Hi guys,
I have still not finished my LPUHPs as I have a problem with the LME49990. This guy has a constant output of about 5mV HF, which is then minimally amplified by the LME49610s.
My question is:
Is this output normal?
The power supply has an output of +-18,4V. The LME49990 has a range of max +-18V. Does the LME49990 have a problem?
I read a comment from Owen. He wrote that removing the LME49990 is counterproductive. Why? Or: What will happen, if I do so?
Many thanks,
Stammheim
I have still not finished my LPUHPs as I have a problem with the LME49990. This guy has a constant output of about 5mV HF, which is then minimally amplified by the LME49610s.
My question is:
Is this output normal?
The power supply has an output of +-18,4V. The LME49990 has a range of max +-18V. Does the LME49990 have a problem?
I read a comment from Owen. He wrote that removing the LME49990 is counterproductive. Why? Or: What will happen, if I do so?
Many thanks,
Stammheim
Howdy,
I've finished my boards and they both work. One is a bit bassier and the volume is noticeably nowhere as loud as the the other. No hum, buzzing or hissing.
.
On the likely chance that I have to replace one of the little SMD components, how hard is it to completely remove the soldier and is there a good chance that the components can be replaced and not damage the board.
thanks.
I've finished my boards and they both work. One is a bit bassier and the volume is noticeably nowhere as loud as the the other. No hum, buzzing or hissing.
.
On the likely chance that I have to replace one of the little SMD components, how hard is it to completely remove the soldier and is there a good chance that the components can be replaced and not damage the board.
thanks.
resistance
I took the amps out of their cases and measured resistance at each component just to see if there were any obvious breaks and I get 5K at the output and 19M at the quieter amp. So bummer. I rechecked all the resistors and checked the capacitors for soldering but I really can't find anything.
I can't seem to find any breaks and, now, I don't want to power up the amp. for fear of possible damage. (Total power up time in bits probably 5 minutes)
I took the amps out of their cases and measured resistance at each component just to see if there were any obvious breaks and I get 5K at the output and 19M at the quieter amp. So bummer. I rechecked all the resistors and checked the capacitors for soldering but I really can't find anything.
I can't seem to find any breaks and, now, I don't want to power up the amp. for fear of possible damage. (Total power up time in bits probably 5 minutes)
I never did any SMD soldering before OPC's projects. I have now successfully completed several. I have removed several SMD parts and replaced them again and again. I have slightly modified a few of OPC's projects to suit my needs. On my first SMD project I did have a few problems and all of them were due to poor soldering. Either no contact, even though it looked good or a short to an adjacent pin. A large magnifying glass is a must or a cell phone with a magnify option and something to hold the cell phone steady. A meter with sharp probes to check for shorts, opens or wrong resistor values is a necessity esp if you have a problem. I have poor SMD soldering skills and the soldering doesn't look very nice and neat. When trouble shooting remember the boards work! Many people have made them without problems. If yours doesn't work, something is wrong somewhere. Because of my old age and poor eyesight I measure every part after I solder it in for shorts or opens. Yes it takes extra time but I am not an expert, so therefore I must take extra time. For the amplifiers I soldered 1 buffer at a time then checked the board for proper operation. If one buffer is soldered incorrectly it is very possible that none of them will work correctly and the board won't work. It is also possible you may have fixed your problem then introduced another problem with the same or different symptoms. This happens more than you think, believe me.
Use resistors in series with the transformer between the transformer and the bridge rectifiers. 10-20 ohms should work. Unsolder the output pin from all but 1 of the buffer chips, does it work now? Unsolder all but one of the input pins, does it work now?
Use resistors in series with the transformer between the transformer and the bridge rectifiers. 10-20 ohms should work. Unsolder the output pin from all but 1 of the buffer chips, does it work now? Unsolder all but one of the input pins, does it work now?
Taking a reasonable resolution picture and zooming in is one of the best method I've found for identifying soldering issues.
Phew!
All eight buffers? I use this big magnifying glass that is mounted on a metal base. (radio shack) I have checked the resistors to see if they are resisting the amount of current they should be. (Yeah - I conjugated - been waiting all day to make that joke)
I have some 10 ohm power resistors for the bridge rectifiers. Is this an all or nothing sort of issue or should I be systematic with each buffer?
