New caps
Dean,
Like all Auricaps this one also will have its outside foil identified. This will allow it to become a natural shield by being connected to ground. From this I would expect an improvement in EMI fields. No, the only potential problems are due to its being larger and should be minor except it does have to fit.
Roger
Dean,
Like all Auricaps this one also will have its outside foil identified. This will allow it to become a natural shield by being connected to ground. From this I would expect an improvement in EMI fields. No, the only potential problems are due to its being larger and should be minor except it does have to fit.
Roger
KLe said:
Yes, I based my PS from Dejan's article.
I have to update the schematic though since I didn't build the PS in my UcD400AD amp exactly like the schematic in my gallery. I ended up increasing capacitance a bit more with 4x 6800uF per rail. The main reason to use many small caps than a few large caps was cost.
I also omitted the small cap parallel to the switch. And the power switch is actually switches both Hot and Neutral, not just Hot.
I also used the snubber circuit that Dejan recommends in his article (I haven't had the time to experiment with different cap values yet).
I find this amp sounds truly outstanding. The best I've had so far.
I might try some fancy bypass caps in the PS rails like Auricaps or Teflon caps in the future.
I'm still debating wether to boost the gain to 32-34dB or not...
Hi,
If you don't like with it with the gain boosted.. at least you can go back right?
It will certainly have an affect on THD and frequency response etc, but "different" doth not equal bad. I wouldnt' do it expecting better sound though, but like those island guys say, it might be good for making up the gain lost in a passive pre amp.
This really has little to do with tweaks but I'd like to point out one simple thing here:
My preferred tweak as far as coupling caps go would be not to have any, and I've yet to see any soundcards that have any DC offset at all. Just a thought.
Regards,
Chris
If you don't like with it with the gain boosted.. at least you can go back right?
It will certainly have an affect on THD and frequency response etc, but "different" doth not equal bad. I wouldnt' do it expecting better sound though, but like those island guys say, it might be good for making up the gain lost in a passive pre amp.
This really has little to do with tweaks but I'd like to point out one simple thing here:
My preferred tweak as far as coupling caps go would be not to have any, and I've yet to see any soundcards that have any DC offset at all. Just a thought.
Regards,
Chris
Hi Julien: Did the snubber circuit, that Dejan recommends in his article, improve the sound of the amp. If it has then how would you rate the improvement?
Yes, Auricaps, Teflon caps or perhaps even Mundorf caps (if they will fit) as PS rail bypass caps sounds good.
If you do not need to boost the gain, then I would leave it as it is. Although, as Chris wrote, if you increase the gain to 32-34dB and you do not like it, then you can always undo the change?
Yes, Auricaps, Teflon caps or perhaps even Mundorf caps (if they will fit) as PS rail bypass caps sounds good.
If you do not need to boost the gain, then I would leave it as it is. Although, as Chris wrote, if you increase the gain to 32-34dB and you do not like it, then you can always undo the change?
Hi,
Deandob, your summary is dangerous. All those tweaks will give you a different sound, but there's nothing to say that it's a sound you'll like. I think this thread would be most usefull if you learn from it how to tweak and why it makes a difference instead of using it as a shopping list, what I've learnt from it so far applies universally to any audio equipment I might want to tweak.
If I use FC caps on my module it's not because everyone else used them but rather because they look OK on paper for the job they'll be doing and they're cheaply and readily available from digikey, right alongside the five others I'll audition in their place.
Anyway...
While we'd all like to buy or design/build a 200$ uber auxiliary supply, what I'd like to discuss now is what exactly it would take that would actually be considered an improvement over the standard regulators which I think have already been posted in this thread.
I think that done right this should prove to be a very cheap and worthwhile modification, but am I taking a step back if I use a 7812/7912 IC regulator in its place? I think that may have the potential to be worse actually, and maybe supplying the given regulators with it's own supply will be enough of an edge. Also the op amps do have some PSRR so there really shouldn't be a need for ultra tight regulation which will end up as money spent for nothing beyond a certain point, it's just that extra edge we're after.
Open for discussion.
Regards,
Chris
Deandob, your summary is dangerous. All those tweaks will give you a different sound, but there's nothing to say that it's a sound you'll like. I think this thread would be most usefull if you learn from it how to tweak and why it makes a difference instead of using it as a shopping list, what I've learnt from it so far applies universally to any audio equipment I might want to tweak.
If I use FC caps on my module it's not because everyone else used them but rather because they look OK on paper for the job they'll be doing and they're cheaply and readily available from digikey, right alongside the five others I'll audition in their place.
Anyway...
While we'd all like to buy or design/build a 200$ uber auxiliary supply, what I'd like to discuss now is what exactly it would take that would actually be considered an improvement over the standard regulators which I think have already been posted in this thread.
