PreAmp Regulated Supply

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I am currently in the process of building a power amp and the preamp will be sharing the same case. After advice from this forum I settled on using http://sound.westhost.com/project97.htm
this project as my preamp and tone control. I have come accross a fair bit of information but need of advice and clarification on the following points.

firstly I am looking for -+15v to feed 3 tl072 (this will be swapped with opa and lm opa amps to paly around)

I am thinking of using 78x regs 1A or 2A a
seprate ei core transforer will be used.


1)Are 78xx regulators sutiable for this task and if not is there a comparativly simple solution that will offer greater benefits.

2) I am thinking of putting a set of 1000uf caps after the rectifier/before regulator and a set after the regulators. Is there a configuration that offers any significant benefits without too much complication.

3) The power op amp is being feed with supply with two rectifiers (one for each rail) and the terminals are being bypassed with
100nf ceramic caps. Its this two rectifier setup suitanble for a pre
amp supplies as well. Soft recovery diodes? are they suitable in this situtaion.

4) Am i correct in assuming that using this supply to feed anything else but the preamp (eg clip indicator, peakmeter) will comprimise the quality of the supply. Is the simplest solution with the least impact to quality of sound is to take power off the main supply and run it through the 78reg to step down. if this is the case what is the best point to tap main supply. I have +-35v being run by 300va torid.

I have a limited understanding of electronics and am really crap at doing pcb layouts so simplicity will be greatly appreciated.
 
well, they are somewhat better than adequate, but remember that 78LXX and 79Lxx regulators regulate and <b>add</b> some tens of millivolts of noise.

there are several regulated power supply threads at diyaudio. From my personal experience, the best <b>simple </b>regulated supply I have built for a phono + preamp stage used a pair of LM317/LM337's in the positive and negative supply rails. I.E. the first LM317 regulated to 18 volts, the second LM337 regulated to 12 volts, and vice versa with a pair of LM337's on the negative rail.

Oh, and by the way, make sure that all the power supply pins of the opamps are decoupled with 100N bypass capacitors.
 
diyAudio Senior Member
Joined 2002
REGS.

Hi,

Generally speaking: the mua78/79xx series is pretty much outdated...unless you use them for a particular purpose.

To me, they're just fine for heater reg. in tube circuits.

Mind you, you can put their limited performance to your advantage.For example, you can take advatage of their poor bandwidth for filtering out digital nastities.

Other than that the LM series is quite superior in performance.

Cheers, ;)

/Horses for courses.:cool:
 
improved 78xxx/79xxx

How about a 78xxx ( or 79XXX) followed by a capacitance multiplier ? It should filter out the noise ?
Cheers.
 

Attachments

  • 78xxx.jpg
    78xxx.jpg
    6.8 KB · Views: 2,196
Double regulation.

You can use double regulation, with the 78xx and 79xx :bulb: .
That is, use 7818 and 7918 first and then 7815 and 7915.
Before the first regulators I would use 3300 uf, 1000 uf sounds weak to me.
After the regulators, something like 100 uf is fine.
Always bypass with 100 nf ceramic caps.
It's a simpler design than the LM317/LM337, but the LMs are better.
And double regulation with LMs? :eek:
Oh man, this never stops... :cool:

Cheers
 
I think that Andy's suggestion for a pre-regulator is a good one. Another worthwhile direction to pursue is constant-current operation, which can be implemented in a variety of ways. This would include folded cascode main circuits (with constant-current sources), shunt / super-shunt regulators, and some types of push-pull regulators.

In my experience, a major benefit of constant-current operation is that it tends to reduce the sonic differences caused by componentry choices in the power supplies.

hth, jonathan carr
 
Constant Current Operation

I'm really tempted by this Jonathan,

In your experience does it reduce the impact of transformer size upon the resultant sound?

The super-reg's were such frustration in this regard, since bigger transformers etc. still sounded better, and it seemed to be primarily impedance-related :(

Even the damn mains quality still makes a difference...

May have to raid the bits box - particularly after your comments on Super-shunts.

Andy.
 
Jonathan's ideas are good ones to try.

If you can limit the peak diode current, a small transformer can be made to sound as good as a big one.

