a shame I didn't get in on this when I saw this thread after it just stared, I'd probably be listening to "the wire" right now 😛. Oh well I ain't.
Cheers,
Kurt
Hello
It would be A full kit for me and the possibility of having some nude boards to work on them.
It would be A full kit for me and the possibility of having some nude boards to work on them.
1. PCBs only
2. No PSU
3. Prefer fully balanced version (SE output optional).
4. 4 board
opc, if time permit, can you consider to modify the PCB to support the use of two LME49600's in parallel on each output? I think that it would be great for builders to have the option to use one LME49600 or two LME49600's in parallel (depending on preferences, applications, budgets…). Thanks for the efforts!
2. No PSU
3. Prefer fully balanced version (SE output optional).
4. 4 board
opc, if time permit, can you consider to modify the PCB to support the use of two LME49600's in parallel on each output? I think that it would be great for builders to have the option to use one LME49600 or two LME49600's in parallel (depending on preferences, applications, budgets…). Thanks for the efforts!
Preferably Kit for me, so I don't have to spend a lot of time sourcing for parts. But whatever you guys come up with, I will be interested in ordering 2x.
Thanks opc for opening 2nd batch group buy.
Thanks opc for opening 2nd batch group buy.
chip and resistor P & A update
To save opc some time... 🙂
LME49600TS-ND
Digikey [646 in stock 5/20/11]: (1) $10.50 (25) $8.61 (100) $7.77
Mouser: not carried
Future Electronics [132 in stock 5/20/11]: (1) $4.94 (10) $4.87 (50) $4.69
Newark [204 in stock 5/20/11]: (1) $9.79 (10) $8.03 (50) $7.23
Farnell: [129 in stock 5/20/11]: (1) 8.05 (10) 6.47 (100) 5.47 [in pounds]
Avent Express: no stock
Arrow NAC: no stock
Allied: not stocked
LME49990MA-D
Digikey [624 in stock 5/20/11]: (1) $5.18 (25) $4.16 (100) $3.79
Mouser: not carried
Future Electronics [218 in stock 5/20/11]: (1) $2.18 (10) $2.16 (25) $2.13
Newark [175 in stock 5/20/11]: (1) $4.81 (25) $3.87 (100) $3.54
Farnell: [175 in stock 5/20/11]: (1) 3.72 (25) 2.99 (100) 2.74 [in pounds]
Avent Express: [95 in stock 5/20/11]: (1) $2.85 (25) $2.72
Arrow NAC: [150 in stock 5/20/11]: (1) $2.91 (25) $2.72
Allied: not stocked
10K 0.1% 0805 RG2012P-103-B-T5
Digikey [282,000 in stock 5/20/11]: [10 piece minimum] (10) $0.41 (50) $0.248 (200) $0.205
Mouser: [20,000 in stock 5/20/11]: (1) $0.41 (10) $0.28 (100) $0.20
Future Electronics: no stock
Newark: no stock
Farnell: no stock
1K 0.1% 0805 RG2012P-102-B-T5
Digikey [156,000 in stock 5/20/11]: [10 piece minimum] (10) $0.41 (50) $0.248 (200) $0.205
Mouser: [13,000 in stock 5/20/11]: (1) $0.41 (10) $0.28 (100) $0.20
240R 0.1% 0805 RG2012P-241-B-T5
Digikey [23,884 in stock 5/20/11]: [10 piece minimum] (10) $0.41 (50) $0.248 (200) $0.205
Mouser: [4,900 in stock 5/20/11]: (1) $0.41 (10) $0.28 (100) $0.20
To save opc some time... 🙂
LME49600TS-ND
Digikey [646 in stock 5/20/11]: (1) $10.50 (25) $8.61 (100) $7.77
Mouser: not carried
Future Electronics [132 in stock 5/20/11]: (1) $4.94 (10) $4.87 (50) $4.69
Newark [204 in stock 5/20/11]: (1) $9.79 (10) $8.03 (50) $7.23
Farnell: [129 in stock 5/20/11]: (1) 8.05 (10) 6.47 (100) 5.47 [in pounds]
Avent Express: no stock
Arrow NAC: no stock
Allied: not stocked
LME49990MA-D