All eight buffers? I use this big magnifying glass that is mounted on a metal base. (radio shack) I have checked the resistors to see if they are resisting the amount of current they should be. (Yeah - I conjugated - been waiting all day to make that joke)
I have some 10 ohm power resistors for the bridge rectifiers. Is this an all or nothing sort of issue or should I be systematic with each buffer?
as you know all 8 buffers are in parallel, if one has some shorted pins then they all do. You can check the operation of the amp after each output/input pin is unsoldered. Careful lifting the pins as they can break. 1 buffer can drive an 8 ohm speaker.
we don't know what test equipment you have so trouble shooting techniques will have to vary with what you have.
If you have a DVM you have to know what it will read with different inputs. On the AC scale will it read AC only or AC + DC, on the DC scale will it read DC or DC + AC or a variation of DC + AC. If all else fails RTFM Read the **** manual as I used to say.
A sound card, with care, can be used as a crude oscilloscope. You can also destroy your sound card/pc with excessive input signals.
I also have a large illuminated lens, I also use an 8x lens for up close and personal stuff. Sometimes the 2 lens together. Connections at 2x or 3x can look ok untill you get to 10-20x. Some times it looks like a part is soldered but it isn't.
we don't know what test equipment you have so trouble shooting techniques will have to vary with what you have.
If you have a DVM you have to know what it will read with different inputs. On the AC scale will it read AC only or AC + DC, on the DC scale will it read DC or DC + AC or a variation of DC + AC. If all else fails RTFM Read the **** manual as I used to say.
A sound card, with care, can be used as a crude oscilloscope. You can also destroy your sound card/pc with excessive input signals.
I also have a large illuminated lens, I also use an 8x lens for up close and personal stuff. Sometimes the 2 lens together. Connections at 2x or 3x can look ok untill you get to 10-20x. Some times it looks like a part is soldered but it isn't.
My best bet seems to buy a few extra buffers (5.60 bucks or so) and lift them all and start with one and move on. I wish I read your post at the beginning. I double checked all the SMDs at the beginning so I can be sure that they are correct. I remeasured the resistors.
.
My DVM is average. It's the radio shack DVM that they offer at 40 or 80 bucks depending on when you go into the store. Fuzed all in one thing. Probably not all that repeatable but hopefully good enough. I work in healthcare so these projects go in little runs here and there. I have to keep notes to remind myself where I was.
.
I have no idea about which oscilloscope to get. I will spend some time over at that forum.
.
Thanks a lot for your response. Off to mouser. ( I probably will redo a few of the 10uF caps simply because after the second amp I know I can solder them better now. I can really tell the difference between the first and second amp because of the solder job)
Good block
Quiet block
.
My DVM is average. It's the radio shack DVM that they offer at 40 or 80 bucks depending on when you go into the store. Fuzed all in one thing. Probably not all that repeatable but hopefully good enough. I work in healthcare so these projects go in little runs here and there. I have to keep notes to remind myself where I was.
.
I have no idea about which oscilloscope to get. I will spend some time over at that forum.
.
Thanks a lot for your response. Off to mouser. ( I probably will redo a few of the 10uF caps simply because after the second amp I know I can solder them better now. I can really tell the difference between the first and second amp because of the solder job)
Good block
Quiet block
Last edited:
Is the nut for the heatsink touching the soldertab in the top left corner, what is the shadow near c104 and the buffer in the top photo. Also in the top photo is r9 solder flowing to the pad beside it?
Is R9 used on the "quiet block" but not on the "good block"?
Are you sure the images are the right way around?
A missing R9 while using input H9 in single ended configuration could explain no sound.
Cheers,
Sebastian.
Are you sure the images are the right way around?
A missing R9 while using input H9 in single ended configuration could explain no sound.
Cheers,
Sebastian.
Sebastian,
Yes r9 is used on both blocks and the resistance has be checked. I called it the "quiet block" because although it sounds good - it's just not nearly as loud as the the second one. The volumes are far too apart to be considered anything but a screwup on my part. I checked with different sources and switched them back and forth.
multisync,
The nut is clear of the solder tab but I did notice that the solder was not bonding with the buffer. I redid that solder joint but still no increase in volume. r9 is clear in both blocks and C104 has no errant solder. I've checked the voltages on the 12 volt PT and that was fine. I'm beginning to suspect that I've goofed up the SE input resistors and the input IC somehow. Will redo those to see what happens.
I won't be able to get back to this for about a week, but I thank both of you very much for your time. I will post my troubleshooting stumblings!
