hi
i only wanted to ask for the right size of the power supply for a gainclone (mono platine).
i've planned a 20-30.000µF supply for each channel.
but i heard that the sound of the lm3886t would decrease extremely if i use such a large supply.
can some 1 confirm this rumor ?
which size would be alternatively the best one ?
____
does the size of the transformator (i want to use 250-300 VA each channel) influence the sound performance of the gainclone in the same way as the capacity of the power supply ?
can i unconsidered use such a oversized transformator ?
____
i also heard that the lm3886t prefers for 4 ohms use 18V than more V.
which voltage would be the best one for a 4-8 ohms use ?
____
sorry for my technical incompetence
and thanks for your effort
i only wanted to ask for the right size of the power supply for a gainclone (mono platine).
i've planned a 20-30.000µF supply for each channel.
but i heard that the sound of the lm3886t would decrease extremely if i use such a large supply.
can some 1 confirm this rumor ?
which size would be alternatively the best one ?
____
does the size of the transformator (i want to use 250-300 VA each channel) influence the sound performance of the gainclone in the same way as the capacity of the power supply ?
can i unconsidered use such a oversized transformator ?
____
i also heard that the lm3886t prefers for 4 ohms use 18V than more V.
which voltage would be the best one for a 4-8 ohms use ?
____
sorry for my technical incompetence
and thanks for your effort
Most people use way larger supply's then are needed.
lets say your LM3886 used with 4ohm load and a 30 V+/- supply will put out 67 watts at 60% efficiency so the amp will need at least 107 watts or 107VA transformer, so now lets look at sag in the supply your little LM3886 can put out a PMPO of 120watts if you try to draw that the supply will sag createing a large voltage drop on the supply making the amp less linear adding distortion ect..... so now take that 60% efficiency figure and apply it to the 120watt peak so now you need around 200 watts per chip. If you want you could add another 20% saftey net, but I think it would be over kill
lets say your LM3886 used with 4ohm load and a 30 V+/- supply will put out 67 watts at 60% efficiency so the amp will need at least 107 watts or 107VA transformer, so now lets look at sag in the supply your little LM3886 can put out a PMPO of 120watts if you try to draw that the supply will sag createing a large voltage drop on the supply making the amp less linear adding distortion ect..... so now take that 60% efficiency figure and apply it to the 120watt peak so now you need around 200 watts per chip. If you want you could add another 20% saftey net, but I think it would be over kill
halo0925
What would you suggest for cap size to drive a 8 ohm load?
I used a pair of 12,000 mfd caps on my Chipamp.com power
supply board to power two channels. I consider this to be overkill.
I also played around with the bypass cap size in the power supply
to arrive at a more pleasing sound. Maybe you could help explain
why some are using 1,000 mfd caps on the amp channel board
and if this large a cap is really needed.
Thanks
Joe
What would you suggest for cap size to drive a 8 ohm load?
I used a pair of 12,000 mfd caps on my Chipamp.com power
supply board to power two channels. I consider this to be overkill.
I also played around with the bypass cap size in the power supply
to arrive at a more pleasing sound. Maybe you could help explain
why some are using 1,000 mfd caps on the amp channel board
and if this large a cap is really needed.
Thanks
Joe
student said:
1. Yes, I can confirm that rumour, I wouldn't use more than 2200uF per rail. If your speakers need more, build different amp.
2. 300VA transformer is rather conservative size. The quality and size of transformers are approx 50% of the caps influence.
3. I did some testing with Variac connected to PS, and the panel of listeners preferred the higher voltage supply (up to 40V per rail) even wiith 4 ohm speakers. Some will still claim that 18V sounds better, but those people also prefer more soothing and less alive sounds
PS
Let get realistic, a power supply with an AC transformer can only be so clean, does a gainclone sound better powered by batteries?? is pure DC better? is regulation better? sound quality is in the ears of the person listening try larger caps you might like the sound someone else might not, sound is subjective.
IMHO the cleaner the supply the better this means more capacitance bypassing ect........
Do what sounds good to you
Let get realistic, a power supply with an AC transformer can only be so clean, does a gainclone sound better powered by batteries?? is pure DC better? is regulation better? sound quality is in the ears of the person listening
try larger caps you might like the sound someone else might not, sound is subjective.IMHO the cleaner the supply the better this means more capacitance bypassing ect........
Do what sounds good to you
Re: PS
I was using GC with batteries for more that a year, in the end I switched back to AC as I find it sounding more realistic.
The big caps kill what GC has best to offer: the freshness of the sound, few other amps do it better, don't kill it with the big caps
... as your mains supply.halo0925 said:
I was using GC with batteries for more that a year, in the end I switched back to AC as I find it sounding more realistic.
The big caps kill what GC has best to offer: the freshness of the sound, few other amps do it better, don't kill it with the big caps
Peter,
Most of my amps have all been very large BPA amps, I heard all this and that about chip amps and gave it a shot, I didnt expect very much because I had a bad taste for chip amps like the original STK modules.
