Hi Eric,
The gain of "The Wire" can be set to whatever you want. It doesn't depend at all on the impedance of your headphones. You need to consider:
- Source output voltage
- Voltage sensitivity of the headphones
- Desired output levels
The Wire can be used with up to +/-18V rails, or up to +/-22V rails if you use a different op-amp in the front end and 610 buffers. That gives roughly 11VRMS output or 14VRMS respectively. I have never seen a pair of headphones require anywhere close to these voltages to achieve over 120dBSPL output. If you think you need more, then the BAL-BAL can swing 20VRMS but you need a differential input to your headphones.
Even headphones with high impedance (600 to 1000 ohms) generally only require a gain of 1 providing you have a source that can drive roughly 2VRMS. There are very few examples that need more voltage than that to achieve ear damaging output levels.
Regards,
Owen
The gain of "The Wire" can be set to whatever you want. It doesn't depend at all on the impedance of your headphones. You need to consider:
- Source output voltage
- Voltage sensitivity of the headphones
- Desired output levels
The Wire can be used with up to +/-18V rails, or up to +/-22V rails if you use a different op-amp in the front end and 610 buffers. That gives roughly 11VRMS output or 14VRMS respectively. I have never seen a pair of headphones require anywhere close to these voltages to achieve over 120dBSPL output. If you think you need more, then the BAL-BAL can swing 20VRMS but you need a differential input to your headphones.
Even headphones with high impedance (600 to 1000 ohms) generally only require a gain of 1 providing you have a source that can drive roughly 2VRMS. There are very few examples that need more voltage than that to achieve ear damaging output levels.
Regards,
Owen
Last edited:
Well, I just broke my soldering iron trying to wick the solder out of the holes while after removing the regulator yet again. I think I'm gonna get a 10 pack of the regulators and a few more PCBs soon. Am I the only one having so much trouble with the PSUs? Both amps I've built so far have worked perfectly from the first power up, but I think I haven't had a PSU for which I can say the same, even though they have less and easier to solder parts! I find it really hard to believe that static discharge could just kill a previously working regulator like that, and even harder that Mouser would have an average of 3 or 4 DOA regulators out of every 10 delivered (again, I did earth myself before soldering in the new one, but I guess I'll have to finally get one of those grounding wrist straps now along with a new soldering station...).
Hi Butizzle,
I've had two of them in constant duty (24/7) since before they were released, and never any problems.
I also use two others for lab testing and troubleshooting, so two of them have had the voltage levels changed at least half a dozen times, and both have been shorted and tripped over-current more than a few times. Never had a single problem with any of them. I have also not yet seem a faulty reg on my end, and I order them either directly from TI, or through mouser.
I'm starting to suspect there's another problem somewhere around your reg. My first guess is that it has to do with isolation, but it could also be another random faulty part that is causing stress on the reg.
It seems unlikely that it's ESD since I've been incredibly sloppy with mine and never had a problem.
The other day I was testing a new output board for the DCX2496, and one of the six channels was hissing badly and producing lower and distorted output. I immediately measured everything for open solder joints or shorts, and everything was perfect. I then replaced the op-amp, and still had the same problem. I suspected the DCX at that point, so I swapped it into another DCX, and still had the same problem.
I re-soldered every joint, and checked each cap for shorts. Still had the same problem. In a totally desperate and confused long shot, I replaced the two bypass ceramics and the bypass cap on VCOM, and low and behold, the problem was gone. Not sure how a faulty ceramic cap was causing the problem, especially when everything was measuring perfectly, but it had to have been one of the three caps.
I also baked these boards properly, so it wasn't overheating that caused the fault.
Sometimes things just don't work the way you expect, and to be honest, it might be time to just start over. I can send you a new free board to replace the old one, and maybe we can eliminate the problem.
Regards,
Owen
I've had two of them in constant duty (24/7) since before they were released, and never any problems.
