This is not just another gainclone

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
frugal-phile™
Joined 2001
Paid Member
Re: Whre is Thorsten's circuit?

GringoAudio said:
How does Quality Kit's LM1875-based kit compare to the Thorsten circuit?

If you built it as shipped it probably has too many parts... no doubt the more you can leave out the closer you'd get.

Does anyone have a take on whether you'd need to make any changes to pop in an 1875 into the Thor circuit?

dave
 
reality check

I've listened to an inverted gainclone for almost a year and there are some things everyone should know before their expectations get sky high.
The gainclone is as fussy about speakers as my SE valve amp. It plays great with my Lowhters and Tannoy Golds but is totally pathetic with my SF Signum. The SF is not a difficult load, having 1st order crossover and it can be very enjoyable even with a little 10W EL84 pp amp but with the gainclone it's vurtually unlistenable.
And to all who seriously compare gainclone to a good SE valve amp, please guys lets get down to earth.

cheers

peter
 
frugal-phile™
Joined 2001
Paid Member
Peter,

If i get it correctly this is the as-built (i've added the optional 1,5k load resistor in blue)?

dave
 

Attachments

  • minimizedgainklone.gif
    minimizedgainklone.gif
    5 KB · Views: 9,018
Re: Re: Whre is Thorsten's circuit?

planet10 said:
Thanks for link to the Forum, about thorsten's LM3875

Found this article mentioned in the Forum.

It is about parallelling LM3886, but could probably be valid for other chips.
Four LM3886 gives more than 200 Watt out!

Also Schematic


DC Electronics -App Note - PARALLEL-BRIDGED CONFIGURATION USING LM3886

This is Thorsten's comment on that article:
Hi,

This Parallel/Bridging principle applies to all Power Op-Amp's. In fact, a number of recent Jeff Rowland Amplifiers employ Bridged/Parallel LM3886.... They call "intelligent output transistors".

If I find Tripath Eval Modules too expensive for my Subamp for my new Speaker Project (18" Eminence Kilomax in URP Subwoofer mode, 15" Tannoy Coax, Supertweeter) I'll be using such a system as "fallback".

BTW, the parallel/bridging can be achieved without any further active components than the Amp Chip's themselves....

Ciao T

/halojoy - has not BRIDGED yet -Is not DEAF enough, to need so much power - but the day may come when he is 80 years old
 
frugal-phile™
Joined 2001
Paid Member
vuki said:
You can switch LM1875 instead of LM3875, but you'll need power xformer with lower sec. voltages (2x18V). Also, I had some difficulties with DC offset (it was 300-400mV) so you'll probably need adj. resistor +cap. at positive input to ground. But it sounds as good as LM3875.

Thanks. I have a set of both. I was going to use 18-0-18 on the 3875 anyway -- from the datasheets that is the optimum voltage.

dave
 
expectations to high?

Just as Peter (SA) I'm listening to gainclones (in various iterations; TDA1514, TDA7293/94, LM1875, LM3875, LM3886 inv & noninv) for more than 2 years. The best results (to my ears) are with TDA7293/94, BUT: my 300B SE amps are miles away from it. Really - no competition! Of course, one has to choose speakers carefully, just as with gainclone but even more so.
I'm listening to Tannoy HPD385's in 180lit BR boxes, but I also have 2-way DIY Accutons that don't go with 300B's but sing with gainclones. And yes-gainclone doesn't work well with power hungry or low impedance speakers.
 
I was also using gainclone with Accutones (midrange, Cabasse for woofer) and they went OK. My friends are using horns and Dynaudio based speakers and they are fine too. They were also using 300B and they say that although gainclone sounds different it doesn't give away much comparing to tube amp. It is just different kind of sound.

Remember that the parts that go into gainclone, make tremendous difference. If you use crappy componenets you cannot expect miracles. The same goes to chassiss construction, damping and layout. I'm using 400VA transformer with mine.

Dave, the schematic you show is the setup I'm using. I didn't try blue resistor yet. I also found that cap is necessary. My DAC has a servo and basically 0V DC at the output, but the amp had 500mV of DC when I wasn't using the cap. The same goes for BOZ with capacitive output. The amp still had some DC offset. When I finally placed BG 4.7u (type N) cap at the input the offset is steady at 20mV at both outputs.
 
Re: Re: Re: Whre is Thorsten's circuit?

halojoy said:

It is about parallelling LM3886, but could probably be valid for other chips.
Four LM3886 gives more than 200 Watt out!

/halojoy - has not BRIDGED yet -Is not DEAF enough, to need so much power - but the day may come when he is 80 years old

I have had quite a bit of experience in bridging these ICs.
The problem is though, i havnt heard enough amps to give any opinion on how good they sound. Ill have to build a Aelph one day to use as a benchmark ;)

For those looking for high power, i managed to squeeze 220WRMS into 8ohms at below 0.08 THD+N out of two LM3886s (bridged). To get this, they were run on a 40-0-40 power supply. At this power though, heatsinking was a real problem. I ended up bolting the LM3886s directly to aluminum bar which was then insulated by a large piece of mica from the big heatsinks.

I use it to power a 12" sub for home theatre... works very well. Cant beat it for cost, size and simplicity.

Assembled PCB looks like this:
An externally hosted image should be here but it was not working when we last tested it.
 
"I had a friend build a discrete version of a gain clone. It sucked, food for thought. There is a lot to be said for keeping the layout as compact as possible."

More than compactness of layout there is something to be said about precision laser etching of multiple semiconductors on the same silicon wafer. That goes a long way in ensuring gain match, symmetry of the layout, thermal stability, interference control etc etc.
 
Retired diyAudio Moderator
Joined 2002
Here is the original thread on AA:
http://www.audioasylum.com/audio/general/messages/172637.html

Here is the more recent thread with the pictures:
http://www.audioasylum.com/audio/tweaks/messages/60943.html
(pictures later in the thread: "Inside of Gaincard? - K.T. 14:37:53 05/20/02 (5) ")

Interesting how simple it looks... It looks like they used the isolated version of the opamp. The pinout is similar to the pinout of the LM3875, with the feedback resistors, but I am not entirely sure that it is the same.

--
Brian
 
Retired diyAudio Moderator
Joined 2002
Here is the original from cfraser about the pics posted to the Chip Amp bbs:

http://pub4.ezboard.com/ffakeidsfrm1.showMessage?topicID=143.topic

"I will be posting some photos of the Gaincard on my site soon. I've been thinking about how to approach this so that there are no hard feelings or ruffled feathers. You will understand much better why when you see... I've come to the conclusion that I'll just post the info as though it were any other piece of audio gear, and let it stand on its own sonic merits, like any other, but *I will not identify the gear*, but you the faithful will know what it is (because everything else IS identified). BTW, there are two filter caps, 2 gain resistors, an input resistor, a Zobel, a FB loop cap, an LM3875, all parts stripped of ID, all resistors are carbon comp 5%, no SMD. No real surprises, more than the sum of its parts... "

--
Brian
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.