Built my first chip useing your schematics, by the way your patek is pure art work
I was amazed but, I was useing two carver 4.0 t's bridged for the bottom end and my KSA250 for the top so the little chip was out classed. So I went on a quest and about 10 BPA units latter I found got a circuit that maintains the sonic signature of a single chip, but the power to drive my setup, and it uses a large amount of capacitance, it may be different for a single chip I used 1500uf fine golds on my original single chip based on your circuit.
Most of my amps have all been very large BPA amps, I heard all this and that about chip amps and gave it a shot, I didnt expect very much because I had a bad taste for chip amps like the original STK modules.
Built my first chip useing your schematics, by the way your patek is pure art work
I was amazed
but, I was useing two carver 4.0 t's bridged for the bottom end and my KSA250 for the top so the little chip was out classed. So I went on a quest and about 10 BPA units latter I found got a circuit that maintains the sonic signature of a single chip, but the power to drive my setup, and it uses a large amount of capacitance, it may be different for a single chip I used 1500uf fine golds on my original single chip based on your circuit.
Re: Re: gainclone power supply
Peter,
Some time ago I took apart the power supply (Humpty) of my original 47 Labs Gaincard. I don't remember exactly after all this time, but the huge transformer inside was set to output something between 34V and 36V or somewhere in between.
It's interesting to me that you're finding some sound improvements at the higher voltages. I always wondered why the Humpty was rated at the higher voltage when talk among gainclone builders was generally of voltages of between 18V and 24V.
Best,
KT
Originally posted by Peter Daniel
...I did some testing with Variac connected to PS, and the panel of listeners preferred the higher voltage supply (up to 40V per rail) even wiith 4 ohm speakers. Some will still claim that 18V sounds better, but those people also prefer more soothing and less alive sounds
Peter,
Some time ago I took apart the power supply (Humpty) of my original 47 Labs Gaincard. I don't remember exactly after all this time, but the huge transformer inside was set to output something between 34V and 36V or somewhere in between.
It's interesting to me that you're finding some sound improvements at the higher voltages. I always wondered why the Humpty was rated at the higher voltage when talk among gainclone builders was generally of voltages of between 18V and 24V.
Best,
KT
Hi,
the supply voltage will determine the reliability.
If you combine high voltage rails and low impedance high phase speakers then the chipamp is capable of killing itself, but it depends on how you listen to your music.
If you listen @ an average of 500mW to 2W then allowing 10db peaks just takes the amp to about 10W on transient peaks for compressed music and even shorter peaks of upto 100W on dynamic source material.
Prolonged high power will cause the chip amp (and any other amp) to heat up. Go on too long and the amp will kill itself.
That seems to indicate that sub-woofer duty and/or party nights and/or DIY discos are not suited to chipamps.
Play at normal levels and into normal speakers and the chipamp will survive on high supply voltages. Abuse it when on high supply voltage (or even recommended voltages) and you risk killing it.
the supply voltage will determine the reliability.
If you combine high voltage rails and low impedance high phase speakers then the chipamp is capable of killing itself, but it depends on how you listen to your music.
If you listen @ an average of 500mW to 2W then allowing 10db peaks just takes the amp to about 10W on transient peaks for compressed music and even shorter peaks of upto 100W on dynamic source material.
Prolonged high power will cause the chip amp (and any other amp) to heat up. Go on too long and the amp will kill itself.
That seems to indicate that sub-woofer duty and/or party nights and/or DIY discos are not suited to chipamps.
Play at normal levels and into normal speakers and the chipamp will survive on high supply voltages. Abuse it when on high supply voltage (or even recommended voltages) and you risk killing it.
I am guessing that to meet this requirement the supply must provide for high voltage when running quiescently and droop to a safe voltage when delivering big power to the bass driver at the other side of the crossover.Nordic said:
If my guess is right then you are suggesting that high dynamic overhead is good.
I suggest that high dynamic overhead is BAD for sound quality.
A stiff supply that does not droop badly sounds better than a limp supply that wilts each time a sustained note asks for some power to be delivered to the speaker.
Power supply combined with adequate heatsinking will govern the life span of the chip amps. I'm not convenced that the 4 ohm load will kill the chip amp because companies like Rock-ola that manufacture juke boxes also use the LM3886 in their amplifiers.
Quote from section "D" specifications and operating features.
"Internal 120 watt RMS at 4 ohms stereo amplifier"
"External 240 watt RMS at 2 ohm switchable stereo or mono amplifier"
The above was taken from a Rock-ola Juke box service and operation manual. I would assume that since Rock-ola was one
of the first to use the chip commercially they would have engineers that were capable of keeping the bugs out of the product. In checking I have found a very low rate of amplifier problems associated with this product.
