ashok said:
Have a couple of bottles of beer while someone takes the trouble of trying it out. It WILL be interesting.
We have seen lots of No, No's working out well !
That's what DIY is all about in the first place .
Cheers.
Ashok.
Indeed.
You need to LISTEN and judge.
Maby if you guys pay the trip, Nuuk could come to my house
and we will compare. There's beer too.
Some mods
Hi all,
I'll be trying the following mods and will post back. Feel free to let me know if there are any others I should know about. This iwll take me about 3 weeks- I have a day job!
1. Raise voltage to +-18V
2. Use MUR860's for the diodes in the PS (currently 15A diode blocks)
3. Bypass ADJ pin of LM317/LM337 to gorund with cap-to lower noise.
4. Try carlosfm's 4700uF cap per supply rail for the GC
5. Use the 200pF cap on the OPA627-I currently use 18pf (as per Pavel Macura)-did not really seem to do much.
6. Bias the OPA627 to class A with a resistor (5k?) between V- and OPA627 output
7. Try the 50ohm resistor to ground. Pavel reports this resulting in tighter bass and less grainy mids.
Anyone else trying these, let us know!
Ryan
Hi all,
I'll be trying the following mods and will post back. Feel free to let me know if there are any others I should know about. This iwll take me about 3 weeks- I have a day job!
1. Raise voltage to +-18V
2. Use MUR860's for the diodes in the PS (currently 15A diode blocks)
3. Bypass ADJ pin of LM317/LM337 to gorund with cap-to lower noise.
4. Try carlosfm's 4700uF cap per supply rail for the GC
5. Use the 200pF cap on the OPA627-I currently use 18pf (as per Pavel Macura)-did not really seem to do much.
6. Bias the OPA627 to class A with a resistor (5k?) between V- and OPA627 output
7. Try the 50ohm resistor to ground. Pavel reports this resulting in tighter bass and less grainy mids.
Anyone else trying these, let us know!
Ryan
Re: Some mods
1. Yes, definitely.
2. I used 1A schottky diodes. You don't need those big ones.
3. Yes yes yes.
4. No. After more tests, I don't recommend passing around 2000~2200uf TOTAL capacitance per rail.
5. Yes. I thought you already had the 200pf cap.
6. I doubt that you'll gain something with that here.
7. I haven't tried it.
Ryan, hurry up, make the other channel.
You don't know the soundstage you're missing.
Dr.H said:
1. Yes, definitely.
2.
I used 1A schottky diodes. You don't need those big ones. 3. Yes yes yes.
4. No. After more tests, I don't recommend passing around 2000~2200uf TOTAL capacitance per rail.
5. Yes. I thought you already had the 200pf cap.
6. I doubt that you'll gain something with that here.
7. I haven't tried it.
Ryan, hurry up, make the other channel.
You don't know the soundstage you're missing.
carlos, are you still using the same schematic as at the beginning of this thread? if not, can you post the schematic you're using right now?
apart from the fact that you're using a single channel chip now, of course ...
edit: oops, thought you were using 2134's ... strip that last comment...
apart from the fact that you're using a single channel chip now, of course ...
edit: oops, thought you were using 2134's ... strip that last comment...
But this thread is called High-end preamp, right? I have spent quite a lot of time trying the get reasonable sound out of 627/634 using premium grade components and the sound is just mediocre. I guess it's about expectations.
ECC88 is probably one of the most hated tubes in audio but i don't really have bad feelings towards it. I even like it in Audio Research SP10 but it works at 300v there.
Yes, i have tried it at lower voltages and it really sucks. Can it make a magic mix in the GC? Maybe, but you'll need a current source or an active load or an anode choke. These measures will allow the tube to work as if the PS voltage is much higher.
Dr H
3. Get rid of the 317/337. A discrete regulator or even better, a battery is the single biggest improvement
6. I've tried a 2mA FET diode - very slight improvement ; a resistor will likely do worse.
The benefits of a seperate toroid
During the process of building the pre (which I have installed inside the GC box...) I tried a separate transformer to feed the op-amps.
What I heard was surprising-even on one channel, there was so much presence in the sound, so much space and depth. Because I had hummmm initially, I though that maybe the toriodal transformer was injecting some rubbish, so I connected the op-amps to the GC supply.
This sound was not so great-the amazing depth of the previous set-up had mostly disappeared and the sound was more attached to the speaker. Still better than no pre-amp. After sorting the hum out (a ground loop had crept in), I reconnected the toroid to the op-amps and we were back in wonder land....
My 2c so far with this amazing little device.
PS> Carlosfm-I used the diode blocks cause thats what I had lying around.
A couple of Q:
1. Has anyone tried swapping the OPA627 for other op-amps (AD8065 and other well regarded entities). I have built the pre using sockets so can swap easily).
2. If using a battery, I guess you'd need 2x12V per rail (+ and -) which would then be regulated down to 18V? Or is anyone using +-12V to run this little beast?
Thanks for sharing.
Ryan
During the process of building the pre (which I have installed inside the GC box...) I tried a separate transformer to feed the op-amps.
What I heard was surprising-even on one channel, there was so much presence in the sound, so much space and depth. Because I had hummmm initially, I though that maybe the toriodal transformer was injecting some rubbish, so I connected the op-amps to the GC supply.
This sound was not so great-the amazing depth of the previous set-up had mostly disappeared and the sound was more attached to the speaker. Still better than no pre-amp. After sorting the hum out (a ground loop had crept in), I reconnected the toroid to the op-amps and we were back in wonder land....
My 2c so far with this amazing little device.
PS> Carlosfm-I used the diode blocks cause thats what I had lying around.
