|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
50.159.77.150
SLA's generally seem popular here but are such a pain. I'd do it if necessary but thought I should at least ask about reasonably priced, off the shelf, offline 12VDC that would be a good fit with my Cambridge 651p's. What is everyone using, liking???
Follow Ups:
This things line frequency noise on its regulated outputs driving a 651p's bound to be right near where I need it to be.
It's got dual outputs. Adjustable to plus and minus 18V. Nice and beefy. Probably less $$$ than I could diy something this nice....
You can DIY for less money. It is open frame, so you STILL have to provide an enclosure, although it is simpler as there is just a frame to mount. At that point, the Belleson regulators look like a better deal.
By the numbers those Bellesons SPM7818 and SPM7918 look like they would be a drop in and do it all for less $$ and not require access holes through the chassis. Verry tempting.
Edits: 10/12/14
As I stated, I had very good results from upgrading my Graham Slee Reflex X phono stage with a Belleson SuperPower regulator. (link to Vinyl Asylum post below)
Belleson now makes the SPM regulator, which is smaller, and has lower noise specs. than the one I used. Maximum current on the SPM is 225mA, and my Slee draws about 30mA quiescent. If the SPM were available then, that's what I'd have used. Check the current draw of your 651P, and if the SPMs work, I'd say go for it!
Oh, and here's (link) why I suggested you try a higher current AC transformer. Even a larger (higher current) AC to 12v AC wall-wart might help. Steve McCormack offers transformer upgrades to some of his older (Micro series) equipment for the same reason.
For instance, this LOOKS LIKE a dual output regulator. I wish I had the patience to WAIT for delivery of one of these. Sure beats breadboarding, what a PITA that is! All you need supply is the case and a transformer. Hard to tell how big it is, though.
The little ∿ on the back of the 651P power input jack is a sine wave. That means you have to input 12 volts AC, not 12 volts DC. The rectification to DC is done internally. Using the wrong supply may have caused serious damage to your Cambridge phono stage.If you're looking for an improvement, get yourself a high current transformer that outputs 12 volts AC. The lower impedance of the big transformer will give you improved sonics, according to Paul McGowan of PS Audio.
If it were me, the next thing I'd be looking at would be to upgrade the voltage regulator(s). I've had excellent results with a Belleson SuperPower regulator in my Graham Slee Reflex phono stage.
Edits: 10/07/14
While you're correct that the unit is expecting AC, no harm will be done by throwing DC at it. You just need to make sure that you provide enough voltage (measure what comes off the bridge rectifier inside, you need to add a little to that, as that bridge will drop your incoming DC voltage a bit) and current (says 0.5A on the back, that seems easy enough).
The nice thing about having that AC socket on the back is that you don't have to worry about the DC polarity. (Tip hot or tip negative, doesn't matter)
There's a slight chance that there's a voltage doubler tucked in there, though it's very unlikely, so you'll need to do your homework before you get too carried away! If there is a voltage doubler, I wouldn't bother trying to get an external DC supply to work.
I reacted, thinking about an AC input on a DC circuit, not the other way around, without actually thinking it through. Thanks for the clarification.
I'd be bypassing the rectifiers and tie power to the regulator outputs if I end up using an external dc source/s. While it wouldn't break anything to pass DC in through the AC input, the circuit wouldn't work that way. This circuit uses bipolar rails and there are only a hot and neutral pin on that ac input.
There is no voltage doubler circuit inside. The ac is rectified, filtered and regulated. It is only missing sufficient filter capacitance to be adequate in my estimation but there is no room in the box for what it needs. Anything from somewhere around plus and minus 5 to 18VDC would make this circuit function well enough I'm guessing.
If you have bipolar rails from your AC wall wart, then you very likely have a voltage doubler, especially considering the low AC voltage already available to the circuit.
What is leading you to believe you need more capacitance? Generally a regulator will provide pretty good noise rejection.
I pulled one apart and measured it while running tonight. At the transformer input, ~18VAC 0-p. Nearly 26V at the regulator inputs. Doubler? Looks like that's it."What is leading you to believe you need more capacitance?"
