DIY a really good class a amp for passive monitors, worth the effort ?

[Axelerator]

Hi, for sure there are many of you did this already 
so i would like to ask some serios ears , are there diy projects out there that could compete with the better pro-amps out there ?

I would need something from 50 watts to little more , that would get me in the area of more than 100 watts consumption for mono only. thats ok for some hours a day i guess.

There are many sites on the net (think passlab etc) (hifi - ) people raving about the amps but i would like to hear the opinion of somebody did/hear some DIY versus pro Amps (Bryston..add
fav. amp here ) on good passive Monitors/Speakers.

cheers
Axel

 

[gyraf]

You need a pretty good design, layout, and build to beat a simple LM3886 circuit.

The recent years I haven't bothered rolling my own power amps because of the development in chip versions.

Your hearing and/or attitude may well be different than mine..

Jakob E.

 

[JohnRoberts]

Any decent modern amp is not likely to be the weakest link in your audio chain... Speakers are harder to do well than amps, etc.

If you feel compelled to DIY an amp, the Class D  amps that Bruno Putzey (Hypex) designed are quite good while about as far from class A as you can get .  ;D

JR

 

[PRR]

Doug Self has shown that a good Class (A)B amp can be cleaner than Class A.

IMHO: Large DIY power amp development becomes a pile of costly blown-up parts, mysterious troubles, uncertain improvements.

Excessively wealthy audiophiles change amps several times a year, so there is always some supply of very-good lightly-used power amps in Hi-Fi markets.

For all that: a good DIY power amp is much easier than a *good* DIY speaker.

 

[JohnRoberts]

Doug Self has shown that a good Class (A)B amp can be cleaner than Class A.

The audiophile argument for Class A is to completely eliminate the hand-off between positive and negative output drive, which indeed removes the crossover distortion caused by finite turn-on/off times. It is not inherently more linear so could easily be higher distortion than a well executed class AB. There is a fair argument that type of distortion matters, but well executed modern designs can make all distortion extremely low.

IMHO: Large DIY power amp development becomes a pile of costly blown-up parts, mysterious troubles, uncertain improvements.

amen... preach it brutha

Excessively wealthy audiophiles change amps several times a year, so there is always some supply of very-good lightly-used power amps in Hi-Fi markets.

For all that: a good DIY power amp is much easier than a *good* DIY speaker.

Designing any product to be better than you can buy off the shelf is never trivial. Why do people think audiophile design is just a simple matter of throwing some expensive parts at any problem?

Consider Class A amps as more of a cold weather (room heater) project. Not summer fun.  ;D ;D

JR

 

[benqbasic]

Speaking from my own experience, I'd say it may be worth the effort depending on the design. But its not worth the price to diy one.
In saying that a diy class D is usually good value for money if you do your research about the prebuilt modules available.

Also don't for get about adding in the cost of a preamp depending on your setup.

 

[pucho812]

I've heard many a good think from the diy community about hypex amps...  Easy to assemble modules to make a power amp. Sound good to from people have said.

http://www.hypex.nl/

 

[audiomixer]

Hi Alex,

now that was a flash-mob! hope you don't feel to bad after being lynched... smile.
build one, class-A just for fun, may be one from the pass lab corner. test against a hypex class-D and a LM3886 chip amp. same power range. you will probably want to keep one of the later, but at least you can participate on a whole other level at the next round of discussion.
GDIY'ers are not impartial judges. on some topics the opinion leaders like to show of their years of experience, but imagine Bruno Putzey would have listened to them years ago....
on the other hand they will invest loads of cash into vintage n^ve technology and tube mixers...

so, don't be fooled, educate yourself, build stuff, listen carefully, enjoy your findings and report back if you find solid evidence.... double check on your listening with a independent listener panel, you might be biased.

cherio,

Michael

 

[JohnRoberts]

Hi Alex,

now that was a flash-mob! hope you don't feel to bad after being lynched... smile.
build one, class-A just for fun, may be one from the pass lab corner. test against a hypex class-D and a LM3886 chip amp. same power range. you will probably want to keep one of the later, but at least you can participate on a whole other level at the next round of discussion.

So your advice is to build all three... 

GDIY'ers are not impartial judges. on some topics the opinion leaders like to show of their years of experience, but imagine Bruno Putzey would have listened to them years ago....

I don't think he was looking for impartial comments. Sharing experience should be a good thing.