A couple of Q:
1. Has anyone tried swapping the OPA627 for other op-amps (AD8065 and other well regarded entities). I have built the pre using sockets so can swap easily).
2. If using a battery, I guess you'd need 2x12V per rail (+ and -) which would then be regulated down to 18V? Or is anyone using +-12V to run this little beast?
Thanks for sharing.
Ryan
analog_sa said:
I can't figure out how you made it sound mediocre.
The OPA627 is very neutral.
It spits out what you feed it.
If the source is not good, it will show.
If the source is a CDP with those nasty NEs, a valve pre may "sweeten" the sound.
What can I say?
Bring your best valve preamp, and let's compare.
One of us may be surprized.
It may be me, it may be you, who knows?
Ryan,
My experience with AD op-amps has always been a negative surprize.
The OPA627 (in fact, almost any OPA) is much superior to the "widely acclaimed" AD825.
The AD826 (double) has the same distinct characteristic.
No midband, very recessed for my taste, and it robs much of the detail.
I haven't tested the new AD8065, but I should say that besides being SMD (no Dip version), it has a max voltage rating of 24v (of +/- 12v).
My advice is:
BE QUIET, YOU ALREADY HAVE THE BEST!
My experience with AD op-amps has always been a negative surprize.
The OPA627 (in fact, almost any OPA) is much superior to the "widely acclaimed" AD825.
The AD826 (double) has the same distinct characteristic.
No midband, very recessed for my taste, and it robs much of the detail.
I haven't tested the new AD8065, but I should say that besides being SMD (no Dip version), it has a max voltage rating of 24v (of +/- 12v).
My advice is:
BE QUIET, YOU ALREADY HAVE THE BEST!
Do you have any specific suggestions (circuits) for a discrete regulator?
I have tried one but didn't note much (if any) difference to the 317/337. In fairness I should say that the discrete regs were used as the second stage of regultion following a 317/337 stage. Perhaps I should try them on their own!
I suppose for a preamp using so little power, there is little excuse not to use a battery (or two).
I had good results using raw supply (rectifiers and caps only) with a CDP output stage. It might work well with those chips as well. After all do you really need that regulation?
Running the stage from a raw supply was actually comparable to battery power (maybe even better). Even a single pass transistor in active regulation (with active reference) degraded the sound.
Running the stage from a raw supply was actually comparable to battery power (maybe even better). Even a single pass transistor in active regulation (with active reference) degraded the sound.
I don't have a cd source at all. If the sound passes through a couple of opamps inside the cdp a few more won't hurt it - maybe that's why our opinions differ.
As this is not likely we'll have to contend with the exchange of long distance opinions. The point is that i can easily compare opamps and valves and even a simple 5687, choke loaded and vaccuum rect/choke supplied sounds incomparable to anything opamp. Don't get me wrong, i am not prejudiced against opamps. I would just love it if something small, light, simple and cheap can replace something ugly and heavy, but it just ain't possible. I didn't touch opamps for almost 20 years and am surprised how little they have progressed soundwise.
I didn't like 825 at all; 8065 is a bit better but similar and 8610 is actually better than 637, which in turn beats 627 to my ears. I have only compared them with a buf637 in the loop and they may well sound different unbuffered.
You may well be happy running at +-12v. The 8610 doesn't like more anyway.
Hmmm, Peter, people here don't like reading stuff like this
A lot of us have been conditioned from years of reading Audio Amateur and similar publications that true high-end demands regulation. After years of pondering i (sometimes) accept regualtion as a necessary evil.Almost all amplification stages, no matter how great PSRR is claimed, benefit from a PS which is not only clean from self/mains generated noise, but more importantly - from the PS hash generated by other stages. You give each opamp/transistor/valve it's own PS and the sound improves a lot. But with all the transformers it tends to get heavy. And expensive. Regulators certainly impose their own sound. Good ones help control the bass, improve the soundstage and keep images steady. The bad ones also help the bass but impose a stranglehold on dynamics and naturalness. The absence of regulation gives remarkable dynamic freedom, good midrange, large images (maybe too large) and generally **** poor bass.
With valves it's easy - use low resistance, preferably nickel chokes and Bob's your uncle. For a bit more sparkle add a couple of gas regulators. Try to avoid series regulators.
Do opamps sound better unregulated? Maybe, but how do you get acceptable bass?
In my limited positive opamp experience i almost reached Nirvana once with a 3 stage opamp phono circuit where each opamp had a separate NiCd supply. All in all 6 bipolar supplies.
200 pF in or out?
Hi Pavel and Carlos,
Perhaps it's just my bad english, but I couldn't figure out if you agree on the 200 pF cap.
Should this cap be used or not?
I understand that PMA uses 18 pF and everybody else 200 pF. Has anybody heard a difference in sound? Which one was best?
Last, but not least: what type of cap should be used in this position?
The reason for my interest: I have the same circuit in my preamp. I currently don't use any cap in this place. I find that the preamp sounds pretty good. Better than my previous opamp-based pre anyway. I couldn't say how it would rate on a absolute scale.
abo
Hi Pavel and Carlos,
Perhaps it's just my bad english, but I couldn't figure out if you agree on the 200 pF cap.
Should this cap be used or not?
I understand that PMA uses 18 pF and everybody else 200 pF. Has anybody heard a difference in sound? Which one was best?
Last, but not least: what type of cap should be used in this position?
The reason for my interest: I have the same circuit in my preamp. I currently don't use any cap in this place. I find that the preamp sounds pretty good. Better than my previous opamp-based pre anyway. I couldn't say how it would rate on a absolute scale.
abo
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.