How many preamps per power supply?
gtzack4
Thread Starter
#1
23rd January 2013
Old 23rd January 2013
  #1
Gear Head
 
gtzack4's Avatar
 

Thread Starter
How many preamps per power supply?

I was wondering the effect of putting many mic pres using the same +-15 volts? I not experienced in building electronics I would like to build a unit with 4 mic pres and was wondering if it would be okay to use the same rails for all of them.

I was also winding the the same thing but for a headphone amp. Can I use the same rails to power 6 Channels of a headphone amp?

I have always imagined the more times you split the rails the more power that is lost.

What are the the effects of having multiple Channels using the the same rails?
#2
23rd January 2013
Old 23rd January 2013
  #2
Lives for gear
 

Hi
You don't 'lose power', but the supply needs to be able to give sufficient milliamps (for a mic pre) or conceivably up to an amp for multiple headphone amplifiers.
Do not use one supply to feed both preamps and headphones in the same unit.
There is a possibility of 'crosstalk' if several preamps share the same supply but it would depend on a lot of factors if it is significant or not.
Matt S
#3
23rd January 2013
Old 23rd January 2013
  #3
Lives for gear
 

you would have greater current draw. You just need to make sure you have enough current or Amps to power all the preamps. That will depend on the circuit. Each design will draw a different current amount. After you find that you multiply it by x number of preamps then add about 20 to 30% for the inrush current “powering the unit on.” You can never have too much current. Your supply will only draw what it needs, so it’s best to overshoot your needs on the supply. It will also help the entire system last longer as its not taxing the power supply too much and providing healthy power to your preamps.
Hope that helps
#4
23rd January 2013
Old 23rd January 2013
  #4
Lives for gear
 
nosebleedaudio's Avatar
 

Quote:
Originally Posted by gtzack4 View Post
I was wondering the effect of putting many mic pres using the same +-15 volts? I not experienced in building electronics I would like to build a unit with 4 mic pres and was wondering if it would be okay to use the same rails for all of them.

I was also winding the the same thing but for a headphone amp. Can I use the same rails to power 6 Channels of a headphone amp?

I have always imagined the more times you split the rails the more power that is lost.

What are the the effects of having multiple Channels using the the same rails?
What Mic pre?
A 2520 for example draws around 15mA.
#5
23rd January 2013
Old 23rd January 2013
  #5
Lives for gear
 
Richard Crowley's Avatar
 

You need to know the capacity (in amps) of the power supply, and the current draw (in amps) of each load (preamp, etc.) Then it is simple arithmetic to add up the current drawn by all the loads and see if the supply can handle it or not.

Regulated power supplies, such as we normally find for this kind of circuit, will maintain their rated output voltage regardless of load. At least until you reach their rated limit. You aren't really "splitting" the power.

Multiple channels on the same rails is the normal way of doing things.
gtzack4
Thread Starter
#6
23rd January 2013
Old 23rd January 2013
  #6
Gear Head
 
gtzack4's Avatar
 

Thread Starter
Thank you guys for clearing that up for me. I wasn't planing to put them in the same unit.

How do to calc current draw?

V=IR? But how would I apply it? Or is the current draw in the specs of the op amp?
#7
23rd January 2013
Old 23rd January 2013
  #7
Lives for gear
 
Richard Crowley's Avatar
 

Quote:
Originally Posted by gtzack4 View Post
How do to calc current draw?
The most realistic way is to MEASURE the current draw, not to attempt to "calculate" it. We typically interrupt each lead (the + side, and then the - side) and run it through a current meter. That gives us the ACTUAL current instead of some theoretical figure that may or may not be accurate for the unit at hand. Of course, you interrupt the power lead and route it through the meter only while the power is OFF. And you always provide all the voltage rails, even if you are only measuring one.