Digikey [624 in stock 5/20/11]: (1) $5.18 (25) $4.16 (100) $3.79
Mouser: not carried
Future Electronics [218 in stock 5/20/11]: (1) $2.18 (10) $2.16 (25) $2.13
Newark [175 in stock 5/20/11]: (1) $4.81 (25) $3.87 (100) $3.54
Farnell: [175 in stock 5/20/11]: (1) 3.72 (25) 2.99 (100) 2.74 [in pounds]
Avent Express: [95 in stock 5/20/11]: (1) $2.85 (25) $2.72
Arrow NAC: [150 in stock 5/20/11]: (1) $2.91 (25) $2.72
Allied: not stocked
10K 0.1% 0805 RG2012P-103-B-T5
Digikey [282,000 in stock 5/20/11]: [10 piece minimum] (10) $0.41 (50) $0.248 (200) $0.205
Mouser: [20,000 in stock 5/20/11]: (1) $0.41 (10) $0.28 (100) $0.20
Future Electronics: no stock
Newark: no stock
Farnell: no stock
1K 0.1% 0805 RG2012P-102-B-T5
Digikey [156,000 in stock 5/20/11]: [10 piece minimum] (10) $0.41 (50) $0.248 (200) $0.205
Mouser: [13,000 in stock 5/20/11]: (1) $0.41 (10) $0.28 (100) $0.20
240R 0.1% 0805 RG2012P-241-B-T5
Digikey [23,884 in stock 5/20/11]: [10 piece minimum] (10) $0.41 (50) $0.248 (200) $0.205
Mouser: [4,900 in stock 5/20/11]: (1) $0.41 (10) $0.28 (100) $0.20
Last edited:
excellent Owen, i'll spread the word, i've already been spruking it pretty hard as i just absolutely love this little amp!!
if doing a redesign, i would perhaps put forward the possibility of a simple bal/se convertor before the input that can be populated or not.
i know it wont gain any performance for the average user and i dont need it, but i have fielded a lot of questions about using the wire with a se input and i have always had to say that i think they would be better placed getting something else more suitable if they dont have a balanced source. if the opa1632 or ths4131 was used it wouldnt be needed, just alternate feedback arrangements.
on that note, i would love a fully balanced in/out version with lme49990 and 2 buffers a channel;
at which point also paralleling buffers for a total of 8 pieces, or 2A becomes lunacy imo and it will get hot hot hot and make a pretty demanding load for a psu. just as it is, with the opamps and 4 x buffers is already pretty demanding and the average super reg wont cut it for current delivery.
then as mentioned above you will probably have some pretty serious current sharing issues unless you also put patterns for current sharing resistors.
i have used it with several psu'd and havent actually felt there was much point as the psrr is so massive and i love psu tweaking so would love an excuse.
I would like the reservoir caps to be possibly changed to a few 7343 size special polymer caps such as PCE3550CT-ND panasonic a few in parallel for a little ultra lowZ supercap . this was the only part that stopped the board from being ultralow profile. they come in higher voltage range
i would also like the possibility of powering the buffers from a separate supply, as they will take up to +/-22vdc (a better option than paralleling them imo),
also leave wideband mode optional.
as far as the psu inputs, perhaps instead of the power inputs in the center, leave them there, but please make it possible to use stacking headers.
i think if you are going to upgrade the psu, you should go all out and make it an onboard unit with sensing etc, otherwise you are compromising performance right there just with the interconnecting distance and the caps will do all the work.