I think that done right this should prove to be a very cheap and worthwhile modification, but am I taking a step back if I use a 7812/7912 IC regulator in its place? I think that may have the potential to be worse actually, and maybe supplying the given regulators with it's own supply will be enough of an edge. Also the op amps do have some PSRR so there really shouldn't be a need for ultra tight regulation which will end up as money spent for nothing beyond a certain point, it's just that extra edge we're after.
Open for discussion.
Regards,
Chris
The gain increase was basically to compensate for my transformer based passive pre and my "bufferless" CD player. My only worry is in increasing distortion.
I have the SMD resistors already, so I'll probably do it soon...
I have the SMD resistors already, so I'll probably do it soon...
classd4sure said:
Chris,
I agree, it is easy to make changes and think they have improved the sound. A proper scientific method is the only way to be sure and this requires a reference and only one change at a time compared to it.
The FC caps are a lot better than most stock caps but not the best. Others that have better specs and sound better can be very expensive and hard to get so the FC’s are a good compromise.
The UcD regulator is basically just a capacitance multiplier with a zener and has good noise rejection. This must have been felt to be the most important parameter and this could very well be right. Like the good doctor “first do no harm” it will be necessary to build something as quiet as what it is replacing. This first requires good grounding practices to keep from introducing a lot of noise (like separate grounds for +/-), then looking into the rest for improvements like source impedance and regulation. Fortunately most power supply improvements are readily heard as sound stage improvements so it should be relatively easy to determine if an improvement has actually been made. I am going to try some small SMPS as the source. This is primarily a space consideration as I am rapidly running out of room. I will have one set of amps with them and compare to a set without. This should tell the story.
On a side note; you are right about the caps being transducers in both directions. Some will sing like crazy. I once had a particularly noisy zobel cap and it took some time to find. As I remember it did have rather high Q and would sing at rather specific frequencies. You have peaked my interest and I will be trying some experiments to check this out. By the way, this was a customer’s amp, not something I built
Roger
Hi,
Yeah I'm uncertain what the pass transistor is but I'm assuming the caps's pass transistor is no slouch in its current capability either.
I'd go with the analog route myself, I'd have more fun making the supply that way.
I suspect the multiplier can stay in place but a more stable Vbe ref rather than a zener... with some measure of feedback, nothing too fancy is probably needed, but it could be done a number of ways too.
You've got me curious too and I'm very much looking forward to those results. Should we define what an acceptable level of reaction is before starting, in terms of what may make for "poor micro-dynamics". If you can't (I can't), maybe the experiment can be designed to help us figure that one out too and put some sort of FOM on it. (We'll tell no one!)
I hope to see a clear relation between testing methods and real world dynamics in order for the results to be meaningful. A little smoke is always cool too. I already gave one idea, another could be if you could hit the case with the same level of force each time, a third and entertaining one if not a bit eccentric..... kick a tuning fork, record it in real time, when it drops to a certain pre specified level, "snap" a pic of the cap voltage with a storage scope and take a DC reading.
I figure if you do it with enough tuning forks you can map the Q across the audio band at least. Any other ideas?
Regards,
Chris
Yeah I'm uncertain what the pass transistor is but I'm assuming the caps's pass transistor is no slouch in its current capability either.
I'd go with the analog route myself, I'd have more fun making the supply that way.
I suspect the multiplier can stay in place but a more stable Vbe ref rather than a zener... with some measure of feedback, nothing too fancy is probably needed, but it could be done a number of ways too.
You've got me curious too and I'm very much looking forward to those results. Should we define what an acceptable level of reaction is before starting, in terms of what may make for "poor micro-dynamics". If you can't (I can't), maybe the experiment can be designed to help us figure that one out too and put some sort of FOM on it. (We'll tell no one!)
I hope to see a clear relation between testing methods and real world dynamics in order for the results to be meaningful. A little smoke is always cool too. I already gave one idea, another could be if you could hit the case with the same level of force each time, a third and entertaining one if not a bit eccentric..... kick a tuning fork, record it in real time, when it drops to a certain pre specified level, "snap" a pic of the cap voltage with a storage scope and take a DC reading.
I figure if you do it with enough tuning forks you can map the Q across the audio band at least. Any other ideas?
Regards,
Chris
Fork it?
Chris,
I think your fork idea would be a little awkward to implement. What I have in mind is to use a precision signal source and put a 10 ohm in series with it then connect up the cap. I will use 15VAC as that is the max my generator will do. I will then sweep the freq and see want happens. This will be done while monitoring the current to look for peaks and dips as well as listening. If there are confusing results I will implement a SPL meter to help sort it out. Will let you know.
As far as a reference goes it is hard to beat the venerable old LM431. I do use and like this part. Have found a lot of uses for it in other applications. The pass element can be included in the feedback to the ref. making a good regulator. This is easy on the + rail and not so easy on the -. Just don’t overlook the 1-10uf feedback cap, it is necessary as is the static 10 milliamps of current.