Modern digital multi-meters typically measure up to 10A. And you can get quite reasonably good meters for very modest prices. For example, here is one for $15: https://www.sparkfun.com/products/9141

Note also that some equipment, particularly when driving large loads like power amplifiers, changes its current draw depending on the amplitude of the audio signal. And remember that safety margins are always a good idea. When working on DIY, amateur projects, it isn't out of line to use at least a 100% "margin". If your the total of all your loads adds up to 2 amps, then use a 5 amp power supply, etc.
#8
24th January 2013
Old 24th January 2013
  #8
Gear maniac
 

Quote:
Originally Posted by gtzack4 View Post
Thank you guys for clearing that up for me. I wasn't planing to put them in the same unit.
With all due respect, you absolutely can put them in the same unit. Actually, with electronics, you can do anything you want to - there are no limits. Any mixing console has many preamps in addition to one or more headphone amps - works like a dream!

All you need to do is have the PS deliver sufficient current and voltage to supply the phones and you're all set.
gtzack4
Thread Starter
#9
24th January 2013
Old 24th January 2013
  #9
Gear Head
 
gtzack4's Avatar
 

Thread Starter
How do you measure current draw of a circuit? Or is this calculated?
#10
24th January 2013
Old 24th January 2013
  #10
Gear maniac
 

Quote:
Originally Posted by rcrowley View Post
Modern digital multi-meters typically measure up to 10A. And you can get quite reasonably good meters for very modest prices. For example, here is one for $15.

Note also that some equipment, particularly when driving large loads like power amplifiers, changes its current draw depending on the amplitude of the audio signal. And remember that safety margins are always a good idea. When working on DIY, amateur projects, it isn't out of line to use at least a 100% "margin". If your the total of all your loads adds up to 2 amps, then use a 5 amp power supply, etc.
Well, first I would comment that when measuring audio, your meter needs to have a wide frequency response and that it's very unlikely that a 15 dollar meter will have such a wide response. If you plan on building a lot of stuff yourself then you need to spend (big) money on a good solid pro class meter with response from at least 20Hz out to about 15 to 20kHz while measuring both AC current and voltage. A scope and a Millivolt meter are also essential.

Second, I would say that, although safety margins are good, if my max (!) load is 2 amps, then I would be happy with anything over three amps. There are a number of reasons for this - one being price, one being size, and the other being availability. If I have a 3 amp unit already in my parts box then that's the baby that's going to get used. Even if I only have a 2,5 amp unit available, I would build that one in before I went out and purchased a bigger one. The thing is, you have to look at things in some kind of reasonable fashion.
#11
24th January 2013
Old 24th January 2013
  #11
Gear maniac
 

Quote:
Originally Posted by gtzack4 View Post
How do you measure current draw of a circuit? Or is this calculated?
The simplest way to do it is to look on the data sheet for the active devices you plan to use - that is - for both the preamps and the HP amps. Second, any LED's, meters, or anything else that draws current you obviously need to include.

The problem with the "realworld" measurement method suggested by Mr. Crowley - sorry man - is that you actually have to build the thing first in order to measure it.
#12
24th January 2013
Old 24th January 2013
  #12
Lives for gear
 
Richard Crowley's Avatar
 

Quote:
Originally Posted by Vitalis View Post
Well, first I would comment that when measuring audio, your meter needs to have a wide frequency response and that it's very unlikely that a 15 dollar meter will have such a wide response. If you plan on building a lot of stuff yourself then you need to spend (big) money on a good solid pro class meter with response from at least 20Hz out to about 15 to 20kHz while measuring both AC current and voltage. A scope and a Millivolt meter are also essential.
It is rarely useful to measure audio with a DMM. Which is why a $15 meter is perfectly fine here. Because we are measuring DC current here, not AC, and not audio.

Quote:
Second, I would say that, although safety margins are good, if my max (!) load is 2 amps, then I would be happy with anything over three amps. There are a number of reasons for this - one being price, one being size, and the other being availability. If I have a 3 amp unit already in my parts box then that's the baby that's going to get used. Even if I only have a 2,5 amp unit available, I would build that one in before I went out and purchased a bigger one. The thing is, you have to look at things in some kind of reasonable fashion.
Of course, but if gtzack4 is doing this from scratch it is always better to err on the side of safety and margin of err vs. cutting it so close to the edge. Especially if attempting to predict how much power you need before even building the circuit. I was also assuming that if gtzack4 is building this from scratch, he would have some sort of bench supply or batteries or something as a temporary power source until he determined exactly how much he will need for the complete project.
#13
24th January 2013
Old 24th January 2013
  #13
Lives for gear
 
Richard Crowley's Avatar
 

Quote:
Originally Posted by gtzack4 View Post
How do you measure current draw of a circuit? Or is this calculated?
Sorry, I missed the detail that you are building these from scratch. It is not very easy to predict in advance how much power something will use before constructing it. Especially for a beginner. When creating new circuits, we typically use a piece of test equipment called a "bench power supply". Because it is used on the test bench and it can be adjustable for voltage and current, and typically it also measures voltage and current. If you don't have a bench supply, then you can use batteries, etc. during development.

