R/E/P Community
R/E/P => Klaus Heyne's Mic Lab => Topic started by: tfish1257 on July 14, 2014, 04:06:22 AM
-
Hi all, new to the forum. Im also due to investing in recording equipment. I've spent my career as a musician on a pretty much no-budget recording method that involved playing straight into my laptop, sitting in my bedroom. While sometimes the flat sound can be endearing in its own flawed way, it makes it hard when Im trying to do full-band arrangements because my computers stock mic just doesnt have the full dynamic of sound.
Granted, i enjoy music that has a 60's retro sort of sound to it, or lo-fi analog, etc., but Im just wondering what mic I should consider buying that is affordable (like $100, yknow, for a poor kid such as myself) that i can make presentable pieces of music with. A mic capable of capturing vintage analog sound would be ideal, but like i said i dont know how much of that matters in regards to the mic as opposed to the software a recording is run through after the fact.
Any suggestions would be awesome. Cheers.
-
Can you fill me in what 'vintage low-fi analog sound' you are after? You don't by any chance meant the sound that was created with a Pultec, an LA 2A, a U67, into a Trident console, and cut on a Neumann lathe?
-
$100? I'd invest in a library card.
-
...where you still have $100 left afterwards, because library cards are free.
So with that $100.- I would buy two sweet spot tickets to a Carnegie Hall (or local equivalent) orchestral performance and enjoy a glass of Champagne with my sweetheart during intermission.
-
Or buy an SM 57(58) & wait for more money to buy the interface you're going to need to plug it into.
(Don't worry about "vintage" right now.)
-
Better yet, audition the Sennheiser e835 against other microphones in this range at your local Guitar Center or Best Buy.
-
...this from a former valued Sennheiser employee (truth in advertising here!)
-
... who is not responsible for making this model outsell any other in this price range ...
The relevant point is 'AUDITION' various microphones in your price range and trust your ears, rather than following blindly someone else's taste and preference.
-
Ears need to be trained before they can be trusted.
No one starts out at the top of their field.
-
The replies in this thread might be rather confusing to the original poster, if he's still with us. I think he asked an honest question and I'd like to try offering a more direct response, in case we haven't scared him off.
Basically, I don't think most people on this forum consider "vintage" to be a style or genre of sound, or that the term conveys any one particular feeling or impression that can be singled out for simulation or emulation. That very concept probably comes across to some people here as an insult, in fact. It's not as if one particular vintage microphone or console has "the" sound quality that all other equipment should strive to match, if that were possible to do.
Instead, I'd say that many/most people here sincerely feel that the best older equipment--in all its diversity--sounds better than most (though by no means all) of what's been made since--even though there is no one consistent, identifiable feature to that superiority. And while much of today's equipment performs measurably better in various respects than the equipment of half a century ago, these people don't respect equipment that they feel is designed primarily to conform to such measurements.
(I don't actually happen to share that viewpoint myself, but I'm trying to characterize it as fairly and respectfully as I can, so that the person who started this thread can get something from having posted here--even if it's not what he expected.)
As it happens, quite a bit of audio equipment now on the market, in all price ranges, does claim to recapture some aspects of vintage sound. Some of that equipment even sounds OK sometimes. So the original poster was on to something real with his question. But equipment specifically made for the purpose of having a supposedly "vintage" sound isn't what people mostly talk about here--or if so, not usually in a positive light.
--best regards
-
I think there is a fairly simple answer to the poster's question: You cannot get audio quality in microphones (or any other audio gear) for peanuts. This business model simply does not exist. We can argue how much truly better more expensive microphones sound than less expensive ones, but I cannot think of a single model where, in comparison, another model at a fraction of its price reaches the same level of performance. It does not exist.
I apologize for some of the snippy, less than helpful answers, but to most who frequent and post on this forum it is quite obvious that at $100 there is no serious contender for a place in a professional microphone arsenal. No matter what advertising or home recordist forums may suggest.
There is a direct correlation between quality and price, as with few other consumer goods (wine is another field where price portends quality, and for the same reason): no superior mic is sold cheaply, and no cheap mic can fetch a high price. The market regulates that correlation between perceived quality and price quite well in the field of products that appeal to the senses: wine, cameras, audio gear...
I don't need to mention names, but a few times in the past copy mics were launched and artificially priced at levels quite close to those of the original vintage mics these products were copying, with the result that, as soon as the mics hit the open, free market, used and even new prices dropped significantly and quickly- a perfect example of self regulation in a market.
