50w @ 8r / 100w @ 4r / 200w @ 2r

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
i desperately need a schematic of an AB/B class amplifier that could give 50w at 8Ù, doubling this figure until at least 2Ù ... yes, a beast in a compact case ... being able to drive the most current thirsty speakers on the world ... shall be a truly symmetrical, stable design... preferably on mosfets....
any help shall be appreciated
 
diyAudio Member
Joined 2002
costiss said:
i desperately need a schematic of an AB/B class amplifier that could give 50w at 8Ù, doubling this figure until at least 2Ù ... yes, a beast in a compact case ... being able to drive the most current thirsty speakers on the world ... shall be a truly symmetrical, stable design... preferably on mosfets....
any help shall be appreciated

Hi,

in Elektor 10/90 was published "MEDIUM POWER AMP"
with some caracteristic:
60W/8E
110W/4E
170W/2E with 0.1%THD
PS is +/-24V
IM (1W) 0.01%
SR 30V/s
DF 100 on 20kHz, 200 on 100Hz
output devices BJT pairs (3x) BD911/912

Regards
 
Ya Don't Get Ought For Nought.

To double power into lowering load impedences, a 'perfect' power supply is required.
This requires over rated transformers and smoothing caps, but pays nice dividends at all power and load impedence levels.
The downside is that this adds much cost.
IOW the supply voltages should not sag with lowering load impedence and power delivery.

Eric.
 
i do know this mr feedback... its 12A peak @ 24v on a 2R load, plus a 50% margin 18A @ 28VDC 18A @ 21VAC -> 378VA per polarity... the point is, it is not going to run into a 2R load all the time.. most likely into the 8r the b&w 603 present.. the overspecification is to ensure that the amp will be stable with any (rational or irrational) fluctuation of the speakers impedance...caps? 4 x 47000uf @ 35V is enough (2ch) i think.....
:devilr: :darkside: :emoticon:
 
"i desperately need a schematic of an AB/B class amplifier that could give 50w at 8Ù, doubling this figure until at least 2Ù ... yes, a beast in a compact case ... being able to drive the most current thirsty speakers on the world ... shall be a truly symmetrical, stable design... preferably on mosfets....
any help shall be appreciated"

Are you hung up on mosfets?

The GAS Son of Ampzilla, Krell KSA50, and the Leach all come to mind.

All are full complementary from input to output using regular bi-polar transistors.

The GAS:

http://home.kimo.com.tw/skychutw/ampzilla/schematics/sonAmpzilla_sch.jpg

A Mark Levinson design, the same basic topology with J-fet inputs:

http://home.kimo.com.tw/skychutw/Circuits/jc3schematics.jpg

If you insist on mosfets for outputs, the Hafler XL280 is the same topology as the Mark Levinson, just with fet outputs:

http://www.hafler.com/techsupport/pdf/XL-280_amp_man.pdf

An all hex-fet variation:

http://www.passdiy.com/pdf/A75p1.pdf

http://www.passdiy.com/pdf/a75p2.pdf

What I use:

http://users.ece.gatech.edu/~mleach/lowtim/graphics/ckt.pdf

http://users.ece.gatech.edu/~mleach/lowtim/

I get 450W/2R from three pair of outputs. I would strongly suggest using a regulted high voltage tier for the front end (like the GAS, XL280, and A75).

A 30-0-30V 400VA transformer per channel with 15,000~30,000µF is all you need for 200W/2R. Two pair of outputs.
 
Leach amp + 4 X 47,000uF = Fine.

costiss said:
i do know this mr feedback... its 12A peak @ 24v on a 2R load, plus a 50% margin 18A @ 28VDC 18A @ 21VAC -> 378VA per polarity... the point is, it is not going to run into a 2R load all the time.. most likely into the 8r the b&w 603 present.. the overspecification is to ensure that the amp will be stable with any (rational or irrational) fluctuation of the speakers impedance...caps? 4 x 47000uf @ 35V is enough (2ch) i think.....
:devilr: :darkside: :emoticon:

Yes 4 X 47,000uF should be all the snot that you need - many high power pro-amps run much less than this - like 30,000uF total for 2 X 1000W into 4 ohms.
With intelligent power supply and earthing arrangements, this should give pretty much limitless performance.