I also use two others for lab testing and troubleshooting, so two of them have had the voltage levels changed at least half a dozen times, and both have been shorted and tripped over-current more than a few times. Never had a single problem with any of them. I have also not yet seem a faulty reg on my end, and I order them either directly from TI, or through mouser.
I'm starting to suspect there's another problem somewhere around your reg. My first guess is that it has to do with isolation, but it could also be another random faulty part that is causing stress on the reg.
It seems unlikely that it's ESD since I've been incredibly sloppy with mine and never had a problem.
The other day I was testing a new output board for the DCX2496, and one of the six channels was hissing badly and producing lower and distorted output. I immediately measured everything for open solder joints or shorts, and everything was perfect. I then replaced the op-amp, and still had the same problem. I suspected the DCX at that point, so I swapped it into another DCX, and still had the same problem.
I re-soldered every joint, and checked each cap for shorts. Still had the same problem. In a totally desperate and confused long shot, I replaced the two bypass ceramics and the bypass cap on VCOM, and low and behold, the problem was gone. Not sure how a faulty ceramic cap was causing the problem, especially when everything was measuring perfectly, but it had to have been one of the three caps.
I also baked these boards properly, so it wasn't overheating that caused the fault.
Sometimes things just don't work the way you expect, and to be honest, it might be time to just start over. I can send you a new free board to replace the old one, and maybe we can eliminate the problem.
Regards,
Owen
Thanks for the reply Owen!
The thing that's bugging me the most is that I've had these problems with 2 out of 3 PSU boards I've built now - the first one outright refused to work (I did get one rail to work after a regulator replacement but the other one remained dead), on the second one I had to replace a regulator first and now one rail has crapped out as described above (perhaps this was not the same one I replaced the regulator on, I can't tell which is which anymore).
I also took care to give the PCBs a good iso rinse after soldering and didn't heat the passives excessively when assembling the boards. So the only thing I can see is really the regulators being the problem, and I'm really wondering what else I could do to prevent any further waste of good PCBs and money.
On the other hand, there's the fact that the most recent failure happened after just moving the PSU around (and using only one rail for a while, but I doubt that would cause any problems with the other rail), which would indicate that something about my handling of the board might have caused the failure. There's absolutely no need to send me any free PCBs, I was planning on getting a few LME49830 amp PCBs anyway and I'm just gonna add some PSU PCBs to that - I'm mainly interested in managing to assemble them without any complications from now on. The fact that regulator replacements have at least partially fixed the problem nearly every time really makes me think that the chips are at fault, but again I can't see Mouser (or TI, in supplying Mouser) delivering several DOA parts that cost $7 each on several occasions - I've already ordered regulators twice with a few months in between the orders.
Thanks again for your help, and for making these awesome circuits available to the public - when the PSU was working, my Wire headphone amp sounded absolutely superb! The other one I've built is serving as a (rather useless due to its unity gain and being followed by a pot negating any possible impedance or volume benefits) preamp in a hum-ridden integrated amp of mine and has not made any problems either, although the ESP P101 power amp stage probably doesn't quite live up to The Wire's fidelity standards.
The thing that's bugging me the most is that I've had these problems with 2 out of 3 PSU boards I've built now - the first one outright refused to work (I did get one rail to work after a regulator replacement but the other one remained dead), on the second one I had to replace a regulator first and now one rail has crapped out as described above (perhaps this was not the same one I replaced the regulator on, I can't tell which is which anymore).
I also took care to give the PCBs a good iso rinse after soldering and didn't heat the passives excessively when assembling the boards. So the only thing I can see is really the regulators being the problem, and I'm really wondering what else I could do to prevent any further waste of good PCBs and money.