Oh nothing. Recall my original post was a request for power supply recommendations. This is just a tangent. I was just speculating about what might be done to solve the noise issues I'm seeing if I were to try and patch the existing power supply. "More capacitance" would likely still be my answer after poking around tonight with my scope a bit. Though the circuits obviously not what I first thought.
Using the AC coupling mode on my scope so I could zoom in on the ac component of a power supply rail noise w/ the scopes 5mV/div scale. I verified the noise riding on the regulator output is essentially identical to the noise I'm seeing at the audio output. Possibly it's some kind of summed version of the ripples on the positive and negative rails. Coincidentally it is also seemingly a scaled down version of the ripple riding on the regulator inputs. The regulators had some decent rejection of this noise, like you say.
The picture is of the output noise in MM mode with 1k terminations right on the inputs. The scope is at 5mV/div and 5mS/div.
Edits: 10/10/14
That's NOT a doubly or a doubler! 18 VAC RMS, actually gives you sq. root 2 * RMS VAC = 26 P-P voltage, which is fine. An actual doubler would give you 2 * P-P or 52 volts! That is great if you need plus and minus rails, +18-0--18, which is a 36 volt difference.
You should measure and report on the scope trace at the output rails AFTER the regulators, because it is unclear whether the preamp runs on balanced rails OR a single rail. You have two regulators in the picture, so I suspect balanced rails, i.e., +9-0--9 and not +18-0. You should trace the diagrams out by careful examination. Here are ways voltage doublers actually work, but I don't think that is what you have:
Yes, in general, they are NOT a good idea. However, I had a tube preamp run off a voltage doubler. With suitable capacitors, it sounded OK. It's usually an issue of which transformer is readily available.
"That's NOT a doubly or a doubler!"It's something like that. I haven't bothered to reverse engineer the thing. I am not planning to reuse any of it.
"18 VAC RMS"
Reread what I wrote. Not RMS, I said zero to peak.
Here is a picture I took last night of the transformer output into the preamp when it was running. 5V/div, 2mS/div.
"You should measure and report on the scope trace at the output rails AFTER the regulators"
Did you read the post you are responding to? I described it in a way I considered to be pretty clear at the time. What exactly don't you get about my description? The picture didn't seem all that interesting to me since it looks almost identical to the one I posted last night.
Here it is. Ac coupled view of noise on +18V rail if i recall correctly. 5mV/div, 5mS/div.
"it is unclear whether the preamp runs on balanced rails OR a single rail."
Hmm. Only unclear if someone hasn't actualy been reading the thread. I mentioned before the audio sections have plus and minus 18V rails.
"I suspect balanced rails, i.e., +9-0--9 and not +18-0"Yes it's plus and minus 18V rails.
"You should trace the diagrams out by careful examination. Here are ways voltage doublers actually work, but I don't think that is what you have"
Not doing it. The chances of me reworking the existing supply are very slim. Doesn't seem worth it at all. My preferred method is off the shelf but if I diy I'm scrapping everything and starting over.
"Yes, in general, they are NOT a good idea"
I get why they would have gone that route but there would be no reason for me to use that type of circuit if were trying to upgrade other than if I were trying to reuse some of the existing circuitry for some reason.
Edits: 10/11/14
Bipolar rails from 9 VAC?? That seems very little to work with, what 6 VDC per side? I had a +15-0-15 supply through a voltage balancer ?? for my phono stage. I replaced the whole thing with a +15 and -15 regulated supply. It was a somewhat well-regarded Black Cube stage. It really opened up the bass response after the mod.
I believe 12VAC -> 12Vrms 0-peak.
Measured a little tonight...
Mines actually putting out closer to 14Vrms 0-peak per my meter when it's unloaded. The regulated rails are sitting at right around + and - 18V.
I'd be happy just addressing the noise I'm seeing but better sound from other performance improvements would be great!
?? Does it have another transformer on the inside? 12 VAC will only get you about 18 VDC off of the first cap (1.414 * VAC = VDC) and so, to get 18 V regulated, it is not a normal regulator as it could be a low offset regulator and then how does it get -18v DC? Gasp, it could be a switching regulator??? Those are VERY bad.
