Amplifier design released (link fixed)

17 replies [Last post]
Offline
Joined: Dec 20 2003
Posts: 851

I have finalized the design for my current project, which many of you have been wondering about for some time. Most of the parts are on order and I can probably begin construction in a few weeks.

Basically, I am making a 40ish Watt stereo amplifier so that I can drive real 8 ohm home audio bookshelf speakers (former my rear channel surround sound speakers) at my computer. I've noticed that even the lowest end stereo equipment sounds VASTLY superior to anything the bears the label "computer speakers". Since my mains were destroyed in my move down from Oregon, the whole system has been replaced and my old rears are sitting on the shelf looking for a purpose in life.

Here are the Schematics:

(link moved down, use that one instead)

Oh and it wouldn't be a Dr. Bob projects without a wild paint job.

http://vader.inow.com/~drbob/amp/gloss.jpg

__________________

Dr. Bob
Applefritter Admin

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
Offline
Joined: Dec 20 2003
Posts: 851
Preemptive strike

I used OrCAD to draw those.

(I just know somebody will ask)

__________________

Dr. Bob
Applefritter Admin

Dr. Webster's picture
Offline
Joined: Dec 19 2003
Posts: 1687
Linky no worky. Sez it's forb

Linky no worky. Sez it's forbidden.

__________________

Applefritter Admin

martakz's picture
Offline
Joined: Dec 20 2003
Posts: 634
Yup, forbidden :(

Yup, forbidden Sad

__________________

Now using x86 machines. Bye bye iMac266 (Strawberry), G3 1400 and the G3 IIsi.

Offline
Joined: Dec 20 2003
Posts: 851
(*&*(^#ing permissions

Somehow the permissions got set differently for those files and not the other 10 or so I copied in the same batch. How very very odd.

Anyway, it's fixed now.

__________________

Dr. Bob
Applefritter Admin

martakz's picture
Offline
Joined: Dec 20 2003
Posts: 634
I never realised building an

I never realised building an AMP could be so simple... How did you manage to get all the freqs amplified by the same level?

Does the ripple from the PSU cause a hum on the speakers?

I might build one of these Smile

__________________

Now using x86 machines. Bye bye iMac266 (Strawberry), G3 1400 and the G3 IIsi.

doug-doug the mighty's picture
Joined: Apr 14 2004
Posts: 1355
you bet!

Audio amplification is really very simple in concept. If you want some real fun, check into tube (aka 'valve') amplifiers. The big difference there is the you will deal with some very high voltage (10kV and up), but you get a much cleaner sound as the harmonic distorition is on around the 3rd and fifth harmnonic vice solid state's 11th and 13th harmonic.

Really fun stuff. I have been desgining mine for the last two years. I have gotten the rough done from the recification all the way up to the bleeder. After that I have been undecided on quads vs pentodes. ;D Choosing the right tube will set my upper rail (voltage) level in my design.

Maybe one day I will get around to up loading my schematics...

Had a complete set I did for a graphic equalizer that used switched-capacitor bandpass filters, but I had overdesigned it as it needed 5 amps just to turn on but it went from .25Hz up to 11kHz. (That and Motorola d/c the chips I needed with no replacement, so I never got a chance to even build the dumb thing.

--DDTM

__________________

--DDTM ('Fritter Critter' since Apr 26 2004 - 18:16)

'If it ain't broke, take it apart anyways. If you can't take it apart, break it so that you can fix it.'

martakz's picture
Offline
Joined: Dec 20 2003
Posts: 634
Well there is alot of debate

Well there is alot of debate whether valve amps sound better than digital amps... Depends on ones tastes I suppose.

Ive used both valve and digital in music production, and im not sure if I prefer the valve sound. They do look cool glowing away though...

__________________

Now using x86 machines. Bye bye iMac266 (Strawberry), G3 1400 and the G3 IIsi.

Offline
Joined: Dec 20 2003
Posts: 851
Re: I never realised building an

martakz wrote:

I never realised building an AMP could be so simple... How did you manage to get all the freqs amplified by the same level?

You really should download and read the manual for the LM3876

martakz wrote:

Does the ripple from the PSU cause a hum on the speakers?

I haven't finished building it, but I keep my grounds done well, it shouldn't. You might want to add a 10 microfarad cap between the signal ground and the earth grounds as well. The LM3876 has over 100dB ripple rejection at 60Hz, so I'm not too concerned about that.

__________________

Dr. Bob
Applefritter Admin

Offline
Joined: Dec 20 2003
Posts: 851
Re: you bet!

doug-doug the mighty wrote:

Audio amplification is really very simple in concept. If you want some real fun, check into tube (aka 'valve') amplifiers. The big difference there is the you will deal with some very high voltage (10kV and up), but you get a much cleaner sound as the harmonic distorition is on around the 3rd and fifth harmnonic vice solid state's 11th and 13th harmonic.

Youa re on drugs if you think a tube amp has less distortion than a transistor amp made in the last 20 years. What do you think gives it that warm tube amp sound that people like so much? That's right! Distortion!

This amp should be around 0.06% THD.

The other differences to include are that tube amps use a lot more power, and the tube's output characteristics vary a *lot* (like as much as 20%) as they age. Plus you do have to replace them regularly and keep the contacts uber clean.

However, if you like that warm tube amp sound, especially for the spoken word, nothing beats it.

__________________

Dr. Bob
Applefritter Admin

CaveNerd's picture
Offline
Joined: Oct 10 2004
Posts: 17
That's a pretty ancient argument

:Smile Didn't Bob Carver settle that argument like, 20 years ago? :Smile

__________________

Look: strange women lying in ponds distributing swords is no basis for a system of government! Supreme executive power is derived by a mandate from the masses, not some farcical aquatic ceremony!

Offline
Joined: Dec 20 2003
Posts: 851
I finally figured out how to

I finally figured out how to make OrCAD print the whole project as one file, and also I added some EMI suppression.

http://vader.inow.com/~drbob/amp/schematics.pdf

__________________

Dr. Bob
Applefritter Admin

Offline
Joined: Dec 20 2003
Posts: 279
Ahh another OrCAD user. I am

Ahh another OrCAD user. I am hoping we will switch to better software at work at some point. OrCAD just sucks imho, would prefer to use Protel. But ohwell.

Tube amps are great if you play jazz or blues. Just gives it the right sound. Otherwise a transistor amp is better in everyway. FAR less power consumption, last longer, much smaller, and less distortion. But then this was all mentioned above...

__________________

G4 in a G5 Case: http://www.mlode.com/~stuka/g45/
iMac in FlexATX Case: http://www.mlode.com/~stuka/imac-atx/

doug-doug the mighty's picture
Joined: Apr 14 2004
Posts: 1355
yessireebob!

I never said that tunes had less distortion. The statement I made is in reference to the musucal fundamental.

Tubes producepredominantly low0order distortion. Transistors, in addition to low-order, produce greater levels of high-order distorition. Distortion which is low order is closer to the musical fundamental (2nd ond 3rd harmonic), while the high order is further away from the musical fundamental (11th harmonic and up). Although both are undersirabel when reproducing an audio signal, hige-order distortion products are the most objectionable as they are dissonant with musical fundamental.

When I said that they had low-order distortion, I never meant that they had less. Absolutely do they have more. It is just that the distortion produced is more agreeable to the ear, hence the 'warmth' you refer to. The warmth was more noticeable when transistors first came out as they had a 'tinney' sound, in part to transistor intermodulation distortion. I agree, transistors have come a long way in the last twenty years, but as for which is better, that is merely personal preference of the buyer, most especially when it comes to high-end implementations.

From outside the box, I disagree that the power requirements are higher - I have a solid state amp that is rated to draw a max of 3 Amps, and it gets just as loud. I trust your remark about power consumption being higher is in reference to the higher voltage. Yes, tubes require a much higher operating voltage, but that is really noghting more than converting 120VAC to 400VAC or 500VAC (higher in some applications). But the current draw is not that much higher than transistors. Still runs off 120VAC, still consumes 2-5 Amps (depending on design of output).

As far as the variance of the tube quality, that goes back to mfr quality. Many tubes are required to be matched, particularily in Class A amplification. You notice fewer variances when you go into tetrodes and pentodes as well, but even then, the performance of the tube abd its natural charateristics can be manipulated by good circuit design. In fact many high end amps using tubes only use tubes with top quality (yes they can be rated, similarily to the tolerance of a resistor or a capacitor). Not all tubes vary so greatly.

Yes, you have to keep your tubes clean, but the problem is not so much dust as it is oil from ones nasty little paws. As the tube heats, the oils can cause uneven heating on the glass, kinda like those nice halogen lights that burn down college dorm rooms when accidenatlly knocked over in a druken stooper (I was not even on that side of the campus that night, God is my witness!).

I am not trying to pick a fight, just clarify my statements.

Much love...
--DDTM

__________________

--DDTM ('Fritter Critter' since Apr 26 2004 - 18:16)

'If it ain't broke, take it apart anyways. If you can't take it apart, break it so that you can fix it.'

Offline
Joined: Dec 20 2003
Posts: 851
more

The easy way to deal with intermodulation distortion on a transistor amplifier is simply to shift it up beyond the range of human hearing. However, this amp I'm making is actually an op-amp design that can source very high current. It's 100% analog with a small switching delay in the nanosecond range.

Tube amps do require a lot more power at normal listening levels as they have to be heated considerably. My statement should not be interpreted to say that tube amp x always uses more power than transistor amp y, but rather tube amps are much lower efficiency with a higher ratio of heat generated vs output power.

On cleaning, I was referring to the contacts that the tubes plug into. They attract dust since they are charged plates, as well as have a propensity to oxidize. They need to be cleaned often in most cases. Also the metal in the cathodes becomes crusted over with time and that causes the performance characteristics to change quite a lot over the life of the tube.

At any rate, my goal for this amp (and any amp) is the most accurate reproduction of the signal possible and that alone simply precludes the use of a tube amp.

__________________

Dr. Bob
Applefritter Admin

Offline
Joined: Dec 20 2003
Posts: 851
Re: Ahh another OrCAD user. I am

Stuka wrote:

Ahh another OrCAD user. I am hoping we will switch to better software at work at some point. OrCAD just sucks imho, would prefer to use Protel. But ohwell.

I don't mind OrCad. I just wish it was easier to find the artwork in the libraries. My version is pretty old, but unless an up to date version of PADS falls off a truck in the near future, it's the best I've got (and it kicks the crap out of gEDA).

I should eat at the Baja Fresh near Mentor Graphics more often.

__________________

Dr. Bob
Applefritter Admin

doug-doug the mighty's picture
Joined: Apr 14 2004
Posts: 1355
cool...

I got a lot of joy out of designing audio ciruits like that, just never could find the time to build them.

Some of my other searches on tubes did find this product... http://home.swbell.net/ronsuthe/faq.html. It is pretty cool if you only need L/R. Too bad there is not option for full surround sound or THX. (Yes, I am still hung up on tubes Smile )

--DDTM

__________________

--DDTM ('Fritter Critter' since Apr 26 2004 - 18:16)

'If it ain't broke, take it apart anyways. If you can't take it apart, break it so that you can fix it.'

Offline
Joined: Dec 20 2003
Posts: 851
OK once again I've done an up

OK once again I've done an update to show both channels. This is the design that will go to the board fab.

http://vader.inow.com/~drbob/amp/schematics.pdf

__________________

Dr. Bob
Applefritter Admin