Yes r9 is used on both blocks and the resistance has be checked. I called it the "quiet block" because although it sounds good - it's just not nearly as loud as the the second one. The volumes are far too apart to be considered anything but a screwup on my part. I checked with different sources and switched them back and forth.
multisync,
The nut is clear of the solder tab but I did notice that the solder was not bonding with the buffer. I redid that solder joint but still no increase in volume. r9 is clear in both blocks and C104 has no errant solder. I've checked the voltages on the 12 volt PT and that was fine. I'm beginning to suspect that I've goofed up the SE input resistors and the input IC somehow. Will redo those to see what happens.
I won't be able to get back to this for about a week, but I thank both of you very much for your time. I will post my troubleshooting stumblings!
Okay, I'm assuming that you connect the input signal return wire somewhere! 😀
A good point for this is J3, pin 2 (input "ground").
It looks like you have a wire soldered there, although we can't see which signal goes where...
Since you have a DMM at hand, why not do some signal tracing with it? 😉
Since The Wire is a DC coupled design, you can use any kind of (low) voltage test signals.
Cheap DMMs work best at true DC in their DC voltage range and at around 1kHz in their AC voltage range.
So if you can, try to generate a 1kHz sine signal with you computer and feed it to the input of each channel respectively. Then take AC and DC voltage measurements of all points in the signal path: input (before and after R9: H9 and pin 3 of U3), buffered (i.e. pin 9 of the U3 or pins 2 of each output buffer) and output (H12) with respect to ground (J3 and/or H13).
In case you can't generate an AC test signal, just use a battery (anything from 1.2V up to 12V) to feed a DC voltage to The Wire. Though no load at the output, please!
Please post what you measure at which node/spot. Both AC and DC measurements please, no output load connected.
Cheers,
Sebastian.
PS: It also can't hurt to take measurements of your supply voltages while powered up.
A good point for this is J3, pin 2 (input "ground").
It looks like you have a wire soldered there, although we can't see which signal goes where...
Since you have a DMM at hand, why not do some signal tracing with it? 😉
Since The Wire is a DC coupled design, you can use any kind of (low) voltage test signals.
Cheap DMMs work best at true DC in their DC voltage range and at around 1kHz in their AC voltage range.
So if you can, try to generate a 1kHz sine signal with you computer and feed it to the input of each channel respectively. Then take AC and DC voltage measurements of all points in the signal path: input (before and after R9: H9 and pin 3 of U3), buffered (i.e. pin 9 of the U3 or pins 2 of each output buffer) and output (H12) with respect to ground (J3 and/or H13).
In case you can't generate an AC test signal, just use a battery (anything from 1.2V up to 12V) to feed a DC voltage to The Wire. Though no load at the output, please!
Please post what you measure at which node/spot. Both AC and DC measurements please, no output load connected.
Cheers,
Sebastian.
PS: It also can't hurt to take measurements of your supply voltages while powered up.
Last edited:
I'm going to take a few steps back and learn a bit more about signal tracing with my new oscilloscope. Well, new to me. Beckman 9020. First build a signal generator. I'll be back.
hello everybody,
I finished my LPUHP !!!
This is not my first project, and due to my long experience in DIY audio, I make some mods to improve the sound.
I want to thank Owen for all the work it needed !
I finished my LPUHP !!!
An externally hosted image should be here but it was not working when we last tested it.
This is not my first project, and due to my long experience in DIY audio, I make some mods to improve the sound.
I want to thank Owen for all the work it needed !
Last edited:
to continue,
I tested my amp today in a mastering room, with an audio engineering and mastering friend's.
This is simply the BEST amp we ever try !!!!
Since a long time we were agree that nothing beat a tweaked Hypex UCD module, but it is wrong, the LUPHP is better , with a transparency (reverb listening ) that we never heard before (due to the very low THD I think).
The only matter is the low power... but I want to make another PCB with something like 40 or 60W, maybe with the Owen's help.
I remember that Owen tried to make a more powerful version but decide to stopped because of bad results, but I think it is possible, but will need absolutely damping resistor.
here is the mods I make on the original card :
- removing all X7r ceramic caps, replaced by sanyo oscon ultra low ESR polymer cap 120µF/20V
- putting damping resistors on all LME49610 output (0,1ohm)
- putting OPA211 instead of LME49990 ( opa211 has more relaxed sound and less aggressive for me)
- putting Elna Silmic as power capacitors (not for filtering)
some pics :
thank you again for this great project !!!
Richard
I tested my amp today in a mastering room, with an audio engineering and mastering friend's.
