DIYHiFi.org

The pre I'm using is the prototype I made almost two years ago, with the SMD AD815.
To take a pic of it would mean disconnecting my whole system and opening the preamp.
This pre has given me lots of work to make, it has 6 inputs, 3 tape loops and two preamp outs (one with the caps - connected to my sub, the other is direct).

Some time ago I've got the non-smd chips and made another prototype board, trimmed dc-offset, tested on my bench but didn't hear it yet on my system. I've been doing other things, I'm very happy with my pre and it's not easy for me to convince myself to fiddle with it again.
Here's that second prototype, non-smd chip. The heatsink is oversized, this doesn't even get warm. This one was meant to be tested with batteries.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein

Another pic.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein

Very nice Carlos
We don't expect you to go taking your main pre-amp to pieces just to take some pictures but at least these give us and idea how you laid it all out.
Noticed you got a couple of Rubycon ZL's in there

Yes, I like them too.
They are 470uF/25V.
As I said, this was to be tested with batteries.
You didn't spot the snubbers?

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein

Snubbers, sorry about that mate
So I should stick your regulator snubber version in mine

Yes, 0.47R + 22uF right after the output cap on the LM317/337 regs.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein

Thank you Carlos

I'm just having a look to see if I can find any matrix boards, I've most of the other needed bits and pieces, the only other thing is the pot, I've got a 20k Alps black I could use for now

Hi Carlos,
good to see your posts of your pre here.
I doesn't look to complex to build and test by me. I will definitely do that but will have to order a couple of chips first.

Pedja's question regarding if you tried any other current feedback amps for a pre is interesting is it is own. What part of the sound characteristic you describe is due to the current feedback type of opamp and what part because of the AD815.

I wonder if a better power supply (i.e an ALW-regulator) would improve the pre, as its PSSR isn't that impressive.

Henk

carlosfm
If you were to use a phono stage on your AD815 pre, how would you do it? Six inputs and three tape loops are alot of connections to make so I am also very interested in how you lay it out and select the sources. I need a fully functional preamp that includes a DAC and phono stage so I really need to be schooled on how to lay it out.

That is fine, I'm using a 20k Alps Blue on my pre.
Just don't go above that, 10k~20k is fine.




Pedja asked about the use as I/V, but anyway, the LM6181 is good but inferior sounding too, used on a linestage.
I also tested the TPA6120A2, also current-feedback, which has impressive specs, but the AD815 sounds better, or to be more precise, much better.



That's not an easy question to answer, but besides current feedback and a high quescent current, the massive output current (even if not 'needed' here) pay it's dividends.
Dynamics is fantastic, a diferent league, really.
We don't know the exact discrete parts this chip has inside, but it surely works wonderfuly for audio aplications.



I'm sure it does.
The PSU on the schematic is what I used back then (and is still there). The unreg. PSU part is schottky diodes and RCRC.
A better PSU is always an improvement, and I have super-regs here and didn't try it on the pre yet. Some did and reported very good results.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein

Hi I am using this layout:
http://www.marklev.com/JC2/jc2schematics.jpg

The best thing you can do is switch the grounds too.
That requires a 3-pole switch.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein

Aha! I stand corrected. Never bothered to look up what this beast of a chip was, assumed it was closer to AD811. Indeed, using a near power amp chip for linestage is not exactly the norm.

It's all coming back to me. I used the AD816 (DSL driver) as RF amps for driving a mass spectrometer. Amazing output drive with those chips. I was pushing 16MHz into some nasty loads, the chips never had a problem.

I'm curious, you have 68uF in parallel with the 33uF on the reg. Or is there ferrite beads or resistance between them? I wonder why the 33u value is so critical when in parallel with the 68u. Also, doesn't the type or construction of cap make more of a difference? Are you using organic electrolytics or tantalums or what? Special low ESR? I'm thinking maybe it's the low ESR on the regulators that can make them go unstable?

jh

_________________
Jim Hagerman

Elso Kwak
Yep, it's hard to fault anything by John Curl. Thanks for the schematic now for the layout :(

Jim Hagerman
Since you seem to be familiar with the AD811, would you tell me how suitable you think this chip would be for a phono stage? As I recall it has an excellent gain to noise ratio so if you can make it RIAA then it should be a pretty good candidate for a MC input. What do you think?

No Jim, no special caps, I'm using Philips electrolythic caps, which are fairly low ESR, decent caps, but no BGs.
I talk from past experience with LM317/337 regs, and also the LM338 (Pedja has the same experience with this reg), and even standard fixed regs (78xx/79xx, etc), big caps on their outputs sound bad.
If for instance you put a 1,000uF cap there, you'll notice right away, bass slows down and treble veils. It's not so bad, but you can already notice quite some difference with 100uF.
These regs have enough output inductance, and consequently lowsy rising impedance towards the high frequencies, you don't want to add more inductance there with big caps.

If the regs are on the same board of the chip, and near it's PSU pins, those 68uF could be discarded. But they had to be really near the chip.
These regs are more sensitive to the cap that they have right on their outputs than other caps at some distance.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein