Aksa 55 testing: output offset

0 Members and 1 Guest are viewing this topic. Read 1754 times.

simbloke

Aksa 55 testing: output offset
« on: 11 May 2003, 01:37 pm »
Hi,
I am testing my Asksa 55, everything has gone ok so far, all measurements have been within the expected range.

I have just got to the point of adjusting the bias and then measuring the output offset. The instructions (section 3.1, step 5) say the offset should be less than 30mV. What about a negative offset?

One of my modules measures  -17mV and the other -38mV. In this measurement is it the absolute value that matters or is -38mV less than 30mV?

I have made the measurements listed in section 5.2, they are within range. Some are at the high end - the supply rails are at 39v.

Regards,
Simeon

P.S.

Paragraph 3 of Section 5 (Diagnostics) mentions R25 and R26, should this be R21 and R22?

AKSA

Aksa 55 testing: output offset
« Reply #1 on: 11 May 2003, 10:41 pm »
Hi Simeon,

Thank you for your post;  good questions.

Offset is the voltage appearing at the output with respect to earth.  It can be either positive or negative, so we can talk simply about the magnitude.

Less than 30mV, positive or negative, is the requirement.  The amp will still function at +/-200mV, and sound almost identical except for a permanent displacement of the speaker cone to left of center, but to be absolutely sure the offset limits are quoted right down to 30mV.

If it's 38mV, it's too high.  You will need to trim R3 to bring it down.  The rule is this;  a negative offset indicates too much stage current through the diff pair;  we must therefore increase the value of R3.  The easiest way to do this is with a variac, but most people don't have one.  I can tell you that increasing R3 by 4K7 from the present 68K will do it.

You are right about R25/26;  these are the fuse resistor designations for the 100W AKSA, not the 55W!  I will amend the instructions;  my thanks for pointing it out.

Cheers,

Hugh

peranders

Aksa 55 testing: output offset
« Reply #2 on: 12 May 2003, 07:49 am »
I myself prefer DC servos but have for a long time had less than 50 mV as value which is not bugging or harmful but off cource the limit is floating.

The disadvantage of the offset volatge is clicks when the speaker is switched on and off but the very high offset causes distortion due to the displacment  but this is depenent of the speaker (bigger speaker less problem) but I have no idea how much would be harmfull. 200-500 mV?

rwalton

While we're on the subject of output offset......
« Reply #3 on: 12 May 2003, 03:01 pm »
Just a couple of additional questions:

1.  Should one strive for zero output voltage offset, or is <30 mV "good enough"?

2.  If one has significant variability in their mains voltage, won't there be a corresponding shift in the output offset?  For example my mains varies between 118 and 126V.  Is it easy to calculate the effect this would have on the output offset (I have a 55 Nirvana and 100 non-Nirvana)?

Just wondering........

Rick

simbloke

Aksa 55 testing: output offset
« Reply #4 on: 12 May 2003, 04:06 pm »
Hello, thanks for your advice Hugh.

I connected a 10K preset in series with R3 to find a good value. I notice that the offset falls for a minure or two after switch on. In the end I used 10K on one module and 3K3 on the other.

At switch on both have an offset of around 20mV but this soon falls and then hovers around zero, +/- 1mV. I figured that was good enough!

Well, I'll soon be biting the bullet and plugging something in to it  :)

Regards,
Simeon

AKSA

Aksa 55 testing: output offset
« Reply #5 on: 13 May 2003, 12:44 am »
Hi Simeon,

thank you for informing us of your positive results.  Your values are fine, and likely reflect, like much of the US, the higher mains supplies we all use these days.

Rick, you are quite right;  offset will rise and fall with mains, but if you set the output offset to 1 or 2 mV at the median of the mains voltage, then the output offset will in almost all case fall within the 30mV spec.

There is no sonic penalty at these levels of offset with any real-world speaker driver I know of.

There is a relationship between applied voltage and offset, but I've never actually divined what it is.  I do know that adjustment of R3 by around 10% of the nominal value fixes all these issues.

Cheers,

Hugh

Hope this helps,

Hugh