HI ALL
I have built an ESP project 19
Single Chip 50W Stereo Amplifier
using the LM3886 PCB
Also his project 97 Hifi Preamplifier
Hi-Fi Preamplifier
From scratch.
I have to say this thing is incredible
According to him this will put out 56 watts at +/-35volts into 8 ohms
BUT I used a 2x28 volt transformer +/-42volts Per Rail
And this thing is giving my Pioneer SX950 85 watts per channel some stiff competition
What is this LM3886@+/-42 volts putting out at 8 ohms (Cooling Fan)
(The Pioneer is still a Monster)
Thanx
I have built an ESP project 19
Single Chip 50W Stereo Amplifier
using the LM3886 PCB
Also his project 97 Hifi Preamplifier
Hi-Fi Preamplifier
From scratch.
I have to say this thing is incredible
According to him this will put out 56 watts at +/-35volts into 8 ohms
BUT I used a 2x28 volt transformer +/-42volts Per Rail
And this thing is giving my Pioneer SX950 85 watts per channel some stiff competition
What is this LM3886@+/-42 volts putting out at 8 ohms (Cooling Fan)
(The Pioneer is still a Monster)
Thanx
Why not read the LM3886 app. notes and find why (not simple) a max +/-35V for 8R loads is recommended and more than that is risking damage that the thermal limiting circuits can't cope with. When the nominal speaker load impedance drops to 4R, the recommended supply limits are then reduced further to only +/- 28V.
http://www.ti.com/lit/ds/symlink/lm3886.pdf
http://www.ti.com/lit/an/snaa021b/snaa021b.pdf
Note comments in sections 4.1-2 in the second doc.
Just connecting the higher rail supplies would not pose much of a threat but when a speaker is also connected and you approach a higher than maximum rated output before clipping, the amplifier will be distorting badly, because thermal limiting will also be acting to keep the device within its SOA. It may eventually fail or not but it won't be doing what is considered safe or recommended operation. You don't actually hear any extra clean watts as by magic. The price is heavy distortion on broad signal peaks.
Maybe you like it that way or maybe you don't notice. The good news is that the difference between 68 and 85W is almost zero - you can't easily pick the difference directly other than by secondary vibrations like speaker rattles, or objects in the room. If you do hear that your amp is louder and much cleaner, a full service to the Pioneer amplifier is probably long overdue.
http://www.ti.com/lit/ds/symlink/lm3886.pdf
http://www.ti.com/lit/an/snaa021b/snaa021b.pdf
Note comments in sections 4.1-2 in the second doc.
Just connecting the higher rail supplies would not pose much of a threat but when a speaker is also connected and you approach a higher than maximum rated output before clipping, the amplifier will be distorting badly, because thermal limiting will also be acting to keep the device within its SOA. It may eventually fail or not but it won't be doing what is considered safe or recommended operation. You don't actually hear any extra clean watts as by magic. The price is heavy distortion on broad signal peaks.
Maybe you like it that way or maybe you don't notice. The good news is that the difference between 68 and 85W is almost zero - you can't easily pick the difference directly other than by secondary vibrations like speaker rattles, or objects in the room. If you do hear that your amp is louder and much cleaner, a full service to the Pioneer amplifier is probably long overdue.
NO WAY you can get that much power out of +/-42 V rails into 8 ohms.
1) the power transistors have a saturation voltage across them, if no better data or mesurements available, 4V drop is a good (real world) estimate, so your +/- 42V now become +/- 38V peaks ... being optimistic.
2) we don't have power supply details, but no regulated one was mentioned, so I assume a regular iron transformer + diode bridge + filter capacitors raw/unregulated one is used, what 99% amps use by the way, and it will also drop, you must also substract peak to peak ripple voltage , both add to at least 20% drop, maybe higher if not generously filtered.
This would leave some 30 or 31 Vpk available, some 110/120W peak or 55/60W RMS .
In any case, the built in thermal protection will quickly turn on as soon as the case (and chip) start to heat up.
Probably no big deal in a home or Hi Fi situation, where amp may be used at , say, 10% of its power most of the time; a definite problem for DJ/PA or Musical Instrument use.