This is simply the BEST amp we ever try !!!!
Since a long time we were agree that nothing beat a tweaked Hypex UCD module, but it is wrong, the LUPHP is better , with a transparency (reverb listening ) that we never heard before (due to the very low THD I think).
The only matter is the low power... but I want to make another PCB with something like 40 or 60W, maybe with the Owen's help.
I remember that Owen tried to make a more powerful version but decide to stopped because of bad results, but I think it is possible, but will need absolutely damping resistor.
here is the mods I make on the original card :
- removing all X7r ceramic caps, replaced by sanyo oscon ultra low ESR polymer cap 120µF/20V
- putting damping resistors on all LME49610 output (0,1ohm)
- putting OPA211 instead of LME49990 ( opa211 has more relaxed sound and less aggressive for me)
- putting Elna Silmic as power capacitors (not for filtering)
some pics :
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
thank you again for this great project !!!
Richard
take a look at
http://www.diyaudio.com/forums/clubs-events/252446-ottawa-diy-winter-2014-a-4.html
post 31 for a hint of what might be coming, 50 watts enough?
http://www.diyaudio.com/forums/clubs-events/252446-ottawa-diy-winter-2014-a-4.html
post 31 for a hint of what might be coming, 50 watts enough?
Another stellar member of the Wire clan
My third family build, a pair of these to power a pair of Mark Alpair 10PeN. I've been breaking them in with a Bal Bal amp, and have been amazed at the drive the BalBal has. (Speakers a relatively easy 90db load). But these, fresh from my desk, are already sounding sweet! Stock, with a few resistor upgrades (probably meaningless), 18 db gain. I'll post pictures in a few weeks—want to build the enclosures first. Owen—thanks again for such a delightful build, and sounding superb, project!
My third family build, a pair of these to power a pair of Mark Alpair 10PeN. I've been breaking them in with a Bal Bal amp, and have been amazed at the drive the BalBal has. (Speakers a relatively easy 90db load). But these, fresh from my desk, are already sounding sweet! Stock, with a few resistor upgrades (probably meaningless), 18 db gain. I'll post pictures in a few weeks—want to build the enclosures first. Owen—thanks again for such a delightful build, and sounding superb, project!
Problems!
After running pretty flawlessly for a day, I decided to organize the shelf a bit better. Turned the amps off, and one channel had a slight pop—not too loud, but unexpected. When I switched it on, there was the same pop. And now, a hum from that channels speaker, audible from several feet away. I looked at the trim of cables, etc., but nothing was awry. So I hit play, and that speaker just sat there, with the same hummm.
I turned the amp off (pop) and unplugged everything and brought it up to my desk to look at it. I decided to hook up my junk tester speaker and check power supply voltages, etc. Well, the junk speaker started to smoke😱. So I turned the amp off, pronto. Unhooked the toast, er, speaker, and measured the speaker outputs. Got 16.63 v dc (no input at all). Also the polarity is switched (measures minus dc when hooked up + to amp ground and meter ground to amp +.
Any ideas what's gone wrong? No dis colored parts, heat, etc. is apparent. I've done a very quick test of the Alpair, and it may still be okay, which seems remarkable to me. (Sort of don't want to really test it just yet...) I'd done a thorough solder check before the original hook up, and will redo this again; I could definitely use a pointer to limit my search.
After running pretty flawlessly for a day, I decided to organize the shelf a bit better. Turned the amps off, and one channel had a slight pop—not too loud, but unexpected. When I switched it on, there was the same pop. And now, a hum from that channels speaker, audible from several feet away. I looked at the trim of cables, etc., but nothing was awry. So I hit play, and that speaker just sat there, with the same hummm.
I turned the amp off (pop) and unplugged everything and brought it up to my desk to look at it. I decided to hook up my junk tester speaker and check power supply voltages, etc. Well, the junk speaker started to smoke😱. So I turned the amp off, pronto. Unhooked the toast, er, speaker, and measured the speaker outputs. Got 16.63 v dc (no input at all). Also the polarity is switched (measures minus dc when hooked up + to amp ground and meter ground to amp +.
Any ideas what's gone wrong? No dis colored parts, heat, etc. is apparent. I've done a very quick test of the Alpair, and it may still be okay, which seems remarkable to me. (Sort of don't want to really test it just yet...) I'd done a thorough solder check before the original hook up, and will redo this again; I could definitely use a pointer to limit my search.