-
Some of my favorite personal recordings were made when I was younger and had no money to spend on Neumanns or whatever brand of nicer mics. I was lucky to have an Ampex 350 machine and a home-brew mixer. I had a few Shure, EV dynamics. My favorite at that time was my Norelco D-24. When I listen to those recordings today, they still have that "something" that sets them apart from how I record these days. And, these days I have the nicer mics that I couldn't afford then. It's more about the passion and the performance than it is the mics. It helps to know how to use what tools you have.
-
Im just wondering what mic I should consider buying that is affordable ... like $100
I would vote for a Shure SM57, original, non of the copies that are around that are said "sounding the same" (they do not).
You can't go wrong with that mic, it's a real classic and it's universal:
voice, el.- and acc. guitars, drums, percussion, ... - almost everything sounds good with it.
For voice get youself the foam windshield with the plastic ring that is available for it.
If you want to see how that looks - the US president has two of those in front of him when he speaks to the public.
Regards
Kai
-
A Shure SM57 is great advice for that price range. Thanks, Kai.
But let's keep die Kirche im Dorf: cumulatively, the SM57 ALONE will whittle down high end pretty quickly in a multi-track recording: with a -3dB point @ 11KHz at best, there won't be much liveliness left piling up a bunch of midrange-heavy tracks.
-
...with a -3dB point @ 11KHz at best...
Here's a measurement I made, using the substitution method.
Reference is a Bruel&Kjaer 4165 mic, source a Tannoy 6", distance 25cm.
Of course it's not a condenser, but there's usable frequency content up to 16kHz.
This doesn't say anything about the sound quality of this frequencies.
-
An alternative is one of those cheapo Chi-com mics with a replacement capsule and perhaps replacement circuit parts or pcb.
The mics tend to sell for around $50~60 or so. The capsules (k-47, c-12, k-67, etc,) sell for around $100. You will need to invest about $200 total into this if you do your own work.
I have about 15 of the older MCA SP-1 mics I bought for $39 each. Each has a k-47, k-67, k-87, k-7, k-1 and c-12 capsules, all are usable and sound very good for the $ invested.
If those capsules are not good enough, there are plenty of higher priced versions available as well. Once you have the mic bodies set up, you can swap capsules all day long.
-
Maybe look at what the TS is using now will make a suggestion to improve the quality of his recordings within budget easier.
What about a nice omni dynamic mike from a well known manufacturer like Sennheiser?
The MD 21 will no doubt be an improvement on what he is using now.
These mikes go for 50- 100 usd in Europe. They were standard equipment in the sixties and seventies for reporters using Nagra III an IV portable recorders.
Nothing against the SM 57 but the old Sennheiser beats the Shure mike anytime.
Vintage is a term meant for items going back to the twenties of the last century.
-
If this thread hasn't grown cold, I'd like to comment on Kai's frequency range analysis of the Shure SM-57. First of all, how representative of ALL SM-57's is the picture? It may be fairly consistent throughout the production run, but without statistical sampling it is not necessarily indicative of what the consumer may find.
And as it has been noted here before, one should be wary of over-reliance on performance stats from the mfg'r as predictive- I have done it myself since I started devouring such specs in the early 60's.
Yet I still believe that performance data can be helpful, it's just that as mic consumers we don't receive it. I'm talking about a distortion curve over the usable audio spectrum, and figures on transient response. Published distortion data is almost universally provided as that amount of THD found at 1000 hz. My Neumann mics almost without exception carry published specs of .05% at that frequency. Comparing this to figures offered by other manufacturers makes me feel all warm and fuzzy.
But I wonder: what's this or that mic doing down around 100 hz, or 400, or- well you get the picture. And what kind of distortion? Even-order or odd-order? Which harmonics? And then- is distortion of a certain kind at a given frequency of a particular amount bad- or is it good?
We have come to learn that the Gain knob on some pieces of gear increases negative feedback which increase marginal (or greater) amounts of distortion, and we think we like it in some cases. Yet if we knew what kind, and how much, and where, we could make more informed decisions about just why we may like or dislike one mic versus another.
Transient response is another performance point. Neumann via their forum has provided me with a technical article discussing transient response in mics, particularly comparing TR in condensers (which?) to dynamics (again, which?). Despite all kinds of issues that could be raised about brand, etc., I think it safe to say that the article pretty well makes the point that condenser performance versus dynamic performance is pretty impressive.
Transient response data is available. While I have never owned or used any of their mics, Earthworks publishes transient response data. All well and good, and I hope they continue to do so, but until industry leaders like Neumann/Sennheiser get on the TR bandwagon, comparability is not possible.
One last observation re specs: Neumann also told me that they lament the fact that there really are no industry standards for performance data. So...until there are standards, we should really take it all with a grain of salt.
-
Complete THD tests are not available from any mic manufacturer. The reason I leave to their marketing departments, but it does ring of the current trend of "reliance on the stupid voter".
THD vs amplitude at 100, 1k and 10k plus THD vs frequency sweeps would tell many tales. Then there are the IMD tests that reveal transient distortions. Most would not be impressed by the results.
Many capacitor mics have THD issues in the low end, Neumann included. It's generated by an insufficient input impedance. AES had an article on this back in the 1990's. 200 meg ohms in a "vintage" tube mic will generate a high level of THD at 20 hz. 1 gig ohm lessens that, it's reduced to .001% at 10 gig ohms.
I have modified AKG 460 bodies here with a 10 gig ohm input impedance. Those measure .001% THD at 20 hz, best I've measured with the Audio Precision, it can be done.
"Industry standards" are created by the industry. All it takes is a few notable manufacturers getting together and setting those standards. Their apparent lack of interest in accomplishing that leads one back to "reliance on the stupid voter".
-
I'd like to comment on Kai's frequency range analysis of the Shure SM-57. First of all, how representative of ALL SM-57's is the picture? It may be fairly consistent throughout the production run, but without statistical sampling it is not necessarily indicative of what the consumer may find.
I had taken an average sample of the 10 or so SM57 that I have, partly from different prod. runs.
There are differences, mainly in the HF range and the 6 kHz peak, but not too big.
...the Gain knob on some pieces of gear increases negative feedback which increase marginal (or greater) amounts of distortion, ...
Turning up gain (louder) REDUCES negative feedback in usual mic amp designs, and yes, distortion can become slightly higher then.
The effect is not very significant (in high quality amps) unless you set up quite high gains.
Transformer coupled designs have an advantage here, as the transformer delivers a voltage gain of up to 26 dB for free (I don't mean $$$, good transformers are costly), so the active electronic part has an easy life.
Regards
Kai
-
Kai, I appreciate your observations, and what you observe with the Shure 57's is more meaningful having tested more than one. My point however, is that manufacturers have ( or at least, should have) meaningful performance data of the kinds I have mentioned. Furthermore, I believe it would be extremely helpful to conduct trully broad- baded tests on the classic pantheon of mics-u47. U49, C12, etc. seeking audio sprectrum harmonic distortion tests at various SPL's, types of noise sources (pink, white, etc.) at varying distances from the source, plus transient response data.
At some point I think we could further identify statistical attributes of what makes us like this or that mike.
In fact, I volunteer to conduct the tests. Just send me your idle U47, U49, C12. Will probably take a little while, but they will be returned. I promise.
-
Turning up gain (louder) REDUCES negative feedback in usual mic amp designs, and yes, distortion can become slightly higher then.
The effect is not very significant (in high quality amps) unless you set up quite high gains.
Transformer coupled designs have an advantage here, as the transformer delivers a voltage gain of up to 26 dB for free (I don't mean $$$, good transformers are costly), so the active electronic part has an easy life.
Regards
Kai
Most transistor mic amp designs don't vary feedback to adjust gain, they are mostly instrumentation amp topologies with gain adjusted between the sections. Some exceptions are older Trident transformer coupled designs that do vary feedback for gain adjustment. Those cause all sorts of variables in secondary loading creating ringing at lower gains and drooping at higher gains.
Transformer voltage gain is offset by insertion losses. -2.5 db is normal. Those topologies offer a maximum noise spec of -127 db EIN using a low ratio 1/2 design. Low noise transistor trans-amp topologies offer much better noise performance, up to -133 db EIN at 50 ohms. THD is also lower as low frequency transformer THD is avoided.
-
Low noise transistor trans-amp topologies offer much better noise performance, up to -133 db EIN at 50 ohms.
I've yet to see a real world (not theoretical) mic amp with -133 dB Equivalent Input Noise (RMS, unweighted) coupled to a 50 Ohms source.
BTW: 50 Ohms does make less sense than 200 Ohms, representing a dynamic mic.
When a condenser mic is connected, the mic's internal amplifier's noise is the dominant noise source anyway.
BTE: 2.5 dB insertion loss of the transformer is a far from an ideal design. There are designs with 5K input impedance (almost no loss) and still 26 dB (1:40) step up ratio, e.g. in the Mindprint DTC.
To come back closer too the oiginal question - in usual setups all these technical parameters are of minor importance.
In my opinion you cannot judge a mic's sound on measurements.
The frequency response gives you an idea of the "voicing" of the mic.
Most other measurable parameters are hardly, if at all, correlating to a specific sound.
