(AK) classA HA/Preamplifier.

[Deleted]

I got a pleasant surprise Tuesday arvo with my boards arriving home from their holiday in Sydney with Alex to see what might be amiss with one.

Just to make things perfectly clear, Jeff's workmanship was superb, and although you can't see it in my photo, my shaky old hands couldn't quite match Jeff's workmanship. A first class soldering job too.
The fault that Jeff did not have time to rectify due to work commitments, was due to having to use local 2 pin terminal blocks that had pins not quite long enough to ensure that solder flowed sufficiently into the PCB hole which provided a through connection from the +VE rail terminals to the output stage.
I had similar problems with my own build, and ended up ordering terminal blocks from the U.K. to replace them. They had 4.5mm long pins compared to 3mm of the local terminal blocks.
Alex

[jeffc]

Aw-shucks Alex, you’re made me blush.

I can’t help myself, sorry, soldering neatness OCD.

Still, the bloody thing isn’t finished yet so plenty of opportunities for more stuff ups yet. And those shaky hands of yours are still doing a fine job, and no matter how neat I bet the amp of the wise old fox will sound as good if not better.

Board layout OK-ish or just suck it and see?

cheers..jeffc

[Deleted]

Hi Jeff
I can't see why there should be any problems, as you have the benefit of a much larger case than me.
With my 15W Class A power amp, I had to use an external supply for the amp to achieve a better S/N, so it made sense to also fit the preamp's toroidal(s) in there too. That then allowed me to use the smaller 1U rack case for the preamp/HA.
Alex.
P.S.
It may be worthwhile to fit upside down PCB pins where the output resistors are, to make it easier to change output resistor value for use with different headphones.You can then use a small soldering iron to tack the output resistors across the pins.
I couldn't fit pins there in my first build as the pins seemed too tight a fit. With my 2nd build,(slowly coming along) the pins could be fitted with a gentle push.

[Spirit]

Mar 5, 2011 5:27:34 GMT jeffc said:

I used a couple of Nichicon FG caps for the offset corrector because a friend had surplus to requirements.

Funny that, I have a few from what I suspect are the same batch sitting on my desk

[jeffc]

Hey Phil,
Long time no talk, must be using the same cap dealer.

And thanks for the tips Alex. All phones I have are 32 ohm, except for my ancient HD414s that haven't seen the light of day for years, but I suppose it would be nice to get that output resistor value set to my liking. Just need a stable USB-DAC source first that I was hoping for with the NFB-12 but now I'm stuffing about with low-pass digital filter settings for the WM8741 chips that markedly affect its presentation of music. More stuffing about now to sort this. I’ll learn to stay contented and not be tempted to tinker one life time.

cheers.. jeffc

[Will]

Jeff,

I've got to echo Alex here, your attention to detail is superb! Very nice work.

The only way to be sure about any hum from the toroids is to try (perhaps mounting them on a seperate plate initially) Do a test fire up and have a listen. Have a look at the scha thread, and you'll see loads of traffos sharing the case. With me, I had a separate traffo case to ensure that I did not have any problems.

Have you not had a cheeky little test listen?

[Will]

After a month or so with no music, my class A, buffalo and cd player are back up making music.



happy, happy, happy!

[Deleted]

Must be bliss Will!

Nothing at home for a while too
Neither is it near the top of "the list" yet. Double

[Deleted]

Mar 5, 2011 14:57:09 GMT Will said:

Jeff,

I've got to echo Alex here, your attention to detail is superb! Very nice work.

The only way to be sure about any hum from the toroids is to try (perhaps mounting them on a seperate plate initially) Do a test fire up and have a listen. Have a look at the scha thread, and you'll see loads of traffos sharing the case. With me, I had a separate traffo case to ensure that I did not have any problems.

Have you not had a cheeky little test listen?

The separate mounting plate is a good idea, as you can then connect mains earth from the IEC socket to that, without having it on the rest of the circuit.
BTW, a cheap and nasty method could be blank copper PCBs from Jaycar, with the copper side underneath, and the head of a mounting screw for the mains earth connection soldered to the copper side .Use short non metal standoffs to support the PCB.
( I hate hacksawing aluminium !)
Alex

[Will]

Mar 5, 2011 15:46:27 GMT @cjarchez said:

Must be bliss Will!

Nothing at home for a while too
Neither is it near the top of "the list" yet. Double

It's rubbish, isn't it? For all the fun to be had in diy, its the music that brings the real enjoyment.

[XTRProf]

Just to ensure the e-caps orientation, + marks on PCB at those e-cap profiles is the positive for the e-caps? Just don't want some volcanic or pyrotechnic show on startup.

Thanks.

[Deleted]

Mar 7, 2011 5:07:35 GMT XTRProf said:

Just to ensure the e-caps orientation, + marks on PCB at those e-cap profiles is the positive for the e-caps? Just don't want some volcanic or pyrotechnic show on startup.

Thanks.

"
Of course.Just double check and triple check your work.

[XTRProf]

Mar 7, 2011 5:12:43 GMT @sandyk said:

Mar 7, 2011 5:07:35 GMT XTRProf said:

Just to ensure the e-caps orientation, + marks on PCB at those e-cap profiles is the positive for the e-caps? Just don't want some volcanic or pyrotechnic show on startup.

Thanks.

Of course.Just double check and triple check your work.

Thanks for the assurance. See whether can put those in tonight if have time.