One problem with those 3-legged things that no one thinks about is output Z vs frequency. Those devices were designed to give a specific DC output, without regard to most anything else. If you look closely at most dat sheets, when they measure transient response, they specify what capacitors are used on the output.

A preamp will most likely not pose a transient load problem, but how a regulator handles the residual audio on the rails is something that must be factored in.

Jocko
 
You've convinced me

That's my spare time disappeared in a box of bits and wire for a while ;)

I'm wondering now, if a good CCS feeding the main reservoir caps is an option. Resistive limiting of peak current was always sonically detrimental whenever I tried it....

Andy.
 
Hi Andy:

>I'm really tempted by this Jonathan.<

That was the idea. :)

>In your experience does it reduce the impact of transformer size upon the resultant sound?<

My experience and subjective impression is that it does do this, but not to the same extent as it diminishes the differences between filter capacitors. Mind you, in either case I think you can expect a worthwhile reduction, but not a total obliteration (I sincerely wish that this were otherwise).

>May have to raid the bits box - particularly after your comments on Super-shunts.<

For similar reasons, folded-cascodes fed from constant-current sources are also worth trying. And cannibalizing from one of my previous posts, it should be possible to adapt the following circuit for opamp duties:

http://www.diyaudio.com/forums/showthread.php?postid=142989#post142989

Whatever you try, I will be very interested in hearing of your results.

best, jonathan carr
 
Andy:

>Even the damn mains quality still makes a difference...<

Unfortunately true. A few years ago a friend and I worked on a flying-capacitor power supply with the goal of cutting the audio unit completely loose from the effects of the mains, but for whatever reason, I didn't like the sound of the results, and we didn't pursue this farther. OTOH, just because I didn't find the results subjectively compelling doesn't necessarily mean that this line of reasoning is faulty. Perhaps _our_specific_design_ was not thoroughly thought out, maybe there were implementation errors, or maybe it was just too much of a kludge. ;)

If you would like, I can try to locate the circuit that I used at that time and email it to you, or if more people are interested, I can post it here as an attachment.

>I'm wondering if a good CCS feeding the main reservoir caps is an option.<

I honestly don't know, as I have always fed the CCS's from the main filter caps, rather than the reverse. On the surface of things, apart from the topological differences, I wouldn't think there would be much difference in the final measurements and sonic results. But again, I have neither tried this myself nor thought about it much, so it is entirely possible that you may discover something interesting.

best, jonathan carr
 
Flying capacitor / charge pump

I looked at the link (that I think you) posted recently (www.never-connected.com), which on the surface seems to be a form of that.

Reading the online literature though reveals the following in the Q&A, which rather put me off pursuing it further, although I have heard good reports from those trialling the technology: -

"Like all power supplies the Never-Connected is dependant on the dynamic source impedance and voltage of the incoming power supply, if undersize or long cable runs are used between your in-coming supply and equipment, this will reduce the dynamic source impedance of the mains supply. The dynamic output impedance of the power supply will be affected resulting in a subjective loss of dynamic attack, and generally poorer definition in the Bass. This effect will be especially noticeable on Power Amplifiers. For best performance the use of a dedicated heavy current power feed (installed by a qualified electrician) will always ensure best results and can make a big improvement to sound quality"

I recently did some in-depth analysis for a friend whose very expensive system was having more bad days than good, with very large sonic variations (from unmusical to captivating in the extreme...). A long-term noise analysis of the mains showed no correlation with sound variation, local RF was eliminated by borrowing the HP kit form work (!), but some voltage variation was spotted, outside of UK specs.

The electricity co. did a log of their own and classed the supply as 'crap' and he's hoping for a new feed soon-ish. I suspect that mains impedance has a far greater effect than almost any other parameter - I've always found mains filtering to degrade sound for the same reason, even on v.low current PSU's.

Even more frustrating filters that appear electrically across the mains affect the other items plugged into the same socket / spur too :(

The annoying thing is I'd really like to find an elegant solution to some of this - super-reg's etc. do help, but not eliminate it.

I'll keep trying though, for a bit ;)

A.

P.S. If you can find the circuit, I'm sure others will be interested too - thanks!
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.