DIY ACA mini

So if you run in parallel mode then you get more current capacity but not more voltage swing so no more power unless getting a tad extra current above any single channel current constraints that might be present in the original. Or maybe you can drive lower impedances? I don't see parallel mode as much of a benefit, Especially for the cost per channel. Please, please correct me if I am off on this. Thanks.
 
  • Like
Reactions: 1 user
Hey guys
Is your bias stable on the ACA Mini
Mine seems to drift a lot
I can have the unit on for an hour or 2
And check and set the bias
Say at 295mv and have 11.51 dc offset
Then check it again in a few hours
And it may be at 230mv with 11.58 offset
I have built a lot of amps never had a bias that was this unstable
Any thoughts?
 
Nota Bene: Bias vs Temp in this post: https://www.diyaudio.com/community/threads/diy-aca-mini.379037/post-7159532
Note that I made an error in labeling the Y-axis in that post (need to multiply the bias by 10 to fix, or see the updated plot below).

That plot is for one particular setting of the pots, so the absolute values will differ for your particular setting, but the variation with temperature should be similar. That means a room temperature variation of about 5°F would give the change of 230mv to 295mv you have seen.

Bias_vs_Temp.png
 
  • Like
Reactions: 1 user

6L6

Moderator
Joined 2010
Paid Member
It’s a very simple circuit with no thermal compensation, and it lives in free air, so there will always be more variation of temperature then something inside a box.

The good news is that it doesn't matter one bit, as this is a single-rail design with a properly-sized output coupling capacitor, so the actual DC offset at the speaker output will be zero.

Which is pretty neat. 😎
 
  • Like
Reactions: 1 user

6L6

Moderator
Joined 2010
Paid Member
So if you are setting .3 at startup, and it lowers over time as everything comes to thermal stability, you could try setting it higher cold, knowing it will come down a bit, or setting bias once it’s been on an hour and warm, knowing it will read higher when it’s cold at startup.

…which is two ways of saying the same thing.


Anyway, proper bias at normal operation is the goal, so tweak as necessary to achieve that.

Bias (in)stability as the amp warms is not a big deal, as this isn’t a DC coupled design and therefore the coupling cap will keep any changes in output offset safely isolated from the speakers.
 
  • Like
Reactions: 1 users
I wonder if putting the Fets on bigger
Heat sinks it would make the bias more stable because of holding ambient temps better?
More thermal mass could slow down responses, but not change the end point.

If you would like to test if the bias changes you are seeing are due to ambient (room) temperature changes or to something else internal to your unit, you could place a room thermometer next to your Mini and take paired temperature and bias readings through the day. If it is the ambient temperature changes, you will see the type of variation I have.
 
  • Like
Reactions: 2 users
Help a brother out?

I was soldering along on my ACA Mini and knee deep in Led Zeppelin III when I realized that I mounted one of the J74s incorrectly. I am a professed neophyte with regard to DIY audio projects and broke one of the pins rendering the FET useless. It looks like that mistake is going to cost me (experience is what I get when I don't get what I want).

I see that the DIY store has matched FETs for sale but I need some guidance as to which offering to select. As you know, the kit came with only 2 of these but it seems that I might have to order 4 or more. Will the Toshiba 2SJ74 work for my needs?