Roger
Chris,
I think your fork idea would be a little awkward to implement. What I have in mind is to use a precision signal source and put a 10 ohm in series with it then connect up the cap. I will use 15VAC as that is the max my generator will do. I will then sweep the freq and see want happens. This will be done while monitoring the current to look for peaks and dips as well as listening. If there are confusing results I will implement a SPL meter to help sort it out. Will let you know.
As far as a reference goes it is hard to beat the venerable old LM431. I do use and like this part. Have found a lot of uses for it in other applications. The pass element can be included in the feedback to the ref. making a good regulator. This is easy on the + rail and not so easy on the -. Just don’t overlook the 1-10uf feedback cap, it is necessary as is the static 10 milliamps of current.
Roger
OPA627 Input Stage
Woah! I've just changed the OPA2134 on the front end of my UCD400's for a pair of OPA627AU's. What a difference! -4dB drop in noise floor across bass region on FFT of THD+N, and general dropping of overall THD harmonics. Result sound wise is phenomonal. The Bass is much tighter, faster and more punchy. Mid-range is now without bloom, and just sounds like liquid silk (If that is possible).
JP, please please redesign your board to use single op-amps on the front end so that the OPA627 is an available option. I know they're expensive, but they're certainly worth it. This is a bigger difference than removing those horrible 22uF AC coupling caps (Bass roll off is never a good thing!).
F.
Woah! I've just changed the OPA2134 on the front end of my UCD400's for a pair of OPA627AU's. What a difference! -4dB drop in noise floor across bass region on FFT of THD+N, and general dropping of overall THD harmonics. Result sound wise is phenomonal. The Bass is much tighter, faster and more punchy. Mid-range is now without bloom, and just sounds like liquid silk (If that is possible).
JP, please please redesign your board to use single op-amps on the front end so that the OPA627 is an available option. I know they're expensive, but they're certainly worth it. This is a bigger difference than removing those horrible 22uF AC coupling caps (Bass roll off is never a good thing!).
F.
Hi Furious,
Did you use a Brown Dog adaptor or something? Which?
Have you tried it with the AD8620's? How do they compare?
Thanks,
Chris
Did you use a Brown Dog adaptor or something? Which?
Have you tried it with the AD8620's? How do they compare?
Thanks,
Chris
I'd love to hear about the differences between the AD8620 and this OPA672AU.
The 8620 doesn't seem to have a sonic character at all, it's very neutral.
Problem is my sources all use inferior opamps... Even the Pioneer 868 top of the line DVD player is running 5532 opamps 🙁 🙁
The 8620 doesn't seem to have a sonic character at all, it's very neutral.
Problem is my sources all use inferior opamps... Even the Pioneer 868 top of the line DVD player is running 5532 opamps 🙁 🙁
FuriousD:
I believe you 😉
That's what I heard when I instaled OPA627AP in my M-Audio superDAC. More clean and detailed overall presentation (even midrange and midbass detail-articulation) much stronger bass, better silences, but on the dark/warm side.
That will be around US$40 for module, wright? 😱
Yves:
Perhaps you should try the BB in your DVD first 😉 (sources first)
regards
Mauricio
I believe you 😉
That's what I heard when I instaled OPA627AP in my M-Audio superDAC. More clean and detailed overall presentation (even midrange and midbass detail-articulation) much stronger bass, better silences, but on the dark/warm side.
That will be around US$40 for module, wright? 😱
Yves:
Perhaps you should try the BB in your DVD first 😉 (sources first)
regards
Mauricio
Max,
I would, but I don't like coming even near a soldering iron, especially on small things like opamps 😀
I'm just half a DIYer 😛
I would, but I don't like coming even near a soldering iron, especially on small things like opamps 😀
I'm just half a DIYer 😛
Yves:
But you'll have to start, sooner or later
Or get a friend or technician done it for you 😉 it is worthy.
I myself don't dear to attack the SMD opamps yet
Ciao
But you'll have to start, sooner or later
Or get a friend or technician done it for you 😉 it is worthy.
I myself don't dear to attack the SMD opamps yet
Ciao
Yves Smolders said:
Yves,
You will need to measure the 5532 supply voltage. Some of the older modules used +/- 15V and this is too much for the 8260. If this is the case Jan-Peter can advise you and supply the correct zeners.
Roger
We will change our UcD700 with a dual shape, so everybody can resolder the standard op amp for an other version.
Probably we will offer the UcD700 with a OPA627, but it will be a very costly upgrade! (whoah just checked the price by www.farnell.com......
)Jan-Peter
Jan-Peter,
"With a dual shape" - do you mean a kind of socket so other opamps can be put in place without soldering?
"With a dual shape" - do you mean a kind of socket so other opamps can be put in place without soldering?
- Status
- Not open for further replies.