On the other hand, there's the fact that the most recent failure happened after just moving the PSU around (and using only one rail for a while, but I doubt that would cause any problems with the other rail), which would indicate that something about my handling of the board might have caused the failure. There's absolutely no need to send me any free PCBs, I was planning on getting a few LME49830 amp PCBs anyway and I'm just gonna add some PSU PCBs to that - I'm mainly interested in managing to assemble them without any complications from now on. The fact that regulator replacements have at least partially fixed the problem nearly every time really makes me think that the chips are at fault, but again I can't see Mouser (or TI, in supplying Mouser) delivering several DOA parts that cost $7 each on several occasions - I've already ordered regulators twice with a few months in between the orders.
Thanks again for your help, and for making these awesome circuits available to the public - when the PSU was working, my Wire headphone amp sounded absolutely superb! The other one I've built is serving as a (rather useless due to its unity gain and being followed by a pot negating any possible impedance or volume benefits) preamp in a hum-ridden integrated amp of mine and has not made any problems either, although the ESP P101 power amp stage probably doesn't quite live up to The Wire's fidelity standards.
Last edited:
Have any of you chaps compared these regs with the Jung / Didden super regs for effects on sound? I've tried neither.
And,
Headphone cable, have any of you chaps compared differently lengths of the same cable with a Wire headphone amp? I'm currently using a Bal-Bal with 2 metres of solid silver 0.7mm wire in PTFE, it was better for transient response and clarity than the stock HD800 cable but I'm still just not getting the bass slam.
1120 hours of about 24/7 music playing ago I changed the output op-amps of the soundcard over to the LME49990 which had less bass than the LME49710HA (TO99) they replaced. They were hopeless for the first couple hundred, or so, hours. No bass, harsh, dull, restrained, annoying, quickly fatiguing, unmusical as the subtlties were ampified and presented in front of the music. To my ears one of the worse op-amps I've heard. But a lot of people like them and that's why I've run them so long and they have improved a lot. They are clearer than the others which are laid back and less distinct at the top and muddled in their larger bass. The 49990 are presenting to me an imaging effect that is better defined and the foreground / background sound levels have come back into balance. The bass is well defined but it may still be a bit lighter, I've not put the previous chips back in to compare.
So, I've just come back from the cinema and the bass there was so good, not ear busting loud, but speed, quiet fast slam, well defined low frequency rumbling, and clear, easy, wafting out but almost visceral but my ears, not body, were sensing it so headphones should do that too?
I know that with my and friends big floor standing TL speakers that the shorter the speaker leads the better the bass transient speed, slam, definition, etc and we're down to 10 inches or what ever is do-able.
Does that apply with headphone cables too with headphones 300 to 600 Ohm load (in my case) with very small current and back emf?
Thanks 🙂
And,
Headphone cable, have any of you chaps compared differently lengths of the same cable with a Wire headphone amp? I'm currently using a Bal-Bal with 2 metres of solid silver 0.7mm wire in PTFE, it was better for transient response and clarity than the stock HD800 cable but I'm still just not getting the bass slam.
1120 hours of about 24/7 music playing ago I changed the output op-amps of the soundcard over to the LME49990 which had less bass than the LME49710HA (TO99) they replaced. They were hopeless for the first couple hundred, or so, hours. No bass, harsh, dull, restrained, annoying, quickly fatiguing, unmusical as the subtlties were ampified and presented in front of the music. To my ears one of the worse op-amps I've heard. But a lot of people like them and that's why I've run them so long and they have improved a lot. They are clearer than the others which are laid back and less distinct at the top and muddled in their larger bass. The 49990 are presenting to me an imaging effect that is better defined and the foreground / background sound levels have come back into balance. The bass is well defined but it may still be a bit lighter, I've not put the previous chips back in to compare.
So, I've just come back from the cinema and the bass there was so good, not ear busting loud, but speed, quiet fast slam, well defined low frequency rumbling, and clear, easy, wafting out but almost visceral but my ears, not body, were sensing it so headphones should do that too?
I know that with my and friends big floor standing TL speakers that the shorter the speaker leads the better the bass transient speed, slam, definition, etc and we're down to 10 inches or what ever is do-able.