Some, like max. SPL or directivity, can still be interesting to pre-judge if a mic will work OK in a specific situation.
Finally a simple A/B(/C/D) test with the candidates will usually show immedeately which one is better for the job to be done.
And then in comes experience - good technicians know what works, giving a good starting point for possible improvements.
SM57 works (in this price range)!
Sennheiser sometimes, sometimes not, but not the ones for 150 bucks.
My 2 cent.
Regards
Kai
-
I've yet to see a real world (not theoretical) mic amp with -133 dB Equivalent Input Noise (RMS, unweighted) coupled to a 50 Ohms source.
BTW: 50 Ohms does make less sense than 200 Ohms, representing a dynamic mic.
When a condenser mi is connected, the mic's internal amplifier's noise is the dominant noise source anyway.
BTE: 2.5 dB insertion loss of the transformer is a far from an ideal design. There are designs with 5K input impedance (almost no loss) and still 26 dB (1:40) step up ratio, e.g. in the Mindprint DTC.
Regards
Kai
Most transformer coupled mic designs use a 150 ohm impedance, Europeans selected 200 ohms. Many transistor output mics are 50 ohms, some less. In that regard mic amp EIN specs at 50 ohms are useful. Some transistor mics have extremely low self noise around 3~4 db, couple that with a low noise mic amp and all the noises heard are random air movements, not electronic hiss.
150 ohms source impedance will give you a -129.6 db EIN, not too bad, better than can be done with a transformer like a JT-16 Jensen at 1/2 ratios. Run that at 50 ohms and it degrades as the noise matching of the transformer is lost. Bill Whitlock and Dean Jensen have AES papers on this subject if any readers want to delve deeper into that subject.
As to real world results, yes there are designs that achieve -133 db EIN at 50 ohms, The Audio Upgrades High Speed Mic preamp does -136 db with a shorted input, -133 db EIN unweighted that shows the true noise contribution of that design. All the noise that is heard are source contibutions from the transducer, the microphone. Transformers are not used as they would degrade those noise specs and severly decrease bandwidth and slew rates while adding low frequency THD.
Audio Precision measured THD is .0005%, IMD is 1.5 ppm, bandwidth is 30 mhz, slew rate is 2000V/us, output current is 110 ma, input impedance 4.5k ohms. It is an all current-feedback design that holds it's 30 mhz bandwidth even at +60 db of gain, all other voltage feedback mic amp designs will show a decreasing bandwidth with increasing gain.
-
...The Audio Upgrades High Speed Mic preamp does -136 db with a shorted input....holds it's 30 mhz bandwidth even at +60 db of gain
These specs are astronomic!
To come back to earth and closer to the topic - noise isn't too much of a problem today.
To have an advantage of such an, probably expensive, micamp you need:
- A very quiet room (no outside noise creeping in, no aircondition).
- The use of an dynamic or ribbon mic (the mentioned extreme low noise transitored mic's usually have a very high output, so the micamp isn't taxed too much).
- A very quiet source, otherwise you wouldn't need enough gain to make noise come into play.
Let's see what contributes to a good sound (in a given room with a certain artist):
The selection of a proper mic and good positioning of it, once you've done that you're 20% there.
Another 20% comes from the processing: EQ, compression, reverb etc., whatever is necessary.
Further 35% you get from working with the artist, giving him or her an inspiring headphone mix, give him the right feedback on his performance etc. etc. ... the whole story of conducting a recording.
14% you get from choosing the best takes and parts out of several runs.
Oops - only 1% left for the mic-amp ;D
Depending on the performer and musical style the percentages may be distributed a bit different - plus - your milage may vary.
But in fact the artistic work is about half of the story.
Best regards
Kai
-
........BTE: 2.5 dB insertion loss of the transformer is a far from an ideal design. There are designs with 5K input impedance (almost no loss) and still 26 dB (1:40) step up ratio, e.g. in the Mindprint DTC.
Regards
Kai
If my memory serves me right 26 dB gain means 1:20 ratio.
Best noise figures I have met were 126 dB for state of the art preamps with stepup transformer.
These figures are of less importance recording music where we can count on relative high output from mikes. The SPL helps quite a lot.
I am not in the business of recording ants pissing against cotton.
-
If my memory serves me right 26 dB gain means 1:20 ratio.
You're right, Mindprint DTC has a transformer specially developed and made for them by Haufe. If I recall right it has a ratio of 1:20 (a gain of 26 dB), while maintaining very high input impedance (5K) and extremly broad frequency response (5 Hz-127 kHz).
Those guys at Haufe are real specialists!