Show of(f)??  Speak for yourself. Years ago Bruno was a real deal design engineer at Phillips, a real company.  Bruno is pretty generous about sharing his knowledge and even co-moderates a R/E/P sub forum, but I haven't seen him around much lately. It looks like he is messing with DACs now and that will probably be good also.

on the other hand they will invest loads of cash into vintage n^ve technology and tube mixers...

Huh...

so, don't be fooled, educate yourself, build stuff, listen carefully, enjoy your findings and report back if you find solid evidence.... double check on your listening with a independent listener panel, you might be biased.

cherio,

Michael

Yes don't be fooled.

JR

 

[Dylan W]

I can't contribute, but found this (semi-relevant) Bruno Putzeys paper to be very interesting. http://www.linearaudio.nl/linearaudio.nl/images/pdf/Volume_1_BP.pdf

 

[ruairioflaherty]

The Hypex amps are fantastic, from the standard UCD180 all the way up through the Ncore, I've used them all.  If performance vs cost is the goal then that is the way to go, if DIY/fun/learning is the goal then I would start with a Douglas Self book. 

 

[Axelerator]

Hi and thank you all for your opinions !
..yes i already thought (hoped  ) for a nice discussion from you well respected engineers !

Speakers used will be very neutral and detailed with very good imaging while beeing easy to feed ..no way to even think of diy-ing better speakers for the same price..
This should end in a serious upgrade of my actual active monitor solution so i am in for microdynamics and fast transients.

- well the chipamp solution should be so cheap it is worth a try , just cannot believe..this things can perform like the really( i mean the hi $ but not the  esoteric equivalent type ) good amps ?- have to build one

- hypex was already on my list, so many big money active speakers use them..

Class A..
..just was dreaming of an industrial landscape  looking-thingy to power the speakers  : ..maybe with smoke also !!  ;D
-But, as i think  ,with 2*50 amps class a the resulting power consumption is 2/3 of my actual PC ...

Hey this seems adequate      :  http://www.audiostream.com/content/opera-only?utm_source=buffer&utm_medium=twitter&utm_campaign=Buffer&utm_content=bufferc7ef7

another strike against the proceeding political correctness everywhere !

 

[PRR]

> chipamp solution should be so cheap

The chip is the least part of a power amp. Your money goes into

Power transformer
Heatsinks
Power capacitors
Box

The PT and Caps will always be a huge chunk of money, and similar for ANY linear amplifier (slightly smaller for a good switching-amp).

Big (linear) amp needs big heatsinks to stay alive. Arguably even bigger in a "monitoring" application where you do not want minor sound-change when the amp gets hot. Switchers need far less sink, but don't under-do it.

Designing a good linear amp is a lot about consistency and matching. Here the chip has real advantages.

And a lot about Assembly. Lots of parts to collect and wire-up. The chip does this once and then "prints" thousands.

Any Real-World amp needs Protection. Classic parts-built amps can only watch the power flow in and out of the big transistors and "guess" what combinations will be "bad". A chip designer can get a lot closer to his big devices and get a better guess. (Which really means he can use smaller devices; the part-built amp can use multiple devices so each stays far away from danger).

 

[usekgb]

Doing a bridged chip solution can get you a really great amp for very little money.  Just watch out that your speaker Z doesn't drop below 4 ohms.  I've done a few with a DRV134 driving two LM3875's push-pull and have been very happy.  Use good sized heat sinks, and lots of reservoir capacitance.

If you are worried about the speaker load, use a pair of LM4780's in a parralel-bridged configuration.  This will supply plenty of current.  Just make sure that you put a 1R power resistor at the output of each amp.

 

[culteousness1]

I built a dual Chipamp for a friend some time ago.
Used Pier's Piccolo PCBs that are based on the LM3886.

Works like a charm. Class AB though.

Best,
Carsten

 

[0dbfs]

What about a tube power amp? Would that satisfy your needs?

Cheers,
jb

 

[Axelerator]

no real Hi voltage needs here !

I need a real neutral and fast amp with no colour.
I looked further into the Hypex story and see real love for this amps nearly everywhere .

I think the Ncore4000 with the special designed SMPS would be a good start.
Not much DIY but that is ok

Hope that they are not too dirty on the EMI side

Hey ,just click on the link i listed in my last post above for some real fun amp-wise .This one is a real Heater !

 

[ruairioflaherty]

Hope that they are not too dirty on the EMI side

The NC400 and SMPS combo is very very good.  There are zero issues with EMI.

 

[Axelerator]

good to hear!

 

[ruairioflaherty]

I've had several different DIY set of NC400 in my room and with PMC I have installed many hundreds of speakers with Hypex onboard with zero issue.