If you MUST calculate the current, you can approximate it from some of the specifications for the active components (integrated circuit op-amps, etc.) But it is practically an upper-level course in electronics just to learn to read your typical spec sheet. For example, you could use the maximum power dissipation from an op-amp IC, and that would give you a worst-case figure with a large margin for error, more than Vitalis and doulos30 are recommending. But when you start getting into circuits with discrete components (transistors, etc.) then you need to really do some complex circuit analysis and run some non-trivial equations. Which is why I recommended just measuring it vs. trying to predict it.

Vitalis is 100% correct that connecting many circuits (mic preamps, summing amplifiers, output amplifiers, metering, headphone amps, etc. etc. to the same power bus is exactly how it is done virtually everywhere. There is no apparent reason for you to use multiple power supplies for the project as you have described it.
gtzack4
Thread Starter
#14
24th January 2013
Old 24th January 2013
  #14
Gear Head
 
gtzack4's Avatar
 

Thread Starter
So if I were to put a hp amp, mic pres, an eq, and a comp in the same unit, how would keep cross talk between elements?

I would think a cap on the rails for each circuit would reduce crosstalk.
Would this work? Or is the a better way to do this?
How important is it to reduce crosstalk on a unit with only mic pres or on a separate hp amp where the signal is generally the same or from the same source?
#15
24th January 2013
Old 24th January 2013
  #15
Lives for gear
 
Lotus 7's Avatar
Quote:
Originally Posted by gtzack4 View Post
So if I were to put a hp amp, mic pres, an eq, and a comp in the same unit, how would keep cross talk between elements?

I would think a cap on the rails for each circuit would reduce crosstalk.
Would this work? Or is the a better way to do this?
How important is it to reduce crosstalk on a unit with only mic pres or on a separate hp amp where the signal is generally the same or from the same source?
As rcrowley has stated, it's common practice to connect multiple audio modules to common power supply rails. If you're planning on connecting several mic pres, and several headphone amps, as well as an equalizer and compressor then it's essential that you take the advice already suggested and measure the current drain of each module (preferably at maximum operating levels). This is DC current and any digital current meter will provide accurate enough results.

You haven't stated the particular components you're considering so we're just guessing on the power requirements. Mic pres can use low power op-amps, or low-Z, high-current discrete component designs with widely different current requirements. The current required for YOUR equalizer or YOUR compressor is a total unknown. As has been suggested, measure or determine from the original circuit designer the current requirements.

Assuming that you're using good, well regulated power supplies, as you've speculated, having local PS rail bypass caps on each module is usually a good idea, and is typically employed. They are used not only to "reduce crosstalk" as you've stated (which really should not be a problem if the individual circuit modules are properly designed), but to more importantly provide a local low-impedance current source at high frequencies for each circuit module. Properly designed circuits will also include local bypass caps on each op-amp, other integrated circuits as well as some discrete components for local decoupling.

Equally important is that the power connections (including ground returns) between the individual modules [the power "rails"] should be of low resistance. That means using a heavy enough wire gauge and good, low resistance connectors for each functional module. Again, knowing the current drain of each module is important, and knowing (actually measuring) the current drain becomes more important as you add more (different) loads on your common power supplies.

There are several approaches you can use to "size" the power supply. You can actually measure the real current using a bench supply and be sure of your current requirements as rcrowley has suggested (which is the only way I'd personally do it). You can try and get worst case current load requirement specifications from the circuit designer(s) and estimate your needs or you can make a very rough estimate of the current loads by analyzing individual circuit components, which as stated is difficult at best and will have a wide margin of error in the real-world.

You can take a "shotgun" approach and make some "worst case estimates" and then double the power supply capacity for safety, but then you'll probably wind up with a supply that is larger, less efficient and costlier than necessary.

As has been correctly stated several times before in this thread, it's best to determine what your current requirements actually are (with some degree of accuracy) before attempting the project you're suggesting. If you don't feel capable of doing that simple task and don't understand the implications, it's doubtful that you should be attempting what you're proposing.
gtzack4
Thread Starter
#16
24th January 2013
Old 24th January 2013
  #16
Gear Head
 
gtzack4's Avatar
 

Thread Starter
So I need a variable bench power supply.

I don't think I would need more the +-35 volt 5amp, Maybe a 5 volt rail and a section for a power amp( say 200 volt?)

What do you guys use for your bench ps? Do you wish you had more power? Or less?

I would like to build my own unless I can get one pretty cheap on eBay. Does any one have any suggestions?
This will only be used for musical equipment.
gtzack4
Thread Starter
#17
24th January 2013
Old 24th January 2013
  #17
Gear Head
 
gtzack4's Avatar
 

Thread Starter
And thanks to everyone for all your input. It has cleared up many things for me.
#18
24th January 2013
Old 24th January 2013
  #18
Lives for gear
 
Richard Crowley's Avatar
 

#19
24th January 2013
Old 24th January 2013
  #19
Lives for gear
 
Lotus 7's Avatar
Quote:
Originally Posted by rcrowley View Post
Your desire for "pretty cheap" and "section for a power amp... 200 volt" are dramatically in conflict. ...
Not to mention that going from a typical 25 or 35 volt bench supply to something that puts out "200 volts" raises serious safety issues. Making a mistake using a 25 v. PS might result in "blowing" an IC or a resistor and a little smoke, but making a mistake using a 200v DC supply can be lethal, so a high-voltage supply must be treated with respect. Be safe!
#20
24th January 2013
Old 24th January 2013
  #20
Gear maniac
 

Quote:
Originally Posted by rcrowley View Post
If you MUST calculate the current, you can approximate it from some of the specifications for the active components (integrated circuit op-amps, etc.) But it is practically an upper-level course in electronics just to learn to read your typical spec sheet. For example, you could use the maximum power dissipation from an op-amp IC, and that would give you a worst-case figure with a large margin for error, more than Vitalis and doulos30 are recommending. But when you start getting into circuits with discrete components (transistors, etc.) then you need to really do some complex circuit analysis and run some non-trivial equations. Which is why I recommended just measuring it vs. trying to predict it.
Sorry again, but I have to disagree with you again. Assuming that an IC type power amp chip is going to be used, the "Output peak current" as well as the "quiescent current" and the "Output power" specs are clearly provided on the data sheet. It's a simple matter then to add up the number of chips or channels that you need and select the appropriate transformer or PS. If you are talking discrete devices, you absolutely have to know how to read a data sheet or else you won't be able to even build the thing you are trying to measure. Engineering is all about calculating and if the OP doesn't have engineering skills then he will never build one single thing - let alone measure it.

Which actually brings me to another topic and, that is, that just because somebody wants/needs a certain piece of gear, well, that's miles away from actually being able to build it. So many people come here thinking it's easy, but it's not. And Lord help you if you actually want the thing to look good when it's finished because that takes a whole different set of skills. I would be willing to bet that 99 percent of the stuff people want to build - never gets built. The fact is that there are no shortcuts and that you do have to understand electronics as well as being able to measure and cut an electronic enclosure without messing it up. All this takes years of practice and there is no way to simply walk somebody through it - especially a novice. So my basic advice really is: If you want it, go buy it and if you can't buy it then do without it because you will never be able to build it anyway.
#21
25th January 2013
Old 25th January 2013
  #21
Lives for gear
 
Richard Crowley's Avatar
 

That's rather amusing. Vitalis claims to disagree with me, and then proceeds to restate exactly what I said.
#22
31st January 2013
Old 31st January 2013
  #22
Gear maniac
 

Quote:
Originally Posted by Richard Crowley View Post
If you MUST calculate the current, you can approximate it from some of the specifications for the active components (integrated circuit op-amps, etc.) But it is practically an upper-level course in electronics just to learn to read your typical spec sheet. For example, you could use the maximum power dissipation from an op-amp IC, and that would give you a worst-case figure with a large margin for error, more than Vitalis and doulos30 are recommending. But when you start getting into circuits with discrete components (transistors, etc.) then you need to really do some complex circuit analysis and run some non-trivial equations. Which is why I recommended just measuring it vs. trying to predict it.
Sorry, man, but I still disagree with you for the following reasons. First you say;

"if you MUST calculate the current...".

