Dynamic mic pre-amp into sound card

[hawk2eye]

Why?, and what are you defining as 'low level'.

And it's not 'my' technique, it's EVERYONE'S technique - can you name a single audio input where impedance matching is used?. If you're thinking of suggesting microphone transformers?, they are simply used to change the impedance, which is then fed into a much higher impedance as normal.

....................................................................................
Nigel Goodwin

Apparently you have not been involved with low level, noisy signals very much.
I am correct and EVERYONE's technique is wrong.

I have emphasized the words "impedance matching"
with ...vvv... marks.

see:
**broken link removed**


.................||||||||||||||||||||||
.................vvvvvvvvvvvvvvvvvv.................
< Excellent impedance matching between the microphone and the pre-amplifier,
< excellent balance and common mode rejection ratios (CMRR), very low distortion,
< and superior bandwidth

see:
**broken link removed**


hawk2eye

 

[Nigel Goodwin]

Apparently you have not been involved with low level, noisy signals very much.

Why would you have noisey signals?.

I am correct and EVERYONE's technique is wrong.

I have emphasized the words "impedance matching"
with ...vvv... marks.

.................||||||||||||||||||||||
.................vvvvvvvvvvvvvvvvvv.................
< Excellent impedance matching between the microphone and the pre-amplifier,

So you're basing your suggestion on an advertising flyer for microphone transformers?.

As I said previously, this ONLY matches the microphone to the transformer, and NOT the transformer to the preamp, not even back in the 1940's where that technology comes from!.

Presumably the advantages you're talking about are from using a BALANCED microphone output? - where any common mode 'noise' on the long cable is cancelled out in the transformer?. It's only an advantage for long cables in poor environments, and 'modern' (last 20-30 years?) electroniclly balanced inputs far out perform transformers. Transformers have too many drawbacks, and are far too expensive to be in common use - modern electronics have rendered them fairly obselete.

Almost any modern PA gear has electronically balanced inputs, and this helps to reduce any noise pickup on the cabling.

 

[hawk2eye]

Why would you have noisey signals?.
As I said previously, this ONLY matches the microphone to the transformer, and NOT the transformer to the preamp, not even back in the 1940's where that technology comes from!.
++++++++++++++++++++++++++++++++++
Nigel:
I repeat:
the mike impedance to the amplifier is (Npri/Nsec)^2*Rgen
as shown by ANY text book that describes impedance matching.

Look it up. there is nothing wrong in being wrong.
............................................................................
The input transformer serves as an impedance matching device.

mathematically, for the transformer

Nout*Iout = Nin*Iin
Eout/Ein = Nout/Nin
then
Eout = (Nout/Nin)Ein
Iout = (Nin/Nout)Iin

Rout = Eout/Iout
Rout =(Nout/Nin)^2 *Ein/Iin
Rout =(Nout/Nin)^2 *Rin

hawk2eye

 

[audioguru]

I worked with very high quality PA amplifiers. They used an input transformer for dynamic microphones as a balanced connection for them. The transformers were used at impedances far higher than the mics and as stepup transformers to reduce the noise of the following opamp since then it didn't need such a high gain.

 

[Nigel Goodwin]

Why would you have noisey signals?.
As I said previously, this ONLY matches the microphone to the transformer, and NOT the transformer to the preamp, not even back in the 1940's where that technology comes from!.
++++++++++++++++++++++++++++++++++
Nigel:
I repeat:
the mike impedance to the amplifier is (Npri/Nsec)^2*Rgen
as shown by ANY text book that describes impedance matching.

Look it up. there is nothing wrong in being wrong.
............................................................................
The input transformer serves as an impedance matching device.

mathematically, for the transformer

Nout*Iout = Nin*Iin
Eout/Ein = Nout/Nin
then
Eout = (Nout/Nin)Ein
Iout = (Nin/Nout)Iin

Rout = Eout/Iout
Rout =(Nout/Nin)^2 *Ein/Iin
Rout =(Nout/Nin)^2 *Rin

And I repeat again as well, what has that got to do with the matching INTO the preamp - quoting formulas from a book doesn't help anything, do you have actual examples of microphone transformers being impedance matched to the preamp input impedance?.

I worked with very high quality PA amplifiers. They used an input transformer for dynamic microphones as a balanced connection for them. The transformers were used at impedances far higher than the mics and as stepup transformers to reduce the noise of the following opamp since then it didn't need such a high gain.

No disrespect Audioguru, but how long ago was that?. Using a VERY high quality transformer on the input stage of a VERY expensive mike stage is probably still the best method available (although it still doesn't use impedance matching to the preamp). However, it would increase the costs VERY substantially for a very small level of improvement. Cheap, and medium price, mike transformers have performance levels below modern electronics.

 

[audioguru]

Hi Nigel,
I worked with TOA brand, a Japanese sound systems company, until 6 years ago. The expensive amplifiers have a high quality input transformer for perfect balancing and for a voltage stepup which feed a very low noise opamp. Although they were small, the transformers have a good low frequency response because they are used with their impedance unmatched.
Their cheaper amplifiers use transistors as an almost balanced input.

 

[Nigel Goodwin]

Hi Nigel,
I worked with TOA brand, a Japanese sound systems company, until 6 years ago. The expensive amplifiers have a high quality input transformer for perfect balancing and for a voltage stepup which feed a very low noise opamp. Although they were small, the transformers have a good low frequency response because they are used with their impedance unmatched.
Their cheaper amplifiers use transistors as an almost balanced input.

TOA? - no disrespect, but the TOA PA gear I've seen isn't anything wonderful (and I've repaired quite a lot of it!). Presumably you're talking about the same sort of gear?, vocal PA, and NOT quality audio for musical PA?. There's no need for any great quality from it, so a cheap transformer wouldn't be any drawback.

I suspect their only reason to use a mike transformer is so they can charge a higher price?, and it's certainly pretty pricey, but all that vocal PA gear is - I've never been able to understand why?, it's certainly NOT because of design or build quality - perhaps just small sales figures?, or because they are selling to industrial users?.

BTW, just checked the mike preamp spec on a cheap Behringer mixer:

Input impedance 2.6K balanced. (for 50, 150, and 600 ohm mikes)
Frequency response 10Hz to 200KHz -3dB. (why bother to 200KHz?)
Signal to noise ratio 110dB
Mic E.I.N. (20Hz to 20KHz)
@ 0 ohms, 134dB
@ 50 ohms, 131dB
@ 150 ohms, 129dB

Pretty impressive specs, although a GOOD transformer will beat it on CMR, which interestingly they don't quote!.

EDIT: BTW Audioguru, you mention the voltage step-up of the transformer, have you ever actually measured how much that is?, or the actual impedances of the transformer?, if not I suspect you will be quite suprised if you do!.

 

[audioguru]

The cheap TOA mixer has recently been discontinued. Its mic input S/N spec is 126dB @200 ohms. I don't know the ratio of the input transformer.

Their products were pretty reliable. Very few items were returned during the 2 years to 5 years warranty. Their best power amps had a "no questions asked" 5 years warranty and none came back for repairs. They had 1500 dealers in Canada.

 

[Nigel Goodwin]

The cheap TOA mixer has recently been discontinued. Its mic input S/N spec is 126dB @200 ohms. I don't know the ratio of the input transformer.

Their products were pretty reliable. Very few items were returned during the 2 years to 5 years warranty. Their best power amps had a "no questions asked" 5 years warranty and none came back for repairs. They had 1500 dealers in Canada.

You're a knowledgeable, experienced man - what is your personal opinion on the actual design quality, and build quality, of the TOA gear - assuming you're prepared to give it here?, no problem if you would rather not!.

None of it has ever impressed me, but the usage doesn't really impress me either, and it doesn't require any great specifications for it's intended use.

I've always been VERY interested in high power audio PA, for band and disco use - and much of that is very well designed and constructed, but it has a MUCH harder life than vocal announcement type PA.

 

[audioguru]

Hi Nigel,
TOA designs are conservative tried and proven ones, nothing new. The high end amplifiers and speakers are used in live show theaters and sound excellent. I heard some background music systems in stores that sound excellent.