The commercial version employs a very large heat sink with takes care of the heat build up from a stereo amplifier and a bridged stereo amplifier. These amplifiers run on 22-0-22 AC before rectification.
My favorite amplifier has always been a bipolar amp if I am going to listen to solid state amps. I find the "chip amp" nothing but an amusement to fill some idle time. No self respecting person could compare the quality or performance of a "chip amp" with a properly designed fully complimentry piece of gear. They don't sound bad and any idiot can put one together and this cannot be said for a complex design.
Quote from section "D" specifications and operating features.
"Internal 120 watt RMS at 4 ohms stereo amplifier"
"External 240 watt RMS at 2 ohm switchable stereo or mono amplifier"
The above was taken from a Rock-ola Juke box service and operation manual. I would assume that since Rock-ola was one
of the first to use the chip commercially they would have engineers that were capable of keeping the bugs out of the product. In checking I have found a very low rate of amplifier problems associated with this product.
The commercial version employs a very large heat sink with takes care of the heat build up from a stereo amplifier and a bridged stereo amplifier. These amplifiers run on 22-0-22 AC before rectification.
My favorite amplifier has always been a bipolar amp if I am going to listen to solid state amps. I find the "chip amp" nothing but an amusement to fill some idle time. No self respecting person could compare the quality or performance of a "chip amp" with a properly designed fully complimentry piece of gear. They don't sound bad and any idiot can put one together and this cannot be said for a complex design.
burnedfingers said:
A little snobbish don't you think?
A well built chip amp can beat most budget/midrange commercial discrete designs. They do lose out to higher end commercial or well designed diy though, my Krellclone drives loads that the chip amps would give up on, and my Symasym easily beats them in SQ. However, I would willingly put a chip amp up against an ESP P3a for instance.
Actually I believe I stated a true point here. I have read thru numerous threads in the chip amp forum and have formed the opinion that a lot of the people building chip amps would be dead in their tracks trying to design or for that matter assembling something more complex. The chip amp is a simple solution for someone wanting to build something on a budget. They can say "I built this" with pride and that is part of the chip amp hype.
Sonically, there is no way this amplifier will compare with a well designed commmercial amplifier or a medium priced hifi amplifier.
When I auditioned my "chip Amp" against a good design from the 70's it got totally blown away. I auditioned a Altec 9440 against my LM3886 and the Altec won hands down. The Altec has a quasy complimentry output stage too. Granted the Altec is one of the better sounding amps even compared against the better amps of the 90's and newer. I have stacks of power amps sitting in my shop that I can audition the Chip amp against. I have owned some of the best over the years and worked on hundreds of different amps.
Sorry, I'm not impressed yet. Like I said I built it for amusement more than anything else. To a person that hasn't owned better amplifiers I'm sure the chip amp will sound like the best they have heard. I will go as far as to say the chip amp is better than the old STK packs.
No offense meant as its just my personal opinion as others have theirs.
hi
nice literature u guys post there
this is a picture of how my gainclone circuit board could look finally. i bought that platine twice for 22€ inclusive postal charges.
on the platine itself, i will use two 1000µF or maybe maximum 1200µF cap.
but why shall i not use a huge external power supply with 20.000µF each chanel ?
im not really a engineer and i only want to build this amp cause of interesting and also cause its the simpliest way to create some acceptable sounding stuff.
but somehow i dont understand why a 20.000µF power supply will might decrease the sound of this amp IF i also use those 1000-1200µF caps on the circuit board additional.
or did those comments concearning the cap-size refer to a circuit board where those huge caps where placed instead of my little ones ?
sorry for my low technical vocabulary and those circumstantially sentences
_______
im also aware of that this little chip amp will not beat up a krell or accuphase or what ever, but as some1 posted here, i think it's a nice try to start building amps.
and i also thnik/hope that the gainclone will keep up with my nad 310 "reference amp" or maybe with my yamaha-ax1050
that would be enough
nice literature u guys post there
An externally hosted image should be here but it was not working when we last tested it.
this is a picture of how my gainclone circuit board could look finally. i bought that platine twice for 22€ inclusive postal charges.
on the platine itself, i will use two 1000µF or maybe maximum 1200µF cap.
but why shall i not use a huge external power supply with 20.000µF each chanel ?
im not really a engineer and i only want to build this amp cause of interesting and also cause its the simpliest way to create some acceptable sounding stuff.
but somehow i dont understand why a 20.000µF power supply will might decrease the sound of this amp IF i also use those 1000-1200µF caps on the circuit board additional.
or did those comments concearning the cap-size refer to a circuit board where those huge caps where placed instead of my little ones ?
sorry for my low technical vocabulary and those circumstantially sentences
_______
im also aware of that this little chip amp will not beat up a krell or accuphase or what ever, but as some1 posted here, i think it's a nice try to start building amps.
and i also thnik/hope
that the gainclone will keep up with my nad 310 "reference amp" or maybe with my yamaha-ax1050 that would be enough
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.