New 200 watt Kit amp features ultra low distortion and Thermaltrak devices

Status
Not open for further replies.
Jump to Latest
#22
Bonsai said:

2. I like to run my amp 'Class A+AB' - so first 5 watts is class A and then it goes into class B (my new iteration will run first 10W class A)
3. Yes, the distortion is higher (just like Doug told us), but this only happens >5W
4. What you get is a sweet sounding 10W class A amp with **** loads of high power class B sitting right behind it. Sounds great.
Click to expand...

Bonsai,
a 4pair output stage using Re=0r15 and biased to Vre=25mV gives 7W of ClassA into 8r0 and is still optimum ClassAB.
Going to 5pair or Re=0r12 gives the full 10W of your ClassA target and satisfies the optimum ClassAB without the GM doubling/distortion increase.
 
#23
smithy666 said:


You should try replacing these - you may be surprised by the result. The fact that it has to handle a fraction of the current is irrelevant. You should do some research on various types of capacitors. Elektor did an article some years ago where they tested each type and provided measurement results (ie objective testing). Electros were almost the worst (only beaten by tantalum caps), and bipolar caps are worse than standard electros, particularly when used in the signal path.

Most high end speakers use polypropylene caps for exactly this reason. Yours probably don't as they cut corners and didn't want to spend the extra $$$ it costs for better quality caps. But hey, if you are happy with those used in your speakers, then no problem. And if you are happy with the circuit/components proposed by Silicon Chip, then no problem either. But you did post here, so I gave you my 20 cents worth, particularly given I have been reading Leo for the past 25 years and know exactly his views of electronics and the world.
Click to expand...

Do they make 300uF and 2000uF polypropylene caps ??

Also I thought you guys didn't take any noticed of measured objective results ??

I just spent a whole thread on the subwoofer forum listening to audio gurus trying to tell me that 30% THD level in a subwoofer was not really relevant to the sound and now you're telling me that this seemingly innocuous coupling capacitor is so much more important :xeye: In any event you can always replacement it with one that you think is a lot better 😉

As well aren't those tantalum caps the same ones used in those English Naim amps that all of the die hard audio gurus swear by ??
 
#25
snoopy said:


Do they make 300uF and 2000uF polypropylene caps ??

Also I thought you guys didn't take any noticed of measured objective results ??

I just spent a whole thread on the subwoofer forum listening to audio gurus trying to tell me that 30% THD level in a subwoofer was not really relevant to the sound and now you're telling me that this seemingly innocuous coupling capacitor is so much more important :xeye: In any event you can always replacement it with one that you think is a lot better 😉

As well aren't those tantalum caps the same ones used in those English Naim amps that all of the die hard audio gurus swear by ??
Click to expand...

Well Solen certainly do make a 300uF polyprop cap for crossovers. I've never seen a 2000uF cap used in a crossover, so you've completely lost me on this one.
I didn't say anything about your 30% distortion subwoofer, nor did I say that I ignore objective results. What I did say is that it is actually possible to measure capacitor characteristics in certain applications, and electros in audio signal paths perform poorly compared to other types.
I think you should find somewhere else to relieve your frustrations, which you seem to have a lot of. You certainly aren't interested in opinions (whether based on science or not), only your own. Goodbye.
 
#26
smithy666 said:


Well Solen certainly do make a 300uF polyprop cap for crossovers. I've never seen a 2000uF cap used in a crossover, so you've completely lost me on this one.
I didn't say anything about your 30% distortion subwoofer, nor did I say that I ignore objective results. What I did say is that it is actually possible to measure capacitor characteristics in certain applications, and electros in audio signal paths perform poorly compared to other types.
I think you should find somewhere else to relieve your frustrations, which you seem to have a lot of. You certainly aren't interested in opinions (whether based on science or not), only your own. Goodbye.
Click to expand...

Infinity Kappa 9's. Check it out.

Also if you don't like the idea of a bipolar electro coupling cap then just substitute it with a polypropylene or whatever. Surely it is not a big deal and it doesn't invalidate the design.
 
#28
Quote

Bonsai,
a 4pair output stage using Re=0r15 and biased to Vre=25mV gives 7W of ClassA into 8r0 and is still optimum ClassAB.
Going to 5pair or Re=0r12 gives the full 10W of your ClassA target and satisfies the optimum ClassAB without the GM doubling/distortion increase.


I am biased just a little higher than this - I've used 0.22 ohms Re and running at a V drop of 33mV. On the new version I have added more output device pairs and will run these at slightly lower Iq to keep them in the 25mV range.
 
#29
Bonsai said:
Quote

Bonsai,
a 4pair output stage using Re=0r15 and biased to Vre=25mV gives 7W of ClassA into 8r0 and is still optimum ClassAB.
Going to 5pair or Re=0r12 gives the full 10W of your ClassA target and satisfies the optimum ClassAB without the GM doubling/distortion increase.


I am biased just a little higher than this - I've used 0.22 ohms Re and running at a V drop of 33mV. On the new version I have added more output device pairs and will run these at slightly lower Iq to keep them in the 25mV range.
Click to expand...

Yes. For maximum linearity, the Iq is NOT important. Important is the 25mV. The Iq then follows from whatever Re you want to use.

Jan Didden
 
#31
Provided the driver and pre-driver do not fall into ClassAB on peak current output it does not matter whether the pre-driver is used or not.
The ClassAB bias applies to the output devices and should be in the range 15mVre to 25mVre.
But, there is a problem. Some modern devices have a built in Re to balance the parallel transistors contributing the "whole" device. How do you allow for this internal Re? Distortion measurement and set up appropriately (with a lower Vre)?
Note that D.Self showed that Vre for various Re are all around 50mV. This figure is for the total voltage drop across both emitter resistors. Even halving his values, he ended up with Vre~=25mVre and not towards the lower 15mV end, that many on this Forum have suggested.
 
#32
Andrew, some time back I played a bit around with some Rotel models which use the T circuit. In these the vre is usually set at around 4 / 8 mv. I found that increasing it to about 10 to 12 mv is sometimes beneficial but anything over that increases distortion, 25mv ???, not taking into account heat dissapation.
 
#33
is it possible that the Rotel had some current limiter to protect the output stage.
The extra mV across the Re would be interpreted as extra dissipation and thus trigger the protection early. Just guessing.

BTW,
4 to 8mV is more typical of a CFP output stage.
 
#40
janneman said:
Multiple unmatched devices only increase the chance that the N- and P- side get better matching.
Jan Didden
Click to expand...

Jan,
reminds me of this power amplifier
Using 44 pieces of LM6171 op-amps as output per channel!
Opamps are an ideal solution, but their current capabilities are too limited. I, therefore, placed 44 of them in parallel.

To drive them in pure class A would demand a high DC-current (per opamp) and increase power dissipation.
I decided to inject only a relatively small DC-current, so each opamp works in class AB.

By using different current values for the various opamps,
each opamp will switch at a different overall current demand.

At any time of operation, the major part of the output opamps will be in a non-switching state, and the "control-opamp" (which is working in class-A) is able to control the output signal continously. TIM-distortion is eliminated, although class-AB functionality is used.

In contrast to a conventional class AB device, where there is no control during switching,
there always will be "control" using many parallel output stages,
each switching at different points.
That's why I called it semi-class A.
Click to expand...

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


http://headwize.com/projects/showfile.php?file=meier3_prj.htm
 
Status
Not open for further replies.
Top Bottom