 

[Dr.EM]

I was quite suprised at how good the live sound was on a Massive Attack concert I went to recently. Hadn't seen an outdoor one before and wondered if the bass would come across well. It was in fact the most powerful bass i've ever heard, literally "trouser flapping" and the vibrations in your chest from it actually made it difficult to breathe . The overall quality of sound was good too, a bit too bass heavy, but at least it wasn't bright and nasty, my ears were fine after.

Generally, in shops the sound is poor. I find the best is when they use Hi-Fi speakers and multichannel amps, though this only works well in smaller shops. Virgin use compact tannoy speakers and they appear to use a lot of amplifiers for them all rather than 100v distribution, they sound respectable. In our local HMV they have a good system on the top floor. It is a pair of big club speakers, they look cheap and are made of plastic but used to sound very good. Now though someone has altered the settings and it has a horrible boomy bass boost sound to it. The speakers look far to big for the room and are mounted on the walls up high but it did manage to sound pretty good.

 

[Nigel Goodwin]

I was quite suprised at how good the live sound was on a Massive Attack concert I went to recently. Hadn't seen an outdoor one before and wondered if the bass would come across well. It was in fact the most powerful bass i've ever heard, literally "trouser flapping" and the vibrations in your chest from it actually made it difficult to breathe . The overall quality of sound was good too, a bit too bass heavy, but at least it wasn't bright and nasty, my ears were fine after.

Know what you mean about feeling the bass they always run too loud at Donut gigs (at least for me!).

You have to bear in mind the massive power they run at these gigs!, I remember reading an article years back about the PA for an ABBA tour of Australia. The first thing they had to do was order (well in advance) an independent 3 phase mains supply to be provided, in order to cope with the demand - they also always carried sacks of salt and large metal spikes, so they could make their own earth system if they didn't like that provided. All the amps (banks and banks of them) were top end, Crown and such. Lighting and sound were on different phases, to help prevent any noise from the lighting getting to the amps.

Powers are even higher (and cheaper) these days, if I remember correctly the first band to run 1,000,000 watts was the Rolling Stones - late 60's early 70's?.

 

[hawk2eye]

impedance matching and noise

< And I repeat again as well, what has that got to do with the matching INTO the preamp - quoting
< formulas from a book doesn't help anything, do you have actual examples of microphone
< transformers being impedance matched to the preamp input impedance?.
[end QUOTE =Nigel Goodwin
.................................................................................
In my final attempt to show that impedance matching is
required for best noise performance the ultimate in low level
amplifiers, I include the following from respected companies
on the Internet
..................................................
1. https://www.electro-tech-online.com/custompdfs/2006/09/an002.pdf
see right side of p2
..............................................
2. **broken link removed**
3. **broken link removed**

.............................................
4. **broken link removed**
quote
Do's and Don'ts of Audio Transformers

Do use a transformer to match impedances.
Do use a transformer to increase or decrease signal level by up to 25 dB.
Do use a 1:1 transformer to isolate problem components in an audio chain.
Do not use a transformer to increase signal level by more than 25 dB.
...............................................
5. **broken link removed**
6. http://www.midi-classics.com/c/c4283.htm
++++++++++++++++++++++++++++++++++++++++++++++++++
It is important to keep an open mind.
Especially when there is so much evidence
that apparently negates what we are sure we know.

hawk2eye

 

[audioguru]

Hawk2eye,
Your 1st example has an 80 ohm source connected to the 42k impedance of an input transformer. That is far from being matched.

 

[Dr.EM]

Know what you mean about feeling the bass they always run too loud at Donut gigs (at least for me!).

You have to bear in mind the massive power they run at these gigs!, I remember reading an article years back about the PA for an ABBA tour of Australia. The first thing they had to do was order (well in advance) an independent 3 phase mains supply to be provided, in order to cope with the demand - they also always carried sacks of salt and large metal spikes, so they could make their own earth system if they didn't like that provided. All the amps (banks and banks of them) were top end, Crown and such. Lighting and sound were on different phases, to help prevent any noise from the lighting getting to the amps.

Powers are even higher (and cheaper) these days, if I remember correctly the first band to run 1,000,000 watts was the Rolling Stones - late 60's early 70's?.

1000W surely? a million watts is madness! I did wonder how much power they used at that concert, and then it's not just the sound but lights, equitment etc. I didn't see any generators so presumably they had a special supply, or they were round the back somewhere out of sight.

