Topic: How to test a preamp? Advice needed!

[sunjan]

Hi all,

I received a Serv4U preamp for testing.
I'd like to make some A-B recording samples, and post some nice plots with spectrograms, noise floor, like the professionals here do.

How do I design the test to eliminate all external factors and make it "fair?

The gear I have at my disposal is:

Mics:
Nak CM-100 (stock, battery powered)
AT853Rx (hardwired phantom adapters)

Preamps:
Serv4U XLR 248HD (phantom powering, analog out)
UA-5 wmod
CA-9100 (miniXLR input jacks)

Recorders:
H120
H320
JB3

Do I simply connect the gear and record a ticking clock, while switching one preamp for the other, while leaving the mics at the exact same place?
How do I set the gain? Just set the recorder to zero gain, and calibrate the gain on the preamp to peak at 0dB (using the safety clip feature in Rockbox)?
And which software makes good/usable spectrograms? I have Wavelab, I suppose that's OK. What settings should I use in the software? Should I normalize/prepare the WAVs in any way, or just view them as they are?

Some concerns regarding the "comparability":
If I compare the Serv4U with the UA-5, it's not entirely fair since the UA-5 has digi-out. Going analog out on the UA-5 is not faultless either, since it doesn't have analog pass-thru.
And comparing the Serv4U with CA9100 isn't straight either, since Serv4U has phantom, while CA9100 provides 9V plugin power. Maybe that's a non-issue? I guess I could run the Naks into both pres, switching off P48. But I don't have any XLR>miniXLR cables ATM... 

Any feedback greatly appreciated!

[illconditioned]

Hi all,

I received a Serv4U preamp for testing.
I'd like to make some A-B recording samples, and post some nice plots with spectrograms, noise floor, like the professionals here do.

How do I design the test to eliminate all external factors and make it "fair?

The gear I have at my disposal is:

Mics:
Nak CM-100 (stock, battery powered)
AT853Rx (hardwired phantom adapters)

Preamps:
Serv4U XLR 248HD (phantom powering, analog out)
UA-5 wmod
CA-9100 (miniXLR input jacks)

Recorders:
H120
H320
JB3

Do I simply connect the gear and record a ticking clock, while switching one preamp for the other, while leaving the mics at the exact same place?
How do I set the gain? Just set the recorder to zero gain, and calibrate the gain on the preamp to peak at 0dB (using the safety clip feature in Rockbox)?
And which software makes good/usable spectrograms? I have Wavelab, I suppose that's OK. What settings should I use in the software? Should I normalize/prepare the WAVs in any way, or just view them as they are?

Some concerns regarding the "comparability":
If I compare the Serv4U with the UA-5, it's not entirely fair since the UA-5 has digi-out. Going analog out on the UA-5 is not faultless either, since it doesn't have analog pass-thru.
And comparing the Serv4U with CA9100 isn't straight either, since Serv4U has phantom, while CA9100 provides 9V plugin power. Maybe that's a non-issue? I guess I could run the Naks into both pres, switching off P48. But I don't have any XLR>miniXLR cables ATM... 

Any feedback greatly appreciated!

If you're able to take it apart, I'd like to see the guts.

Thanks for testing...
  Richard

[sunjan]

Phew! I think I need to attend a uni course in audio engineering to understand all this. Most of this went way above my head! 
Thanks a million for your thorough advice, mshilarious.
Be prepared for a lot of stupid questions along the way...

I've got this multimeter, but I have no idea if it's RMS capable:


Regarding the dummy load, can I use my multimeter to apply resistance between the pins??? I don't have any resistors lying around 

Or for feeding the preamp with a test signal, I can use a 3,5mm stereo>dual XLR cable and connect it from the computer line out to the pre inputs?

Using Wavelab as a signal source (Tools/Audio Signal Generator), create a 1kHz sine wave.  Route that out of your interface and calibrate to 1VRMS. 

How do I do this calibration? In Wavelab? Heck, I'm not even sure I know what 1VRMS is... 

OK, so try setting the level at -30dBV, and send that to each unit.  Note that for the balanced input units, you need to have pin 3 terminated to get a valid measurement. 

Another stupid question: how do I terminate pin 3? On the pre? At this stage, isn't the XLR connected to the cable? Do I need to open the pre and measure from within the box while it is connected?

It's best if you can route a differential test signal to the unit, if your interface has balanced outs. 

How do I make the signal differential (and what does it mean in layman terms)?

Note that you need to measure output between pins 2 and 3 (or tip and ring) to calibrate to 1VRMS, but for the unbalanced input pres you'll have measure between pins 2 and 1 (tip and sleeve).

This is the output of the pres, right? Not the output from the computer??? Again, the calibration parts is a bit hazy for me.

Set each pre to minimum gain and measure the output voltage from the pre. 

Between which pins do I measure the output voltage? Since the output is joint 3,5mm stereo, does this measure one channel at a time, or both somehow?