Anyway, as I explained, a 1 Amp, 24 volt center-tapped transformer and very simple fixed voltage regulators and reasonable capacitor size, 3300 uF, 50V or so (I'm trying to recall) should get you to a pretty good power supply without having to go to the Belleson's. EBay will sell you regulators for $8 each. I tore up an old CD player and broke off the power supply section for mine. The circuits are really simple if you know how to use a soldering iron. The transformer is the most costly item for this and, probably, why you have to spend big bucks to get reasonable performance.
Xfmr regulation voltage depends on loading. I didn't measure but am guessing with the light loading this one sees it is sitting somewhere between it's no load voltage, the 14V I measured, and it's full load voltage of 12V. LDO's can be had with 0.1V drop and with 0.7V across what seems to be a half wave rectified circuit you only need18.8V from the tranny for 18V rails.If I do end up diying I'd most likely run with no regulation other than the transformer regulation but pump up filter capacitance so big ripple gets kept in the noise floor. What's even better than a really good reg? No reg at all.
Edits: 10/10/14 10/10/14 10/10/14
OK. I think it must be a typical voltage doubler, which makes sense, which means it is only getting half-wave rectified per rail, which is not as good as a center-tapped full wave circuit.
You should check the output rails compared to the input rails and see that there is, in fact, about 3 volts DC lower on the output side, which means they are standard regulators (+15volt and -15volt regulators are < $1 apiece).
Yes, the simplest mod is to just increase the cap size, maybe double it. I think having regulators is good, too as I find nothing wrong with them and they must be good as they are used in most audio circuits.
Still completely confused. It is good to take your measurements both before and AFTER the regulator to know what you are dealing with. Usually the ICs have VCC and GND pins that are easy to find, being the first and last pins. Having a spike AFTER the regulator doesn't seem very good to me -- but as I don't have a scope, I am not sure how the waveforms typically look. Usually, they just specify a ripple voltage %. So 5 mV for 15 volts is .03 %, which seems kind of high to me.
Thanks for the post.
The plan was always to apply the dc internally onto the pcb at the regulator outputs so nothing was ever really at risk.
Since my problem is not sagging voltage but line frequency ripple. I'd expect more benefit from adding more filter cap than I'd expect to receive from a bigger transformer. There is just no room inside there for enough more cap to be added to make much difference. Not that I wouldn't expect any off the shelf supply worth looking at to have a bigger transformer.
In my case it would seem that filter cap truly is what I need since the profile of the noise I'm aiming to eliminate is line frequency plus a couple odd higher order harmonics. The stock regulators could be perfectly adequate for this task up to the level of their own and the circuits opamps Psrr limits if only there were much more pre regulator filter capacitance. The problem is there just isn't much filter cap.in there now. Though regulators with better Psrr could be part of my solution I think I have too much ground to gain and so will likely need a multi tiered approach.
I'll definitely have a look at the bellesons.
Cheers
I've had good luck building my own supplies. At some point, it came to that I didn't want to spend $1K on each component just to get good sound, so DIY I went. It's actually very simple. The biggest problem is getting the power transformer. Next is getting a suitable circuit board unless you are just rebuilding the existing circuit, if there is space to do so.
So far, I've built PS's for a phono stage, a headphone amp and a DAC. You are on the right track, to get a bigger transformer and use a larger sized capacitor at the input to the regulator. Those Belleson's appear to be a good idea, although then I would be in for a $250 PS whereas, I was looking at a $35 PS!
Going all out on a DIY phono stage with custom power supplies is probably the long term plan. For now I was hoping to band-aid these Cambridge units well enough to move forward, do some recording for a buddy, without much cost/time outlay. This leaves me a bit torn on putting to much effort into custom power supplies for these cambridge units when I will likely go a different phono stage direction in the end.
Finding a cheap Chinese but well enough built off the shelf supply would put me right in the sweet spot I think for this project but this has eluded me so far. Maybe some cheap SLA's off eBay really is the way to go for now.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: