DIYMobileAudio.com Car Stereo Forum banner

Does anyone Actually know?

7K views 45 replies 13 participants last post by  MercuryFlint 
#1 ·
I’m wondering does anyone actually know why when setting gains everyone says use 40 or 1000hz?
I can’t comprehend why I would want to set the gains of a amp running tweeters with 1000hz because the amp is ever even going to see 1000hz. It is only going to be amplifying whatever frequency range I send it. NOT THE ENTIRE SPECTRUM.
So how does that make any sense at all?
I can understand it if running passive system. But active it make zero sense.
Say i run my tweets from 3000-20k. It doesn’t matter when the amp clips at 1000hz.
Same thing for setting midrange and midbass. It makes more sense to use tones within the givin signal the amp will be seeing.

And another thing, who chose these couple tones 40 and 1000. Why would we set the gains to not clip at 1 frequency instead of playing pink noose or something playing the whole spectrum?
Yeah it may not clip at 1000hz but what about the other 19999 frequencies?
 
#2 ·
you will want to set the gains in the range that they will be playing. setting sub gains at 1khz is pointless as setting tweeter gains there. they wont play it.

personally, set the gains with your ears. if it is too bright turn down tweeters. etc
 
#4 ·
Some background is that amp power is universally specified using 1,000 Hz since it is unknown what speaker will be playing. Actually, I can’t find where it says amp power is taken at 1,000Hz but the Texas Instruments paper below mentions the FTC requires 1/8th power preconditioning at 1,000Hz so maybe that has something to do with the 1,000Hz “standard”. I also looked at CEA power guidelines but while it covers power and S/N, it doesn’t specify at what frequency power should be rated, at least not that I could find in a quick search.


However, if you want to set your gain via a test signal, pick a set and play a tone in that range. So for the 1,000Hz test tone, you’d be setting the baseline gain of your system to the mids. Then you wouldn’t use any other test tones, you’d level set the gain on the other channels (subs and tweets) based on the output level of the mids. So your mids at that point would drive the gain structure for the other channels.
 
#7 ·
I went looking. Here is what I found.

Audio frog

Points to Harvey Fletcher. (Dutch?) Na.. BRITISH!
263817
Harvey Fletcher’s study of acoustics led to many innovations, and he continued learning and teaching about his field throughout his life.


From BYU:
Sound Science - BYU Magazine




THIS DUDE WAS THE TESLA OF AUDIO! And he even had time for 7 kids!
263818





Article BYU
The Father of Stereophonic sound.


Something else you should hear. This man did this. Please read and listen. AMAZING

From this 1923 Edison cylinders:
<iframe width="100%" height="166" scrolling="no" frameborder="no" src=""></iframe>

To this 1932 FLETHERFIED!:


Fletcher and Munson curves

WILDEN MUNSON, BELL LABS
263821




263822


LINK:

ISO Standard both developed



Human response curve with loudspeakers


Specifically on Audio level measurements
Audio system measurements - Wikipedia

Output powerOutput power for amplifiers is ideally measured and quoted as maximum Root Mean Square (RMS) power output per channel, at a specified distortion level at a particular load, which, by convention and government regulation, is considered the most meaningful measure of power available on music signals, though real, non-clipping music has a high peak-to-average ratio, and usually averages well below the maximum possible.

The commonly given measurement of PMPO (peak music power out) is largely meaningless and often used in marketing literature; in the late 1960s there was much controversy over this point and the US Government (FTA) required that RMS figures be quoted for all high fidelity equipment.

Music power has been making a comeback in recent years. See also Audio power.Power specifications require the load impedance to be specified, and in some cases two figures will be given (for instance, the output power of a power amplifier for loudspeakers will be typically measured at 4 and 8 ohms). To deliver maximum power to the load, the impedance of the driver should be the complex conjugate of the impedance of the load. In the case of a purely resistive load, the resistance of the driver should be equal to the resistance of the load to achieve maximum output power. This is referred to as impedance matching.


RMS or Root mean square
Root mean square - Wikipedia

In frequency domain
The RMS can be computed in the frequency domain, using Parseval's theorem.