Double check before every soldering, yes, have done that ....... Will triple check all joints and spacing after completion of stuffing.

[jeffc]

Hi Will,
No lash-together listening tests done yet, sorry, but there will be soon. Twin transformers now mounted on an aluminium block with stand-offs for ease of attachment to the case. Now only case drilling, board fixing and wiring and I'll be done with an amp in a case. This amp must be special as mostly anything I build gets to live out its life on a pine board.

cheers..jeffc

[jonclancy]

Hey, this thread is now bigger than the Panda thread!!!

[Deleted]

Wish I understood it......

[Deleted]

This amp must be special as mostly anything I build gets to live out its life on a pine board.

Hi Jeff

Way to go

can't wait for you to hear it

take care

[Deleted]

Mar 7, 2011 12:26:11 GMT jeffc said:

Hi Will,
No lash-together listening tests done yet, sorry, but there will be soon. Twin transformers now mounted on an aluminium block with stand-offs for ease of attachment to the case. Now only case drilling, board fixing and wiring and I'll be done with an amp in a case. This amp must be special as mostly anything I build gets to live out its life on a pine board.

cheers..jeffc

Jeff
Remember for added safety , that the toroids mounting bolts should go back to IEC earth, but don't have that earth appear on the main case metalwork.
Regards
Alex

[XTRProf]

Feb 10, 2011 20:05:51 GMT Will said:

Setup

Check for dc on the output. Place a 2k7 or 3k3 resistor to the signal input terminals, and using a DMM set to mV, see what voltage is on the output.


Fine Tuning

If anybody is feeling adventurous, and they have a function generator that can output decent square waves at greater than 20KHZ and a CRO, they could experiment by reducing the value of the 68pf polystyrene compensation capacitor to 56 pf or even 47pf. You could also use a Silver Mica type there.Just ensure that there is no waveform overshoot (spikes) with a lower value.

Just to clarify, for the dc part, is it mandatorily to load the input with a 2k7 or 3k3 resistor? Any problem with no load as I will need to find or purchase a +/- 3k load.

Also, without going into signal generator and CRO, what's the possibility of having a good square wave without overshoot at 68pf? For those that had tried it, perhaps we can get a good idea just from this without even trying it. Of course, when we have the time, we can be more adventurous.

Thanks.

[Deleted]

Mar 8, 2011 8:30:28 GMT XTRProf said:

Feb 10, 2011 20:05:51 GMT Will said:

Setup

Check for dc on the output. Place a 2k7 or 3k3 resistor to the signal input terminals, and using a DMM set to mV, see what voltage is on the output.


Fine Tuning

If anybody is feeling adventurous, and they have a function generator that can output decent square waves at greater than 20KHZ and a CRO, they could experiment by reducing the value of the 68pf polystyrene compensation capacitor to 56 pf or even 47pf. You could also use a Silver Mica type there.Just ensure that there is no waveform overshoot (spikes) with a lower value.

Just to clarify, for the dc part, is it mandatorily to load the input with a 2k7 or 3k3 resistor? Any problem with no load as I will need to find or purchase a +/- 3k load.

Also, without going into signal generator and CRO, what's the possibility of having a good square wave without overshoot at 68pf? For those that had tried it, perhaps we can get a good idea just from this without even trying it. Of course, when we have the time, we can be more adventurous.

Thanks.

Chong
You could always do it with your volume control set to around 3K ohms between the wiper (centre) and the earth side of the volume control.
I posted recently that it gives excellent square waves as standard
with no overshoot, and even 100KHZ is still quite good, so I would leave the compensation capacitor at 68pF.
Alex

[jonclancy]

Chong, if you need some 3K3 resistors (matched, of course!) I can send you some in the post...

Cheers

Jon

[Deleted]

Mar 8, 2011 9:49:03 GMT jonclancy said:

Chong, if you need some 3K3 resistors (matched, of course!) I can send you some in the post...

Cheers

Jon

C.O.D. ?

[XTRProf]

Mar 8, 2011 9:49:03 GMT jonclancy said:

Chong, if you need some 3K3 resistors (matched, of course!) I can send you some in the post...

Cheers

Jon

Wow, I didn't expect that to be matched as well as that's not critical to sound. I really like your style, man! I will find some way to get that 3K. Heh, heh, heh, maybe joint a few resistors in series or what so as not to trouble you. I don't like troubling people when not critical.

Btw, are you very near to firing up your AHA yet? For me, I just need to match e-caps and put them in and then matched transistors and put them in. Maybe about 1.5 labour days away.

Oh, yeah, where's my progress picture? I had forgottern to take it.

[XTRProf]

Mar 8, 2011 8:45:13 GMT @sandyk said:

I posted recently that it gives excellent square waves as standard
with no overshoot, and even 100KHZ is still quite good, so I would leave the compensation capacitor at 68pF.

Ah great! Then it will be 68pf for the moment ........

[Deleted]

[Setup

Check for dc on the output. Place a 2k7 or 3k3 resistor to the signal input terminals, and using a DMM set to mV, see what voltage is on the output.]

In the case of this design without a DC servo being used this would make sense as the amp has a transistor input and 3k setting would be a 'mostly used' vol pot setting assuming 20k to 50k volpot is used.

The DC output voltage, however, is independent of the input resistance in this design and solely determined by DC servo in this particular amp.

You can use (matched !!) input shorts ....