[Uncle Bob]

Quote:

Originally Posted by tangent /img/forum/go_quote.gif Half correct. The fact that it isn't drawing lots of extra power means it probably isn't oscillating wildly. It could still be misbehaving. The fact that it sounds good mitigates against that possibility; I'm just being picky. Most likely, that extra heat is just due to the higher Iq of this chip.

thanks for that feedback, the ad8397 was bought from you recently (just as an impulse purchase when ordering some other parts

) and the amp is currently being powered by a TREAD based supply and both 'upgrades' have improved the performance of the amp substantially.

(now I've just got to get round to building that PPA v2)

 

[SmellyGas]

Quote:

Originally Posted by Pars /img/forum/go_quote.gif The blue box caps would be read as u22k63 = 0.22uf 63V and u1j63 = 0.1uf 63V I believe. I would not replace these with electrolytics. The 0.1uf appear to be used as power supply decoupling caps, so what is here appears to be adequate. Hard to say, but I would guess the stock ones are polyester or more likely metalized polyester. Replacing these with polypropylene, particularly the input capacitors (0.22uf) or with orange drops, etc. since you have the room would be beneficial (as Tangent noted). The output caps should be as Tangent suggested. Tangent’s Parts Shop Lots of goodies here, so poke around.

I tried straight jumpering these u22k62 caps like Tangent suggested = BAD NEWS. There was no output until you cranked up the potentiometer, and then all I got was horribly distorted sound. I cut the jumpers, and thankfully, the sound returned to normal. Any idea why these caps are apparently necessary? (I'm using AD843 x2 via Brown Dog adapter, if it makes a difference.) I initially thought this may have been due to major DC offset introduced from the input, but I measured 0mv offset at the headphone out.

 

[tubes]

I have some .22uf 450v auricaps that I can use in place of the input caps in my OBH-11. Is the 450v rating going to affect the sound adversly, the original caps are 63v.

Phill

 

[yzriver]

Quote:

Originally Posted by SmellyGas /img/forum/go_quote.gif I tried straight jumpering these u22k62 caps like Tangent suggested = BAD NEWS. There was no output until you cranked up the potentiometer, and then all I got was horribly distorted sound. I cut the jumpers, and thankfully, the sound returned to normal. Any idea why these caps are apparently necessary? (I'm using AD843 x2 via Brown Dog adapter, if it makes a difference.) I initially thought this may have been due to major DC offset introduced from the input, but I measured 0mv offset at the headphone out.

I measured about 12V at the NE5532 input. i didn't track where it is from, but whatever, there is it. If you short the cap, you change the bias.
There is a cap before the headphone out. It blocks the DC, so the DC offset is 0.

 

[yzriver]

Quote:

Originally Posted by tubes /img/forum/go_quote.gif I have some .22uf 450v auricaps that I can use in place of the input caps in my OBH-11. Is the 450v rating going to affect the sound adversly, the original caps are 63v. Phill

I use orange drop 715P 0.33pF, 600V rating. They are working well.

 

[Girlsound]

This is a crappy sketch of a more complete schematic. You missed the power supply section. You can now see why the input cap is needed.

The first thing I did was to use a proper regulated supply and remove that awful zener. The 5532 went next. I swapped in an AD8620. I got a bit warm, but sounded great.

Input and output caps need to be upgraded. But you've already got that covered.

The output buffer is kinda wimpy, but it's not a job for the meek to find a way to heatsink the BD139/140 transistors. I replaced the 30ohm output resistors with nice RN60 25ohm ones.

I was even thinking of a way to convert it to a bipolar power supply in order to DC couple the amp. It was about that time that I realized it would be a lot easier to go with something else. I was actually quite happy with my DC output coupled Szekeres.

 

[millwood]

it looks like a fairly good amp. the use of a discrete buffer (to225 transistors) is a good sign - many of the amps built here don't even have that.

I would try to figure out how the output buffer is configured and if it is underbiased - it looks like a class AB buffer biased with 3 diodes per channel. if true, the output is underbiased.

I would leave the ne5532 alone: it was a pretty good opamp then and it is a still a good opamp today. it may not have as impressive spec as other newer opamps do but it is better than what most people can tell with their ears.