 

[audioguru]

1kW of audio power sounds twice as loud as 100W.
In TOA's demo room I had a subwoofer with two very efficient 15" drivers connected to a bridged amp producing 1800W cleanly at max and got clipping sometimes. The sound was so clear that it didn't seem like so much power.

 

[Nigel Goodwin]

1000W surely? a million watts is madness!

1000W is common place, my small Behringer PA is 800W - I can't remember the year the Stones first used 1,000,000 watts, but it was a LONG time ago, certainly 70's and may have been 60's.

I would expect many modern club disco systems to output in the 10's of kilowatts?.

Here's a quick 'modern' quote I found:

"The Rolling Stones are also one of the most agile organizations in the world. They move 250 employees with every concern, have the capacity to generate four million watts of electricity, lay eight miles of sound and electric cable at every venue, assemble the world’s largest television screen, position 1,500 spotlights, and build a stage so tall that it actually requires aircraft warning lights under U.S. Federal Aviation Administration rules, all in record time."

What do you think they do with 4 megawatts of electricity

I did wonder how much power they used at that concert, and then it's not just the sound but lights, equitment etc. I didn't see any generators so presumably they had a special supply, or they were round the back somewhere out of sight.

As I mentioned earlier, the ABBA concert arranged for their own supplies to be laid on - but any regular venue would probably have it permanently there.

As you say, a generator is a possibility, a truck sized one can generate an enormous amount of power, and they can be whisper quiet.

 

[Nigel Goodwin]

It is important to keep an open mind.
Especially when there is so much evidence
that apparently negates what we are sure we know.

All those are literature from transformer manufacturers, and none seemed to suggest feeding into a matching impedance preamp?. What they did suggest, is a loading resistor on the output of the transformer (to reduce ringing in the transformer), this then feeds in to a higher impedance input preamp - as all valve microphone preamps did.

I'm quite prepared to keep an open mind, but you've yet to give a single example of matching source impedance to preamp input impedance in ANY microphone preamp?.

 

[systemloc]

I tried out building a new circuit around the LM386, which I found at the RadioShack, since everyone pretty much agreed an op-amp would work out better. I checked out it's , to come up with my circuit. The 20x gain the LM386 defaults to is reasonable, but I think I'll modify it with a rheostat to tweak it at will. I have two questions that strike me immediately. First, the input 10uF cap is electrolytic, and the second page referenced shows it with the positive towards the input. Looking at the internal schematics LM386, this seems acceptable since that side goes to ground. Is that acceptable? Secondly, what's the function of the 0.1uF cap and 100 ohm resistor to ground on the output? The original spec was for a 0.05uF cap, but I substituted it for 0.1uF which I had on hand. I'm guessing this circuit acts as some sort of filter to lessen noise. Is this the case, and is my substitution acceptable?


In response to the discussion of impedance coupling, for microphones, at least, wouldn't having matched impedance create a load on the mic itself, thus causing distorted input?

TJ

 

[audioguru]

I tried out building a new circuit around the LM386, which I found at the RadioShack, since everyone pretty much agreed an op-amp would work out better.

It isn't an opamp. It is a power amp. It is pretty noisy if it has a gain more than 20.

the input 10uF cap is electrolytic, and the second page referenced shows it with the positive towards the input. Looking at the internal schematics LM386, this seems acceptable since that side goes to ground. Is that acceptable?

The input DC voltage of the LM386 is 0V so if the capacitor is fed from a grounded volume control then it doesn't matter which way the capacitor's polarity faces.
At first I thought 10uF is way too high a value. Then I realized that it would help reduce noise if the volume control is 1k to 10k ohms.

what's the function of the 0.1uF cap and 100 ohm resistor to ground on the output?

It is 0.05uF and 10 ohms. 0.1uF might work but I would use 0.047uF instead. It stops the LM386 from oscillating at a high frequency when the inductance of a speaker causes a high load resistance, like your application without a speaker.

In response to the discussion of impedance coupling, for microphones, at least, wouldn't having matched impedance create a load on the mic itself, thus causing distorted input?

A matched impedance doesn't cause distortion, it causes the output to be half.