Chipamp working OK, then bad distortion

[AndrewT]

With properly built amp and even 40V rails, 4 ohm speakers should not be a problem. I'm talking from experience using such system for a year.

so why you bother what I'm asking?

because it worries me that you are condoning high supply voltages into low impedance loads, when we hear from many builders that precisely this situation leads to overheating.

I think that your specific installation is much less common compared to what most builders intend doing with their amplifiers.

 

[Peter Daniel]

How about you tell us what is your actual experience with those amps: how many did you build and what different loads did you test them with , over what period of time?

 

[pacificblue]

We still don't know what Gecko's asking from his amps, so why you bother what I'm asking?

The differences are obvious.

Btw the speakers are 88db efficient

96dB efficient.

That is 8 dB difference or more than six times the power requirement for the same SPL.

I was using 3 x 12V, 7.2Ah batteries per rail, which produced approx 39V DC, there was not much voltage drop at full power.

So i guess the capacitance would be 3000uf per channel.

That means no power supply ripple at all with the batteries, but an enormous ripple with only 3000 µF per channel, heating the amp further.

I later switched to bridged configuration, using 300VA transformer per channels and 1000/50 smoothing caps. That means 2 ohm load per isolated cheap

 

[Peter Daniel]

Re: Re: Chipamp working OK, then bad distortion

The differences are obvious.




That is 8 dB difference or more than six times the power requirement for the same SPL.




That means no power supply ripple at all with the batteries, but an enormous ripple with only 3000 µF per channel, heating the amp further.

Whay don't you get yourself acquainted with real life measuremens: http://www.diyaudio.com/forums/showthread.php?postid=1521304#post1521304

 

[Gecko123]

We still don't know what Gecko's asking from his amps, so why you bother what I'm asking?

I didn't expect this amp to put out huge, but it seemed to be doing so for awhile, thats why I was driving it hard, but I will be using it mostly to run small bookshelfs, 8 ohms, 86db efficient, run as computer speakers in small office, So the power thing wont be an issue, but I see now why it was distorting bad, thanks for the help.

 

[pacificblue]

Re: Re: Re: Chipamp working OK, then bad distortion

Whay don't you get yourself acquainted with real life measuremens:

Thanks, I got acquainted with them about 26 years ago for the first time. So, what have measurements of DC offset or of the power output into a resistive load to do with "belching speakers" after one hour of operation?

If you prefer to use those ICs outside of their specifications, suit yourself. The probability that it could be the exact reason, why they behave out of specs increases by the square of the abuse.

 

[Peter Daniel]

I didn't expect this amp to put out huge, but it seemed to be doing so for awhile, thats why I was driving it hard, but I will be using it mostly to run small bookshelfs, 8 ohms, 86db efficient, run as computer speakers in small office, So the power thing wont be an issue, but I see now why it was distorting bad, thanks for the help.

And this is how it goes: in real life very few people are actually pushing those amps. Most of the time only few watts are being used and then the amp does not really care what rails you supplying it with or what load is connected. This is why I'm not really opposing to using the amp at full rails with 4 ohm speakers; because in real life, most of the time it will work fine.

Myself, I had only problems with few recordings when at certain moments on a disk there was really deep bass and then the amp went into clipping, but the protection circuit never kicked in.

So am I condoning high supply voltages into low impedance loads? Not really, as you might noticed in my other posts. But if by any chance you end up with high rails, small heatsinks and low impedance loads, chances are you still be fine.

 

[lanchile]

Wow! He must have run the amp really hard.I have never "pushed" my amp to the limits, because it is "wayyyyyyyy" to loud for me.and if I need "MORE" power than this...I would need to move to a place with no neighbors around me lol. I am not concentrated in the power but in the sound quality.if He needs more power to "BLOW" the windows,He should get a professional amp like Crown getting 1000 watts per channel and speakers over 98db spl.

 

[pacificblue]

in real life, most of the time it will work fine.

Yes. And in real life, any device that only works fine most of the time and not all of the time, is crap.

Oh! You had an accident with that car? But hey! The brakes work really well, while you don't push them, don't they?

 

[balaboo]

Interesting - I have a FirstAct 50W guitar amplifier that uses a LM3886TF (the insulated package) as the power amplifier. Even when playing in a SMALL practice room with the band, it will crap out whenever i try to turn up the volume for leads, etc.

So, to those of you who find the LM3886 "loud enuff", goodie for you - my Traynor TS-25 (25 watts discrete !) plays plenty loud, except for having no effects. All I can say is I am NOT impressed with the LM3886TF; perhaps the LM3886 has better thermal/clip capabilities...

 

[AndrewT]

the lm series are designed for domestic duty in low fi applications.
Builders have found that medium fi performance can be obtained from these cheap chips but they are still domestic duty.

Do not expect them to take the power damands of a PA/live music application

 

[Peter Daniel]

Yes. And in real life, any device that only works fine most of the time and not all of the time, is crap.

Oh! You had an accident with that car? But hey! The brakes work really well, while you don't push them, don't they?

pacificblue,

Silly comments like that bring nothing to the discussion, I know you can do much better:

The best amplifier from a technical point of view, is the one best suited to its task. If you need an amplifier for civilized listening at home, it makes little sense to use one that can put out several kW, just as little as using class A amps for PA systems.

The best amplifier from a psychological point of view, is the one that satisfies its owner most. There are so many reasons, why an amplifier may do that. Some are related to sound, some to looks, some to technical data or properties, to price tag, brand name, others to being self-made. There are even people, who satisfy themselves more with the act of building, disarming and building again, than with the actual result, eh, Carlos?

The best amplifier in the world is always the next that you build (or buy).

 

[balaboo]

AndrewT;

So you feel that a standard discrete BJT/MOSFET amplifier is the way to go ? I am experimenting with a simple Hitachi style power amplifier that uses a MOSFET VAS stage driving a pair of TIP142/TIP147 output transistors (hey they ARE "integrated" [chip] devices). It has been very interesting to see the distortion spectrum as the amplifier is swept from no input to gross overload - I have found that low voltage/high-current capability has terrible distortions as compared to high-voltage/low-current using an output transformer (!).

 

[AndrewT]

If you want/need PA/Live music duty, then you must design the amplifier to survive the heat demands that will be imposed.
Chipamps cannot get anywhere near that.

I doubt TIPs can either.

 

[audiojoy]

I agree wholeheartedly with peter

 

[balaboo]

Gentlemen;

I was leading up to the idea of using the existing LM3886 as a driver for one or more pairs of TIP142/TIP147 output devices. In my own application I don't need more than 50w, but it is simple enough to use the transistors in series (to share higher supply voltage) and use a pair of MOSFET drivers as level shifters. And there ARE a lot of music/PA commercial amplifiers out there that do use these Darlington transistors. You'll never see them in my home stereo, but they are rugged and perfect for these bandwidth-limited applications.

 

[Gecko123]

well I had a chance to reground placing all ground points to the center of the copper bar, then removed (chg) grounds from boards, plugged in, and VERY loud buzzing/distortion from the same channel that seemed to fail in the first place, so the chip must be must be fried, guess I gotta get a new chip. And another thing, even after star grounding I still get hum out of the speakers, any other remedies for this?