4 layer board
i think thats it haha, sorry, i LOVE it how it is and just going balanced would be great, but these are some ideas that wopuld make it even more interesting to me, i'm already toying with a mod to supply the buffers separately with 22v on the current boards
so count me in for blank boards and perhaps semis and npo caps, only a power supply if its full on and onboard, otherwise without
if doing a redesign, i would perhaps put forward the possibility of a simple bal/se convertor before the input that can be populated or not.
i know it wont gain any performance for the average user and i dont need it, but i have fielded a lot of questions about using the wire with a se input and i have always had to say that i think they would be better placed getting something else more suitable if they dont have a balanced source. if the opa1632 or ths4131 was used it wouldnt be needed, just alternate feedback arrangements.
on that note, i would love a fully balanced in/out version with lme49990 and 2 buffers a channel;
at which point also paralleling buffers for a total of 8 pieces, or 2A becomes lunacy imo and it will get hot hot hot and make a pretty demanding load for a psu. just as it is, with the opamps and 4 x buffers is already pretty demanding and the average super reg wont cut it for current delivery.
then as mentioned above you will probably have some pretty serious current sharing issues unless you also put patterns for current sharing resistors.
i have used it with several psu'd and havent actually felt there was much point as the psrr is so massive and i love psu tweaking so would love an excuse.
I would like the reservoir caps to be possibly changed to a few 7343 size special polymer caps such as PCE3550CT-ND panasonic a few in parallel for a little ultra lowZ supercap . this was the only part that stopped the board from being ultralow profile. they come in higher voltage range
i would also like the possibility of powering the buffers from a separate supply, as they will take up to +/-22vdc (a better option than paralleling them imo),
also leave wideband mode optional.
as far as the psu inputs, perhaps instead of the power inputs in the center, leave them there, but please make it possible to use stacking headers.
i think if you are going to upgrade the psu, you should go all out and make it an onboard unit with sensing etc, otherwise you are compromising performance right there just with the interconnecting distance and the caps will do all the work.
4 layer board
i think thats it haha, sorry, i LOVE it how it is and just going balanced would be great, but these are some ideas that wopuld make it even more interesting to me, i'm already toying with a mod to supply the buffers separately with 22v on the current boards
so count me in for blank boards and perhaps semis and npo caps, only a power supply if its full on and onboard, otherwise without
Last edited:
Hi Guys,
It's great to see so much interest so quickly! Lots of good suggestions, all of which I will have to consider. I'll see how much work and risk a fully balanced version is going to be, and I'll also consider taking a closer look at the PSU. Putting it right on board would generally make more sense, but there are always pros and cons.
Anyhow, I'll compile a new list with everyone on it, and what each is interested in. when I get a few more people on there, I'll start making up my mind on what exactly I'm going to run.
Cheers,
Owen
It's great to see so much interest so quickly! Lots of good suggestions, all of which I will have to consider. I'll see how much work and risk a fully balanced version is going to be, and I'll also consider taking a closer look at the PSU. Putting it right on board would generally make more sense, but there are always pros and cons.
Anyhow, I'll compile a new list with everyone on it, and what each is interested in. when I get a few more people on there, I'll start making up my mind on what exactly I'm going to run.
Cheers,
Owen
oh yeah perhaps consider proper optional placement of the volume pot inside the gain stage feedback loop for those who want to make it standalone or dont have a sabre or similar volume controlled dac
ie. optional gain and in the case of the balanced version, do so with a single resistor placement in between the input fb loops. this means cheaper 2 layer can be used in place of 4 gang pots and no resistor matching needed other than L/R channel matching, not quite as vital as phase
ie. optional gain and in the case of the balanced version, do so with a single resistor placement in between the input fb loops. this means cheaper 2 layer can be used in place of 4 gang pots and no resistor matching needed other than L/R channel matching, not quite as vital as phase
Last edited:
i should clarify for those that might consider the mod to supply the buffers with a higher +/-22v power supply, that driving very low impedance loads may mean the heatsinks for the buffers are not sufficient, perhaps opc can comment there as i dont know the exact spec of the pcb. probably it applies to a degree to all loads as the buffers operate in class A