Noir, a two transistor headphone amp: class-A, single ended, 150mA bias

PRR

Member
Joined 2003
Paid Member
single end klasse-A, so the rendement is only 12,5%.

Off-topic: this is resistor loaded single-ended and the efficiency can't be much over 8%.
Efficiency-Trail.gif

A detailed analysis by "kvant" gives ~~8.58..%.

The 12.5% figure is cited many places, even (at least two) writers who should know better (or can rough it out as I did above).

This is not new. This result appeared in a letter in Wireless World from John Vanderkooy in 1969 (though he does not give the number):
http://www.americanradiohistory.com...1969-08-OCR-Page-0089.pdf#search="vanderkooy"

responding to-
http://www.americanradiohistory.com/Archive-Wireless-World/60s/Wireless-World-1969-06.pdf

Vanderkooy's critique of "CirCard" info:
http://www.americanradiohistory.com/Archive-Wireless-World/70s/Wireless-World-1973-12.pdf
page 78 of the PDF

disputes Carruthers et al, June 1973 page 79 of PDF
http://www.americanradiohistory.com/Archive-Wireless-World/70s/Wireless-World-1973-06.pdf
 
So, I tried to put together the Noir, using the PCB from Scotsky.
Unfortunately, it did't work.
Had one part mislabeled (apart from a couple of other faults that Scotsky warned me about).

In particular, the E153 current regulating diode was marked the other way around and didn't notice until power up...

When i did, BC516 Darlington transistors started overheating almost immediately.
And when checking voltages, bias voltage was zero and remained zero even when turning the pot clockwise.

Also, a weird symptom was a pretty serious drop in supply voltage.
Instead of measuring -24VDC right at the NEGPWR spot, I was getting something like -14VDC instead.

Even after reversing the E153 current regulating diode to match the schematic correctly, little changed.

I suspect because of the CRD being mislabeled i've fried the Darlingtons and possibly the E-153 too.
Hope the Mosfets are OK though.

Anyway, would you please let me know what you think folks?

Should i try it once again with fresh BC516 and E-154?
Or get a couple of STP16NF06 just to be sure?
Between witch points you measure those 14V?
I'm have it connected here now, and can do measurements if you like.
Just tell me where 😺
 
Scotsky, check your diodes orientation.
E-153 should be with the cathode (ring) side looking towards NEGPWR.

In the PCB you sent me, they are marked the other way around.

That's why I think it blew both the diodes and the Darlingtons.
Also probably the reason bias remained at zero (Mosfets couldn't begin conducting?).
 
Yes, I forget to mention them. Apologies again for that.
On the build, you can see that I soldered them in the correct way.
I must make an extra note of that, for eventually future builds.

What about the measurements, I thought you needed extra information, about different voltages on different places for reference.
I hope you'll get it soon right.

Yesterday evening I've used it with a proper build powersupply and pre-heated.
The the difference was great compared the first try's I made with a bad supply.
I totally forgot the time, when listening, so it was late last night 😹
 
I am waiting for the Darlingtons and the E-153 diodes to arrive.
Just to make sure, I ordered the Mosfets as well.

As I stated in a previous post, the E-153 was soldered the other way around.
The cathode (ring) side of the diode was unfortunately not facing NEGPWR.

So my question is:
At power-up, would that be sufficient reason to blow the diodes themselves and/or the Darlingtons?
 
Folks, it seems this kit doesn't want to cooperate.

The reordered parts arrived (E-153 diodes, Darlingtons and the MOSFETs).
I placed the E-153 diode the correct way around but when I tried to fire it up, a wierd thing happened (again).
Using a typical LM337-based PSU to provide negative power (-22VDC), I got a drop of 10VDC across the Connection Points.

In the following picture, you can see the where exactly is connection point on the 'revised' schematic:

1675955187762.jpeg


The X-marked parts to the left and to the right are left unpopulated.

Is it normal to measure a drop of around 10VDC (I measured -12VDC) right across the Connetion Points?

Thanks in advance!

PS:
The PCB has been thoroughly checked point by point and agrees with the original schematics.
 
Because you used my PCB, I can send you high-res and more detailed pictures of mine if it helps?
(But I thought they were allready sharp enough.)
It works like a charm.
I only left out the L1, because I made a mistake with the powersupply and blew it, so, I shorted it.
How many current does the circuit use now?

But that wil be tomorrow.