 

[millwood]

well, having a schematic helps a lot.

it does seem that the output stage is properly biased: the to92 transistors form a constant current sink to drive the diodes.

the design is a single rail type and you have to have the input / output caps.

a few things to do:

1) you can bypass C8/C10/C4 (if electrolytic) with a small cap (.01u - .1u).
2) you can turn the output into a class A buffer, by cutting the pnp transistor's emitter out, and then taking one of the diodes out. now, R11 and the supply voltage determines the Class A amp's idle current: I=(Vc/2)/R11 so you will need to reset R11.

you may also wish to put a small resistor (around 110ohm on the npn transistor's base).

 

[SmellyGas]

After my little fiasco jumpering the input caps, I decided to do an all-out upgrade.

NOW the amp sounds fantastic! Pre-modification, the headphone out jack on my $200 cheapo Sony receiver sounded BETTER than the Creek. Adding a regulated power supply only seemed to improve the SNR. The Sony receiver still sounded better. I then tried various opamps - LM4562, OPA2132, and finally dual AD843's. The AD843's sounded the best of the 3. I then tried a pair of OPA627BP's (normally $27/each) that I got cheap off ebay, but they turned out to be COUNTERFEIT - ironically, they sounded better than the AD843's!

My last step was to replace the input caps with AudioCap polypropylenes and all of the electrolytics with Elna Silmic II's, except the big power supply cap, which I replaced with a Nichicon FG. I also replaced the counterfeit OPA627BP's with real OPA627AP's from digikey. The only problem was that the caps were too big to fit in neatly, so they're bent into funky directions.

The headphone amplifier sounds way better now. Before the Creek's bass was weak, and the upper highs were muted compared to my cheap receiver. After the upgrade, the receiver now sounds considerably worse - "dry" and less pleasant. Bass is improved, and vocals sound buttery smooth. There is much more detail and separation in the upper highs. I'm very happy with the way it sounds now with my HD650's.

Total Cost of mods:
Elpac WM080-1950-760 regulated PS - $19
OPA627AP x 2 = $38
Brown Dog adapter = $10
Elna Silmic II and Nichicon caps = $20

I hope this additional information will help someone. The OBH-11's were selling new for <$100 on ebay a while ago, so this might be a worthwhile project.

 

[rogerlike]

Nice. Does it still fit into the case?

 

[SmellyGas]

Quote:

Originally Posted by rogerlike /img/forum/go_quote.gif Nice. Does it still fit into the case?

Yes, with the creative capacitor bending. I guess I could have also used smaller caps (with lower voltage rating). Oh well.

 

[rogerlike]

You should send the pic in to Creek, see what he says!

 

[tubes]

I have some 220uf Rubycon ZL low impedance caps that I can use to replace the existing output caps. Are low impedance caps the wrong choice for the output caps, I have heard that low impedance caps can cause oscilation? Thanks for your help.

Phill

 

[yzriver]

Quote:

Originally Posted by tubes /img/forum/go_quote.gif I have some 220uf Rubycon ZL low impedance caps that I can use to replace the existing output caps. Are low impedance caps the wrong choice for the output caps, I have heard that low impedance caps can cause oscilation? Thanks for your help. Phill

the Rubycon ZL is low ESR caps, which is good for power decoupling. For audio application, the caps with low tan d value are prefered. Usually, the power caps have higher tan(d) then audio caps, for Electrolytic caps, but you can try. You still can hear the sound.

 

[tomano58]

Hi all, I'm new to this forum but have read it for a while. I know the thread is old but I recently got my self a Creek OBH-11 v1 for coffee money and decided to upgrade it a bit. I contacted Creek and asked nicely if they could provide me with a schematic diagram and they did (thank you). So for those of you who might still want it, here it is. This is the v2 version, which most of you seem to refer to and I have the v1 version but my guess is all that differ is that in v1 it's a TL071 and only one set of RCA sockets. If publish it like this is not ok, please let me know.
 
I noticed the zener diode, ZD1, is a 22V one. Do you think I can skip it and use the 24V/1A supply I have (switched and stabilized) without compromising anything? Or do I need to keep the zener?
 
My plan is to do as most of you have, upgrade the in/out capacitors, better power supply, probably a better pair of RCA sockets (wire connected) and the opamp. I'll start with a OPA2134 (I was actually going for a LM4562 but it was currently out of stock at my supplier)