The Mini-A

Hello!

I changed the resistors on the board for Vishay CFM 1%. All 220R in the circuit are now 221R too. The difference is very visible especially in the high frequencies. It is not yet perfect but much better already.

At the moment, I haven't touched the capacitors.

On the other hand, I have an offset of 56mV that I cannot lower. I installed a cermet provisionally for testing in place of R3.

I think I'll have to take the two IRF9620 apart to measure them. Do you think this is the first thing to do next?
 

Attachments

  • 100Hz-new_resistors.jpg
    100Hz-new_resistors.jpg
    51.8 KB · Views: 282
  • 1KHz-new_resistors.jpg
    1KHz-new_resistors.jpg
    50.2 KB · Views: 270
  • 10KHz-new_resistors.jpg
    10KHz-new_resistors.jpg
    53.5 KB · Views: 259
Last edited:
Hello!

I removed the two IRF8620 on the heatsink. They are not well matched, 3,616 and 3,545.

I matched two fresh IRF8620 and put them in place. With a 221R resistor, I got an offset of 25mV. With the 25 turns cermet, I can goes down around 4mV. Just adding some mOhms.

I also changed the 1nF capacitors, replaced the 39nF with a 47nF and replaced the 2 power resistors of .51R with .47R.

The curves are better. At 1.5vPP, I start to have distorsion.

Seems difficult to do more.
 

Attachments

  • 10KHz_1vPP.jpg
    10KHz_1vPP.jpg
    47.9 KB · Views: 228
  • 10KHz_1.5vPP.jpg
    10KHz_1.5vPP.jpg
    49.3 KB · Views: 96
Last edited:
Hi Zen!

I think I found the problem. One of the new IRF9620 has burnt down (Q2). A fresh one bought at mouser. Bad luck, he's part of the pair that I matched. I no longer have IRF9620 to remake a pair. No Luke. I have to re-order some.

In the meantime, I used the second pair that I intended for the second board.

The last thing I did before the blackout was to change the C1 and C3 220uF capacitors. When I switched on again, it broke down.

While I'm at it. I did not understand the management of power resistance. Currently I have .47R for R21 and R22 and 0.33R for R18 and R19. For a BIAS of 1A, what are the right values?

The last schematics with new parts : PASS-AM-OPTIMIZED - EasyEDA open source hardware lab
 
Last edited:
moving my question from another thread where we were talking Alephs so I got a thought:

I have Aleph30 pcbs fully assembled and collecting dust (they were used only for a year). Next Winter I hope to be making a mini Aleph to drive the mid/top section of my new speakers.

I was planning on making AlephJ with SemiSouths but the speaker section will be 4ohm impedance. When I see amp power output into 4ohm dropping below that which it does into 8ohm I kind of have a feeling that the amp is not really suited to drive a 4ohm load even when it comes to distortion (wrong?).

I also remember that the midrange was very nice; (bass handling was not that good). So I thought rather than selling off those A30 boards maybe I modify and try them in a miniA: What do I need to change?

As I remember my original Aleph30 ran on +/-25.5V rails and had 3 output fets biased at 0.7A each. On say +/-17V rails do I only pull two out and re-bias the one left for something like close to 2A and then reset AC gain or is there more to it? Thanks!
 
Official Court Jester
Joined 2003
Paid Member
Alephs are having more power in 4R than 8R, under condition that they aren't current starved

see factory user manual

newer Papamps - with Mu Follower for OS, are tending to be wimpier on tougher load, matter of choice
though, some measurements I did are saying that even those aren't too shabby regarding tougher load cojones

what was exact schm you did build your A30, with subpar bass capability, and what speakers were driven?

as I see it , now is main thing to find reason why you found that deficiency, and to see proper way of solving that ........ was it mistake in construction, or you're simply having taste for something else
 
Last edited:
Member
Joined 2016
Paid Member
Hi koja,

the ability of an Aleph to supply more watts into a lower impedance all has to do with the bias current.

I built a Mini Aleph on BrianGT boards and have been running these with a single pair of IRFP140 at about 19V and 2 - 2.2 amps (big case, similar to 4U/400). That's why I don't call it Mini, but "Aleph Single".

So, I think it really is mainly just removing two pairs from the board.

With regard to the Aleph current source: First, I did use 0R33 for the source resistors of the IRFPs and 2 pcs. 0R33 in parallel for the current sensing resistors; later 0R22 for the source resistors and 0R11 for the current sense.

So, from what I gather from the various schematics, the current sense resistance should be half of the source resistance.
You may need to adapt your current sense resistance when you remove two of the three pairs, and best measure the actual current gain when you have done the amp conversion.

I have attached a few files that might help you in setting your parameters. First, a schematic of the Aleph Mini as realized in the BrianGT boards, then, the still "reference" Aleph 30 schematic by algar_emi, and then a calculator for power in various load impedances that helps setting the parameters (last from wuffwaff).

Hope this helps, have fun with the conversion,
Claas
 

Attachments

  • mini-A schematic.JPG
    mini-A schematic.JPG
    58.3 KB · Views: 303
  • Aleph30 by Algar_emi.pdf
    29.9 KB · Views: 115
  • Aleph-power-xls.zip
    12.4 KB · Views: 94
I have to check my files but I am pretty sure that I used the same schematic by algar_emi.

When I said less power into 4ohm load I meant AlephJ. That is what made me think that I should probably reconsider a mini version of regular Aleph30 and not of AlephJ with SemiSouths (maybe save those for J2 or whatever).

I am planning on the 4U/300mm housing (like the vfet amp housing which soon will be offered in the store with a new face-plate with vertical lines and no markings so it can be used for other smaller amps). I think it should be able to deal with ~60W dissipation per channel.

Thanks for the miniA pointers.

p.s. as for the bass: I used Aleph30 with Elsinores. Midrange fantastic, warm voices but flabby (not tight) bass. When I put BA3-bal monoblocks on Elsinores I could appreciate the absolute control the amps have over the speakers top to bottom. However the same monoblocks for example struggle with LX521s which sound a bit overdamped. When I put the AR D150 amp on those speakers they got a new life (but that amp has a buzz which I am still to address by changing output tube bases etc.). So horses for races I guess. There is a conceivable gain to be had in matching gear. When one adds room and speaker placement to the equation the hobby becomes even more interesting.
 
Last edited:
After many, many years, I might finish my mini aleph project.

I have a set of BrianGT's surface mount boards, with resistors installed. Need to order a few parts and I can put it together.

In resurrecting this project, I found some one files on my PC, they are attached to this post.
I also annotated a schematic to add some notes for TBD resistor values, and resistors to change for bias and dc offset (or to add pots instead).
Also attached routing I had for both SM and TH boards.

There are a couple drawbacks to the SM boards. One is that the bias pot wants to install on the bottom of the board. I'll see if I can figure something out there, I've drilled holes in PWBs before.

Other is that it only supports one pair of output devices per channel, the TH board lets you use 3 pairs.
I would like to bump the bias up a bit, seems like it wants to be at least 1.2A from what I gather.

So better late than never, maybe this info can still help someone.

Randy
 

Attachments

  • miniAleph-sch v01.pdf
    193.7 KB · Views: 115
  • miniALEPH-parts_list.zip
    17.3 KB · Views: 62
  • smd-miniA.gif
    smd-miniA.gif
    94.1 KB · Views: 123
  • miniA th.gif
    miniA th.gif
    107.8 KB · Views: 122