The Leach amp that Djk uses, although not mosfet ought to be good - I have not heard one but I do bellieve that they are pretty good.
Use 2SA1943 and 2SC5200 output transistors and it will never let you down.

Eric.

Eric.
 
Re: Ya Don't Get Ought For Nought.

mrfeedback said:
To double power into lowering load impedences, a 'perfect' power supply is required.
This requires over rated transformers and smoothing caps, but pays nice dividends at all power and load impedence levels.
The downside is that this adds much cost.
IOW the supply voltages should not sag with lowering load impedence and power delivery.

Eric.

Eric,

that's right. And have a look at the SOAR of output transistors, take to the account reactive compound of load impedance.

Pavel
 
the reason for the heavy oooooveeeerspeeeeciiiiifiiiiicaaaatiiiiooon of eveything is the quest for the perfect by the book looooooow impedance supply... thanks to all for the help...
no, im not stuck to mosfets.... i like tube, ic opamp, fet, mosfet, hexfet, bipolar, vfet designs (in random order here)
 
Hi,

I still working same project, but with 60W/8ohms - 240W/2ohms. For this amplifier I use +/-40V power supply, with 800W transformer, and 4pairs of MJL21193/94 output.
The prototype is working and gives 400W with 1ohm load. :nod: But this application needs fan for cooling, because 50cm heatsinks, is not enough:redhot:

Sajti
 
costiss, I suggest that you start from the speakers and then go backwards.

Max voltage, min impedance => gets you max ampere + add some more

PA use or party level => Gets you the smoothing package and transformer => heatsinks, output transistors.

If don't have a any working figures it's very hard to really design.... if you don't are made of money that is.

Note that you can't get lower impedance than the DC resistance by that you some idea of max current. Maybe there are expections but I think this is a good starting point.
 
my speakers are the DM603-S3 ...............................................
b&w.......... not exactly the most sensitive speakers on earth, and certainly not what i would use with an el34 se..........
the figure of doubling increases in importance by the time you first come in contact with impedance curves ..... arghhhh .... and understand a speaker is by long not a resistor.....
thats why i wat it...
moreover... i know 102 dB is , for me, deafening levels for my home. the 603 is 90dB efficient /2.83Vrms(that is 1w/8, 2w/4 etc)
two speakers output 93db with 2 X 1w (@ 8r always!!) at 1 meter distance. at 2 meters its 87 db (square of distance).so, with 32 w you can get ~102db.. thats enough for me... plus an optional headroom of , say, 50%, ~50w @ 8R. i believe it is more clear now.now, say the lovable b&w has a dip in impedance around ~50Hz to 3,6 R (and a -180 phase...ok, thats far fetched)... that means, if the amp should output 18vrms @50hz (that on a 8r load is 41w), the load would req. 18/3.6=5a rms (!!!)
thus a power of 90w. to output this power, it should be able to double its power precisely at least up to 3R, if a 50w amp.
its not irrational dreams, its just reality.. if i were a figures freak, i would test my amps only with dummy loads ... i received a 500w 8r resistor today, really fine.... hehehehehehehehehehe
 
...and you can certainly have a lower impedance than the dc resistance.... the simplest example i can think of? a purely capacitive load on the amp... results? infinite(ideally) (some 10 x MOhms anyway) dc resistance, 1/2p*f*c (if i recall right) impedance... a 1uF cap?
1/6.28*1*10^-6*1000Hz=1000000/6.28*1000=159R only....
bit more complex now... say the behaviour of ur load resembled that of a resistor and inductor paralleled.. z@35hz would be (say r=8, l=12mh) XL=6.28*35*12*10^-3=2.63R
z=8*2.63/sqr(2.63+8)=6.4R > 8R DC
:bawling::bawling: unfortunately, IT CAN HAPPEN....:bawling: :bawling: :bawling:
:dead: :cannotbe: :clown: :bored:
 
Disabled Account
Joined 2003
I think the questions should be: 1) will the amp be stable at 2ohm? and 2) will the output devices be able to deliver enough current into a 2ohm load? power dissipation doubles on the output devices at 2ohm vs. 4ohm and 3) will the power supply generate that much juice?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.