Compare with the 1VRMS level (you can use a loop from the output to the input of your converter to calibrate that). 

The converter - is this my computer basically, or a separate piece of gear I should get?

So during all these tests, it's just the PC + pre + multimeter? No mics or recorder involved yet?

Once I've digested this, I'll probably have more questions on the second part of the test...

Again, thanks for your patience mshilarious. And if anyone else has explanantions or opinions, chime in!

[digifish_music]

Hi all,

I received a Serv4U preamp for testing.
I'd like to make some A-B recording samples, and post some nice plots with spectrograms, noise floor, like the professionals here do.

How do I design the test to eliminate all external factors and make it "fair?

The gear I have at my disposal is:

Mics:
Nak CM-100 (stock, battery powered)
AT853Rx (hardwired phantom adapters)

Preamps:
Serv4U XLR 248HD (phantom powering, analog out)
UA-5 wmod
CA-9100 (miniXLR input jacks)

Recorders:
H120
H320
JB3

Do I simply connect the gear and record a ticking clock, while switching one preamp for the other, while leaving the mics at the exact same place?

The ticking clock certainly has an important place for subjective assessment of preamp noise.

You probably remember this thread (aimed at blind comparisons, but the principles for making the recordings are sound IMO)

http://taperssection.com/index.php/topic,113862.0.html

Summary...

A couple of basic things (all intended to improve listener sensitivity to differences)...

1. Levels need to be matched. Louder almost always sound better. The threshold for loudness differences is in the order of 0.5 dB (under the best listening conditions in an A/B back-to-back change. It rises to about 1.5 dB at 1 minute separation between the examples, again under ideal listening conditions). In the real world 1 dB is fairly close to the just noticeable limit.

2. The samples need to be spliced back-to-back with an instant (click free) transition between the recordings. The duration of the selections should ideally be around 5 seconds (however this is open to debate and will depend on the qualities being assessed. Generally, short selections A/B no gap usually produce the most sensitive discriminations).

3. The listener should be allowed repeated presentations of each comparison paring until they make their decision.

4. There should be an objective response indicator. That is they should be asked to choose based on preference - either 1st or 2nd half, or pick the modded one or whatever. But they should be given a A/B 'forced choice'.

5. The number of comparisons should exclude chance from playing a significant role in the outcome. It's typical to shoot for 95% confidence in the perceptual world. So that means a performance of ~7 consistent identifications out of 8. Ideally 10 randomized A/B or B/A comparisons would make a good standard.

5. The comparison should be done blind. That is the listener should not know which example they are listening to on any given session.

6. No feedback during the assessment session should be given. The internet makes this easy as you can download a file and listen to it without seeing the experimenter or interacting with them.

7 The recordings should be simultaneous from the same set of mics. Alternatively, the same the same microphones used in the same position recording the exact same sound. That is why the recording of a Hi-Fi source, ticking clocks, test signals or whatever is a great paradigm.

[Church-Audio]

Phew! I think I need to attend a uni course in audio engineering to understand all this. Most of this went way above my head! 
Thanks a million for your thorough advice, mshilarious.
Be prepared for a lot of stupid questions along the way...

I've got this multimeter, but I have no idea if it's RMS capable:


Regarding the dummy load, can I use my multimeter to apply resistance between the pins??? I don't have any resistors lying around 

Or for feeding the preamp with a test signal, I can use a 3,5mm stereo>dual XLR cable and connect it from the computer line out to the pre inputs?

Using Wavelab as a signal source (Tools/Audio Signal Generator), create a 1kHz sine wave.  Route that out of your interface and calibrate to 1VRMS. 

How do I do this calibration? In Wavelab? Heck, I'm not even sure I know what 1VRMS is... 

OK, so try setting the level at -30dBV, and send that to each unit.  Note that for the balanced input units, you need to have pin 3 terminated to get a valid measurement. 

Another stupid question: how do I terminate pin 3? On the pre? At this stage, isn't the XLR connected to the cable? Do I need to open the pre and measure from within the box while it is connected?

It's best if you can route a differential test signal to the unit, if your interface has balanced outs. 

How do I make the signal differential (and what does it mean in layman terms)?

Note that you need to measure output between pins 2 and 3 (or tip and ring) to calibrate to 1VRMS, but for the unbalanced input pres you'll have measure between pins 2 and 1 (tip and sleeve).

This is the output of the pres, right? Not the output from the computer??? Again, the calibration parts is a bit hazy for me.

Set each pre to minimum gain and measure the output voltage from the pre. 

Between which pins do I measure the output voltage? Since the output is joint 3,5mm stereo, does this measure one channel at a time, or both somehow?

Compare with the 1VRMS level (you can use a loop from the output to the input of your converter to calibrate that). 

The converter - is this my computer basically, or a separate piece of gear I should get?

So during all these tests, it's just the PC + pre + multimeter? No mics or recorder involved yet?

Once I've digested this, I'll probably have more questions on the second part of the test...

