LME49810/STD03 parallel pairs

[jan.didden]

I am agree with you about the sound quality with more bias..
Just I'd like to try, but I am afraiding to destroy the diodes, because Sanken gives 2.5ma for the maximum Ibb level.. Anyway, I guess it would be better to keep it at 40mA bias (or 60 like yours)..
What about the sound quality of your amp? Have you ever compare it with a gainclone or a good discrete amp?

You don't change the 2.5mA when you change bias current, you change the pot and measure the bias across the emitter resistors of the output devices to get 40mA through them (or 60mA).

As for the sound of my amp, see last post here:

http://www.elektor.com/forum/elektor-forums/recent-main-projects-and-articles/pax-(april-may-2008)/invitation-to-discussion.432667.lynkx?highlight=paX&pageStart=21

Edit: paste the whole link in your browser.
Jan Didden

 

[Dxvideo]

I think there is a problem with your "whole" link.. Because my IE says "bad request".

 

[peranders]

Click on the link and then add the "black" text, then you'll get it right.

http ://www .elektor.com/forum/elektor-forums/recent-main-projects-and-articles/pax-(april-may-2008)/invitation-to-discussion.432667.lynkx?highlight=paX&pageStart=21

Remove the spcaes.

 

[Dxvideo]

Hmmm...
May you share your design with us? So we can make a comparison may be..

 

[jan.didden]

Hmmm...
May you share your design with us? So we can make a comparison may be..

You can download the complete article series from the Elektor website for a small fee.

Jan Didden

 

[Dxvideo]

Ok.. This is a way too!

 

[jan.didden]

Ok.. This is a way too!

I'm sorry, they published my article, they hold copyright.
It would be very bad of me to go around their back.

Jan Didden

 

[klewis]

An externally hosted image should be here but it was not working when we last tested it.

Dear Ken,
This is my application schematic. However PCBs are still in manufacturer. I am expecting to finish my amp in next week. So I will report the results after the first tests...

Dxvideo,

Did you get your amp up and running?

Ken

 

[sasmit]

This is a bit off topic..but has anyone tried a CFP output stage with the LME 49810. I am working on a schematic with MJE 340 and 350 driving stages and
2SA3281 and 2SA1302 in the output stages...

 

[Dxvideo]

You don't change the 2.5mA when you change bias current, you change the pot and measure the bias across the emitter resistors of the output devices to get 40mA through them (or 60mA).
.....
Jan Didden

May be it will be a bit late discovery! But are you sure about that? Means, how could you keep the 2.5mA while changing the pot value while it depends on this pots value!
HOW?

 

[jan.didden]

May be it will be a bit late discovery! But are you sure about that? Means, how could you keep the 2.5mA while changing the pot value while it depends on this pots value!
HOW?

That is what I have been trying to say. You, as the designer, must make sure that the current is 2.5mA, for instance with a current source. That is the current where the thermal tracking is correct. Then you fiddle the pot to set the output stage bias current. But the diode current should be 2.5mA nominal.
Doesn't anybody read data sheets anymore???

jd

 

[Dxvideo]

Ok... So you (or Sanken) say if I set the diode current around 2,5mA the bias should be ~40mA but I can still adjust it an exact value while the diode current is still ~2,5mA!
Is it?
By the way, I was already read the manual (both STD and SAP) however this point is not good explained..
Thx.

 

[Sebastiaan]

Dear DXvideo,

THe bias forward current should be fixed on 2,5mA. With the diodes of the STD/P and STD/N in series the voltage drop over the total diode string should be 2245mV. This is an easier way to verify the bias current by measuring the voltage.

With kind regards,
Bas

 

[AndrewT]

Dear All!
I want to buy an amplifier with LME489xx driver one or two pair SAP15s. What LME chip I will choose? LME49810 or 49811?
Thanks.

I am currently working on a project with SAP15s and LME49811. I think 49811 is the best solution for SAPs. Because they already have a biasing systems and do not need Vbe multipliers. So if you consider 49810 and 49830 has bias pins then 49811 (or LM4702) is the best for SAPs. Another point is 49811 has Class A driving stages..

I don't agree.
The 811 is designed with an output stage that suits low current demand.
The 810 is designed for about seven times as much current demand and better suits a two stage (Darlington) output pair.
I think the 811 needs a three stage output pair for anything that demands output current, i.e. any speaker rated at less than 8ohms.

 

[klewis]

Andrew,

If the driver had a hFe min of 160 max 320, would that change your mind?

Ken

 

[AndrewT]

Hi,
work back from the valid audio signal maximum peak transient current.

Let's for sake of this example assume +-50Vdc supplies and an 8ohm speaker.
Peak current ~[50-3] / 8 / 0.35 ~16.8Apk.
If you assume 4 to 8ohms speakers then the transient maximum exceeds 33Apk.

There is some evidence that the 0.35 factor gives too low a result for some music types and some speaker types. I have seen some claiming 0.2 as a worst case factor.

 

[klewis]

If the LME49811 makes .007A, and the driver makes min. 160 HFE, that's 1.12A, then if the output makes min. 55 HFE, that's 61.6A. Now, my knowledge and math may be all wrong, that's why I posing the question.

Ken

 

[AndrewT]

find that transistor that has a gain of 55 for an Ic=61.6A.

I said

work back from the valid audio signal maximum peak transient current

to find the current requirement at the preceding stage

 

[klewis]

This is about needing predrivers with the lme49811 when driving 4 ohm loads. If you pick a driver with a high hFE you won't need a predriver. You will need multiple output transistors to keep within the soa and the will need decent hFE for the current driven.

 

[AndrewT]

let's try doing it in steps.

What output stage do you want to use for a 4ohm capable amplifier fed with +-50Vdc supply?