Does that apply with headphone cables too with headphones 300 to 600 Ohm load (in my case) with very small current and back emf?
Thanks 🙂
Last edited:
I had similar experience with my first PSU. Turned on fine, worked for a bit, one rail failed, followed by the other rail rising, then failing. If i discharged the PSU, then restarted, the voltages would rise to about +0.7V and -0.7V.
In my case though, the problem was entirely my own doing. I didn't isolate the regulator tab from the heat-sink like I should have, using the recommended ceramic screws. The strange thing was that even though I fixed that particular problem after my first failure, follow-up attempts to replace components and regulators just kept burning up more regulators (6 in total).
Finally, I ordered new PSU board, and everything worked great. Been using it 24/7 for a few months without any problem (Just measured it now, and it's +15.00V/-15.01V, very nice). The lesson for me, of course, was to be more cautious about the tabs.
Thanks for the advice - unfortunately, I've already checked the isolation between regulator tab and heatsink. On the positive rail, there's no (measured) connection between the two and on the negative side, they're shorted - looking at the schematic and the datasheet tells me that's the way it's supposed to be, as the tab is connected internally to the regulator's GND pin (which goes to the ground plane on the negative side), and the mounting pads for the heatsinks are also connected to the ground plane. I'm using plastic shoulder washers and mica tabs for isolation.
Just a small update: I replaced my cheapo soldering station (costs $30 locally these days, this model is sold under a bunch of different names/brands - mine was called AP2) with a proper nice Weller WS81 and got an anti-static mat as well as one of those little grounding wristbands to go with it.
Equipped with all this stuff, I just soldered a new Wire PSU and this time, everything worked perfectly right from the beginning! I'm still not sure why all the other ones I built were acting up, but I'm certainly gonna take these precautions whenever I have to solder high-performance ICs from now on. Better safe than sorry I guess, and the 40 bucks for the mat and armband were probably a good investment. The Weller station is awesome anyway, but I don't think I have to tell anyone about that. I'm still amazed what a difference a good soldering station and good tips make.
Agreed. I've been using Weller's propane soldering irons for many years now.
Besides not having to fiddle with the cord, it has the advantage of being portable and reaching operating temperature in about 15 sec. The flexibility to use it as hot air torch for fine smd work, melting hot glue, closing heat shrinks, and lighting candles is a plus. It's a nice thing to have in the soldering toolkit.
But I digress from the thread topic. 😱
Last edited:
Just got done enclosing my "Wire" headphone amp/preamp - it sounds great (but I already knew that from running it unenclosed 😉
Here's some pics:
As you can see from the pictures, there are some unused RCA jacks and it's not quite complete yet - I'm going to put a primitive, relay-based input selector in there as well as a pupDAC for direct USB digital input.
I really like using oversized cases - stuffing everything in a small enclosure has caused me nothing but trouble so far and making this one in a 1U 19" rack enclosure has been a real eye-opener.
Here's some pics:
As you can see from the pictures, there are some unused RCA jacks and it's not quite complete yet - I'm going to put a primitive, relay-based input selector in there as well as a pupDAC for direct USB digital input.
I really like using oversized cases - stuffing everything in a small enclosure has caused me nothing but trouble so far and making this one in a 1U 19" rack enclosure has been a real eye-opener.
Hi,
I'm building a BAL-BAL amp and it's almost finished, and as a first time smd soldering project I must say this was easier than I thought (I'll never forget the day I unpacked the boards and components 😀). But still I have 2 questions (might sound stupid but didn't find the info):
1- do the heatsinks really need to be soldered or can it be glued ? I don't really know how hot it can get with PSU @15V (and 32ohms headphones) and I'm having trouble soldering them. If not, any tip regarding soldering (T°, specific way to go etc.) ?
2- what type of transformer must I use to run the PSU @15V out ?