See:
http://mindprint.de/cms.php?scr=show&tab=prod_anhang&id=7&r=p
Next time I service one of my units I'll take the time to measure the real values.
Regards
Kai
-
A transformer does not have an input impedance, it simply converts the impedance it sees on one side to the other side according to the ratio of the transformer.
An input impedance of 5k is not the figure I would expect for optimimum noise figures.
It means the input impedance of the circuit following the transformer with step up ratio of 20 needs to be 1 Mohm.
I must admit I do not follow modern solid state developments but this does not look like an easy figure to achieve with minimum noise in mind.
Given the system of minimal load on sources as used in European professional audio 2 Kohm input impedance would be a more apropriate value.
-
A transformer does not have an input impedance, it simply converts the impedance it sees on one side to the other side according to the ratio of the transformer.
Yes and no.
A transformer is designed to work within a certain range of source and load impedances.
Outside of this range, specially at higher source or lower load impedance, frequency respose and sometimes even LF distortion figures suffer.
Otherwise you're right, the transformer even "transforms" the load and source impedance in the 2nd potence of it's ratio:
E.g. a 1:3 transformer transforms the connected impedance by a factor or 9 into the step up-, by 1/9 into step down direction.
An input impedance of 5k is not the figure I would expect for optimimum noise figures.
It means the input impedance of the circuit following the transformer with step up ratio of 20 needs to be 1 Mohm...
I works very well, I'll provide my own measurements ASAP.
As explained above the 1MOhm isn't the impedance the active electronic is shunted, but the much lower impedance that results from source impedance transformation.
The whole design was developed with by Mindprint in close collaboration with Haufe, who build this special transformer that is constructed much more complicated then usual.
The idea of this 5K input was to avoid damping of impedance peaks that can be found on dynamic mics, and I can say it works.
An Shure SM58 e.g. sounds brighter, more open on a DTC compared to any micamp I know.
Regards
Kai
-
Kai,
5000x20x20 equals 2 Mohm load at the secundary of the transformer.
(You state the input impedance is 5Kohm.)
It follows the input impedance of the electronics is 2 Mohm.
Distortion does not vary as a function of the source impedance for low frequencies.
It is most of all dependend on power and frequency.
Power is low as long as the load of the source is ten times higher than the source.
I designed quite a number of inputstages as a young engineer.
The basics for good designs have not changed, components have but they are still based on laws of nature.
Input transformers are not rocket science with all due respect for Haufe.
By far the best input transformers I ever got my hands on were made by Studer.
These were used in the much improved green consoles designed by Polygram in the seventies.
Green followed the black generation that suffered from serious problems.
These green mixers were installed at Wisseloord studios in 1976.
They were superceded by SSL.
-
Do you believe Studer made its own? That would be uncommon for what primarily was a tape machine manufacturer.
Anyway, would be interesting to find out who made the transformers. There are never enough good ones out here!
-
If a manufacturer is able to make recorders like the C37, A62, A80 mixing consoles like the 900 series they are certainly able to produce input transformers.
Yes Studer made their own audio transformers.
Polygram ordered large quantities for the mixing consoles that were designed and build in Baarn, the Netherlands.
Studer and Polygram are history, no more excellent analogue equipment from these companies.
btw Studer also manufactured their own recording and playback heads.
For a designer small transformers that do not transfer power are easy to design.
Sonic quality depends largely on the sort of metal used for the core.
Designing output transformers for tube amplifuers is a different piece of cake.
Minimum distortion at low frequencies and a large bandwith like 150 Khz are not easy to combine with power transfer.
Partridge in the UK and Unitran in the Netherlands made those.
Famous American tube amps like MacIntosh and Harman Kardon were based on superior output transformers.
-
Input transformers are not rocket science with all due respect for Haufe.
By far the best input transformers I ever got my hands on were made by Studer.
Not exactly rocket science, although I wouldn't be astonished if some made it into the Sojus rockets :)
Building a good sounding audio transformer is high science.
It's even harder if you want to combine the extremes in ratio, bandwidth and level/distortion.
It's no wonder that well made, maybe like your mentioned Studer's or the famous Neumann BV8, are so sought-after.
Regards
Kai
-
No idea what you mean with high science.
The guy who time and time again astonished companies with their own scientific labs designing transformers they could not develop let alone produce was what Germans call a Diplomingenieur.
He was not educated at university. I am referring to Mr. Y Drost at Unitran in Weesp, the Netherlands. He was my supervisor during the time I was there as part of my study.
He could predict with maximum 5% tolerance what the results of a new design would be meaning, banwith, distortion etc.
A pity this knowledge was lost when Unitran closed in the eighties.