Again, thanks for your patience mshilarious. And if anyone else has explanantions or opinions, chime in!

In other words its not something for the do it your self-er... It has taken me along time to get all the test gear I have. I would be more then happy to run some noise tests on the preamp as well as frequency response tests free of charge just to be able to look inside the thing. As far as a listening test goes its hard because the cheap preamp has fixed gain but how much gain does it have? nobody seems to know.. you would to be objective at the very least have to match the gain levels... that does not seem possible with this preamp. I suspect it will be noisy and a big piece of turd. While it certainly is possible to build a preamp this cheaply its not really possible to do it and actually have it be a worthwhile product from a quality standpoint. The design is weak because of the use of a cheap molded jack who knows what else lurks inside!

Taken from his ad "High quality silver plated 1/8 inch (3.5 mm and 2.5 -- > PLEASE ASK!) stereo mini plug "

I say silver plated my ass.... I had some of these jacks I made the mistake of buying 100 3.5 to 3.5 cables.. They were junk and made such a poor contact with anything I tried to use them with I had to throw them out..


And I sure as hell hope that frequency response graph is of a set of microphones and not his preamp because if it is he has some serious design issues with it I would say by looking at the graph its +- 12db! from 20hz to 20khz not good for a preamp not great for a set of mics Actually not really good for anything.

[sunjan]

Taken from his ad "High quality silver plated 1/8 inch (3.5 mm and 2.5 -- > PLEASE ASK!) stereo mini plug "

I say silver plated my ass.... I had some of these jacks I made the mistake of buying 100 3.5 to 3.5 cables.. They were junk and made such a poor contact with anything I tried to use them with I had to throw them out..

Hey Chris, you're confusing two different preamps!
I don't have the el cheapo Spyshop $40 pre, but the Serv4u $165 one with dual XLR and phantom. So your comments above are not valid for what I want to test, strictly speaking.

To get back on topic, could you suggest any kind of testing procedure that is easy enough to perform for an amateur???
Like I said, I don't know if my multimeter has RMS measuring capability, so keep it basic if possible...

[Church-Audio]

Taken from his ad "High quality silver plated 1/8 inch (3.5 mm and 2.5 -- > PLEASE ASK!) stereo mini plug "

I say silver plated my ass.... I had some of these jacks I made the mistake of buying 100 3.5 to 3.5 cables.. They were junk and made such a poor contact with anything I tried to use them with I had to throw them out..

Hey Chris, you're confusing two different preamps!
I don't have the el cheapo Spyshop $40 pre, but the Serv4u $165 one with dual XLR and phantom. So your comments above are not valid for what I want to test, strictly speaking.

To get back on topic, could you suggest any kind of testing procedure that is easy enough to perform for an amateur???
Like I said, I don't know if my multimeter has RMS measuring capability, so keep it basic if possible...

LOL sorry.....  I would simply use my computer to send the signal and then return the signal from the preamp back into my computer and analyze it and adjust the preamp gain levels in real time but that is because I have a balanced interface on the input and output of my measurement card. Then I use my Lab version of Winaudiomls to measure the preamp. Set gain levels and move on to an actual test with mics... making sure that the gain levels were untouched so as to give a fair apples to apples comparison of the two preamps. But I forgot to mention the fact that there is about $2,800 in software and hardware being used to do the tests. This is not really something that can be done with the tools you have at hand.

You must find a way to set the gains of both preamps at the same levels or any comparison is moot.. You need a computer with balanced inputs and outputs and some software or a 1k test tone and a RMS ac volt meter and some termination resistors you can skip the resistors if you open up the connectors and test from the pins when they are plugged into a recorders input. This loads the preamp down to the same exact level as both preamps under test. This is very important because a change in the load = a change in the overall output level of the preamp. Sure you could normalize the tracks in post so they are the same but there might be performance advantages due to the fact that one preamp is not set at exactly the same gain as the other thus making your comparison unfair.

The other issue of concern would be how much current the phantom supply can render the spec for phantom supplies is 10mA per input if one or both preamps fail a simple load test with a amp meter inserted between a load resistor then that preamp would not really be of value for mics like earthworks or other mics that require almost all of the 10mA on the 48 volt supply.

Noise tests are easy to do if both preamps are set to the same level you must do a baseline measurement of your recorder terminated with a load resistor of equal value to the load the preamp will put on it and simply record the silence of your recorder set to the levels you will be using to record. Then take the wav file and upload it to your computer and put it into a program that can measure noise floor from a file * not many programs do this * then once you have your baseline for tests you can then move on to terminating the preamps inputs with load resistors of equal value that represents the mics that you will be using for the test. Then do two recordings one for each preamp with just the loads connected and both at the same gain levels.. This will give you an idea of gain vs noise. But again this test should be performed at various gain settings this will require multiple calibrations to make happen. Not something easily done at home. You are better off to hand this test over to me or Jon or someone that has the gear then to try and do it your self.

Chris