Cheers
I'm building a BAL-BAL amp and it's almost finished, and as a first time smd soldering project I must say this was easier than I thought (I'll never forget the day I unpacked the boards and components 😀). But still I have 2 questions (might sound stupid but didn't find the info):
1- do the heatsinks really need to be soldered or can it be glued ? I don't really know how hot it can get with PSU @15V (and 32ohms headphones) and I'm having trouble soldering them. If not, any tip regarding soldering (T°, specific way to go etc.) ?
2- what type of transformer must I use to run the PSU @15V out ?
Cheers
Last edited:
What Wattage is your soldering iron?
Some solders run better than others. Maybe a flux pen would help?
Some solders run better than others. Maybe a flux pen would help?
Certainly not a hard requirement. You can just leave them without any solder, or you can glue them. It won't have any real impact on the thermal performance or anything else really.
It is a massive thermal load (as it was designed to be) so many irons will struggle to get everything hot enough. I have an 80W Weller WD 1 and it manages just fine, but smaller or lower power irons may not fair so well.
For +/-15V you'll want anywhere from dual 12VAC to dual 15VAC secondaries depending on the regulation of the transformer. It's probably best to lean towards the 15VAC variant to be safe. The supply caps are rated at 25V before the reg, so you'll land safely under that.
Something like this would be perfect:
AN-0215 - 25VA 15V Transformer - AnTek Products Corp
Regards,
Owen
Thanks to you 2 for the answers, I think I'll go the glue way 🙂
Regarding the transformer, I would love to be able to purchase Antek but unfortunately shipping and duties costs to France are big dealbreakers.
I'll ger back here as soon as I can put this all up together and test 🙂
Thanks for the support!
Regarding the transformer, I would love to be able to purchase Antek but unfortunately shipping and duties costs to France are big dealbreakers.
I'll ger back here as soon as I can put this all up together and test 🙂
Thanks for the support!
Since you're in Europe try toroidy transformers, made in Poland.
Hey Owen just in case youre not receiving my messages, just wondering if you've managed to ship the latest round of orders, including mine, yet? 🙂
Hey Owen just in case youre not receiving my messages, just wondering if you've managed to ship the latest round of orders, including mine, yet? 🙂
Hi,
I need SE-SE stereo board, 1 peace. Since I didn't get any kind of respond via PM, I have to ask here.
Thanks,
Tomislav
I need SE-SE stereo board, 1 peace. Since I didn't get any kind of respond via PM, I have to ask here.
Thanks,
Tomislav
You can also preheat the PCB using a hot air gun set to 250°C when soldering parts to ground planes and the power of your soldering iron is nut suffcient.
I assembled a headphone buffer that had two chips that needed soldering to the PCB to perform as a heatspreader.
I used a hot air gun (for stripping paint etc), standing on the bench pointing up in the air.
Supported the PCB horizontally a short distance above the hot air nozzle with the solder pad uppermost. Allowed a minute or two to preheat the PCB from below.
Laid the two chips on the solder pads and applied the soldering iron to the top side chip tab and the solder just flowed under.
Job done.
But I had overheated the bottom side.
The GRP had bubbled slightly, electrically it still worked.
More distance between gun and PCB should solve that.
I used a hot air gun (for stripping paint etc), standing on the bench pointing up in the air.
Supported the PCB horizontally a short distance above the hot air nozzle with the solder pad uppermost. Allowed a minute or two to preheat the PCB from below.
Laid the two chips on the solder pads and applied the soldering iron to the top side chip tab and the solder just flowed under.
Job done.
But I had overheated the bottom side.
The GRP had bubbled slightly, electrically it still worked.
More distance between gun and PCB should solve that.
Yeah, these transformers are really nice. I've used them in my The Wire. They are made especially for audio use. Quality is pretty good, they are sealed with epoxy and shielded with silicon steel. Talema 55111-P1S2 is also pretty good and should be easy to purchase in Europe.
http://sklep.toroidy.pl/en_US/p/TTSA0030-Transformer-AUDIO-TSA30VA-voltage-to-50V/307