Dylon Amps, an Argentine MI amp manufacturer, had big problems because they rised supply voltages to get true 70W RMS into 8 ohms (it's datasheet rated around 50/8) and amps started muting at random times when used live, yet most worked very well on rehearsal duty.
So that's an example of what you may find.
This Amp is crystal clear and has plenty of Backbone.
Has no negative effects at all @ 42 volts.
The spec sheet says 84 volts max voltage.
I have 2 more of these transformers that are scheduled to feed some more of these LM3886's.
At no load. You need to measure power supply droop at high output.
With a musical program ( transient in nature) and forgiving speakers (nice impedance curve) you can get away with stunts like this. I bet some popular speakers would give your amplifier fits.
And if you did a steady state test with sine waves, you would see your amplifier exhibit some of the characteristics you've been warned about. All the caveats you've been warned about are real.
As long as this thing sounds as Great> as it does>>(Make that Excellent)
Plenty of Power and doesn't blow up or catch on Fire. I'm Happy.
I custom made a 12"x15" chassis made out of 1/8" aluminum and bolted the heat sink directly to it with holes drilled thru the chassis right under the heat sink, also Drilled holes all across the front of the Chassis for evacuation of accumulated heat from the transformers and the power supply(s). Cooling fan in the middle of the back Panel drawing out. I know what these chips do when they get HOT because I disconnected the cooling fan and let it get hot. It's protection circuit kicks in. It does not do that with the configuration I have. The Aluminum Chassis is also a giant heat sink its self. Also there is no evidence of the SPiKe Protection Circuitry kicking in because it would be very Audible.
Ah. He only has 42Vdc, meaning 0-42Vdc or ±21Vdc.
So this statement was just a lie
His chipamp is well safe at only half the maximum supply voltage.
.
Pity he does not understand how to present the information in unambiguous format, so that we can understand.
That makes his claims for enormous power output even more ridiculous.
42 volts is 42 volts.
Well to be fair, that's not what you said above
That post reads as though you are running it on half its rated maximum.Play nicely.
+-42 VDC, and the amplifier can handle an absolute of 84 volts.
so its fine.
the supply dropps a few volts anyways under a heavy load and that is possibly enough to keep the amp within SOA.
in my experience chips feed with close to the stated maximum sound the best anyways.
and again, its working properly, it sounds good, then what is all the fuzzzzzz ?
i call that a job well done. does what its supposed to do, so its just fine as it is.
so its fine.
the supply dropps a few volts anyways under a heavy load and that is possibly enough to keep the amp within SOA.
in my experience chips feed with close to the stated maximum sound the best anyways.
and again, its working properly, it sounds good, then what is all the fuzzzzzz ?
i call that a job well done. does what its supposed to do, so its just fine as it is.
Most people don't like to run at the absolute max (2*42=84). If the speakers are efficient enough and benign enough a load that it works then great, but you will get more average power out at lower voltage rails.
No derating though so anyone of an engineering bent will point that out, which has happened.
But if OP is happy and understands the risks all is good.
To begin with 28+28VAC is not +/-42V rails , rather +/-38 to 39V , no load, so all this boasting about "look Mom!!!! I run a +/-35V rated amp at +/-42V !!!!!!!!!!!!!!!!" has no base.
Just sayin' .
As a side note I already mentioned a Real World Music Instrument amplifier which caused its makers endless trouble (negative comments started to pile up in Music Forums, first hurting then killing sales) because they did about the same as what the OP claims.
Of course, MI/PA/DJ amps are used in a Professional environment, not relaxed at home (nothing wrong, just lighter use) and then you *quickly* learn to trust and respect datasheets.
Just sayin' .
As a side note I already mentioned a Real World Music Instrument amplifier which caused its makers endless trouble (negative comments started to pile up in Music Forums, first hurting then killing sales) because they did about the same as what the OP claims.
Of course, MI/PA/DJ amps are used in a Professional environment, not relaxed at home (nothing wrong, just lighter use) and then you *quickly* learn to trust and respect datasheets.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.