What I'm saying is that you MUST (LoL) calculate the current because there is no other way to do it. You can't build the thing first and then measure it because how will you know how to build it if you haven't calculated it first? Measurement is for a circuit that already exists. As I understand, the OP desires to build the circuit from scratch and not simply measure an existing one - or maybe I misunderstood him.

You also say;

"For example, you could use the maximum power dissipation from an op-amp IC, and that would give you a worst-case figure with a large margin for error..."

The data sheet does not leave anything to guess work, but gives you the Minimum, Typical, and Maximum values for all specifications. Understanding these values will tell you whether that part is actually correct for the application or not.

You also say;

"But when you start getting into circuits with discrete components (transistors, etc.) then you need to really do some complex circuit analysis and run some non-trivial equations."

On a DIY level, what you would do is search the web, or where ever, for a schematic that you believe can do the job that you need to be done. At that point you can began to collect the data sheets for all the semiconductors and look them over, paying special attention to max voltages, but also to those parameters which are important to your project, for example, audio performance, heat dissipation, part size, cost, and so on. Your advantage is that the engineer who designed the circuit will have done all the complex circuit analysis, so all you have to do is copy it and package it. The question is if you have the skills to copy it and package it.

And lastly you say;

"Which is why I recommended just measuring it vs. trying to predict it."

Here again I say that your biggest asset is your ability to predict how a circuit will perform before you actually build it; you have to plan it from top to bottom. After you plan it, you draw up a schematic and build the circuit according to your - or someone else's - schematic.

Hope that clears up my side of things.
#23
31st January 2013
Old 31st January 2013
  #23
Lives for gear
 
Richard Crowley's Avatar
 

Quote:
Originally Posted by Vitalis View Post
What I'm saying is that you MUST (LoL) calculate the current because there is no other way to do it. You can't build the thing first and then measure it because how will you know how to build it if you haven't calculated it first?
Sorry, that simply makes no sense at all. Of course you can build a circuit and THEN measure it. Many of us here have been doing this for decades. Our reality does not agree with your theory.

Quote:
The data sheet does not leave anything to guess work, but gives you the Minimum, Typical, and Maximum values for all specifications. Understanding these values will tell you whether that part is actually correct for the application or not.
The question at hand is NOT component selection. The question is power requirements.

Quote:
On a DIY level, what you would do is search the web, or where ever, for a schematic that you believe can do the job that you need to be done. At that point you can began to collect the data sheets for all the semiconductors and look them over, paying special attention to max voltages, but also to those parameters which are important to your project, for example, audio performance, heat dissipation, part size, cost, and so on. Your advantage is that the engineer who designed the circuit will have done all the complex circuit analysis, so all you have to do is copy it and package it. The question is if you have the skills to copy it and package it.
Apparently the reason we are disagreeing is that we think we are attempting to answer different questions. My assumption (reinforced by the very title of this thread) is that the gtzack4 is asking about how to power his circuits.

Quote:
Here again I say that your biggest asset is your ability to predict how a circuit will perform before you actually build it; you have to plan it from top to bottom. After you plan it, you draw up a schematic and build the circuit according to your - or someone else's - schematic.
Perhaps. But the gtzack4's questions caused me to assume that he is nowhere near that point in his electronics experience. Which is why I am attempting to offer pragmatic and useful suggestions like simply measuring it.
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
jeff deff / Geekslutz forum
59
tamasdragon / High end
87
Pfhuck / Remote Possibilities in Acoustic Music & Location Recording
6
soundguykyle / So much gear, so little time!
3
parky / Remote Possibilities in Acoustic Music & Location Recording
3

Forum Jump
 
Register FAQ Search Today's Posts Mark Forums Read

SEO by vBSEO ©2011, Crawlability, Inc.