ITUR-56

The origin of the current ITU-R 468 weighting curve can be traced to 1956. The 1968 BBC EL-17 report discusses several weighting curves, including one identified as D.P.B. which was chosen as superior to the alternatives: A.S.A, C.C.I.F and O.I.R.T. The report's graph of the DPB curve is identical to that of the ITU-R 486 curve, except that the latter extends to slightly lower and higher frequencies. The BBC report states that this curve was given in a "contribution by the D.B.P. (The Telephone Administration of the Federal German Republic) in the Red Book Vol. 1 1957 covering the first plenary assembly of the CCITT (Geneva 1956)". D.B.P. is Deutsche Bundespost, the German post office which provides telephone service in Germany as the GPO does in the UK. The BBC report states "this characteristic is based on subjective tests described by Belger." and cites a 1953 paper by E. Belger.



i-TUR56


DIYMA GAINS



RF Tech

still no real answer....
 

Attachments

#8 ·
So After doing even MORE looking around. Sleeping a bit. THE SHORT answer is that at 1-Khz, was agreed upon
was...

CUZ MAGAZINES MADE IT POPULAR.

Here is Gene From Audioholics take on it.

263844
Covid-19 Edition.
263845


(1kHz PSweep) - popularized by the print magazines, this is an instantaneous power vs distortion test at 1kHz. This is the test most manufacturers (and many print publications) do when rating their amplifiers " Continuous All Channels Driven". The problem with this test is it often masks slew related and or frequency response problems some amplifiers exhibit at the frequency extremes, and thus inflates the measured power results. It does provide an instant gratification number for consumers to argue over on the forums so we are now incorporating this test to please the masses. It also gives the consumers an apples to apples comparison between our ACD measurements and the measurements that the major print magazines do.







Now WHY would a Magazine do that? Would it just be Monkey see Monkey do? Or was it something that was repeatable? Something easy? The truth is, that it was the most Simple error free way to do it.

You have to think back to back in the day when getting a reference grade signal generator cost more then some homes. Getting one that was based on the Audio band was not that much cheaper. Sure you could Build one. But then you have all these other things to think about. Such has line level out, Impedance mismatch etc.

From INFOCOM:

There are two methods for setting system gain: Unity and Optimized. Both methods are relatively simple and neither requires expensive equipment. For a typical presentation/conference room/boardroom type environment, Unity Gain will provide an adequate system signal-to noise ratio – around 60 dB using professional audio components. If the first device shows a signal level output of 1.23 V, you should be able to measure a 1.23 V signal all the way to the power amplifier inputs. For a more critical listening environment such as a small studio, broadcast facility, performing arts center, lecture hall, etc., the System Optimization Method will provide the optimal signal-to-noise ratio allowed by the equipment in the signal path. This method takes more time and skill to complete than the unity gain method.

So if you think about it, testing something like an amplifier is easy if you are just doing one thing. POWER OUTPUT. But Amplifiers are not Linear output devices that perform one function so to speak. Unlike a Power Invertor that we want a set mode of parameters, (50-60hz, (cycles per second at the Zero Cross point of an Alternating current output?) @ 120-240 VAC at 15-20-30 or 50 Amps Rated Load; Audio amplifiers have to do the min of 20hz to 20 Khz at any level demanded at any power level from a Micro volt to 1000 watts; And do it with the least amount of distortion of the signal possible.

Now if you remember back in the day, how much gear could really do 20hz-20Khz FLAT White noise? Sure you could make electroinics that could do that. But did you have Playback gear that could? The Answer to that is No.

Heck your Tape decks can BARELY do this out of the box. Let alone FLAT. Let alone without adding anything else to signal (Distortion or extra noise) Let alone your Phono machines of the day.

But? You could build a Signal Generator that put out one tone at the time. And do it reasonably well for a modest cost that could be used as a reference. Then you run into another problem. Audio Unity.

Andrei Martinez Agras:


Unity Gain, also called Unity Volume or volume at unity, is when the bypassed output level of a device that can be a guitar pedal, a studio hardware unit or a plugin is the same as when the device is engaged. Typically this can be achieved by using the output level control of the device or the auto gain function if its included. The importance of using unity gain levels when applying a process or effect is that a positive difference in volume can be perceived by our ears as an increase in fidelity and sound quality tricking us to believe the effect makes our tracks sound better even when the only change is volume. In this example we used a Maxon SD9 distortion pedal and Audified U73b v.2 compressor plugin on guitars and drum loop respectively.

This link goes into Audio Unity and the white paper by AES. Thanks to Amrim for sharing this.
263850


263851
Amirm

Staff Member CFO (Chief Fun Officer)
Founder/Admin · From Seattle Area


Basically, It ain't easy getting a solid reference grade signal from one device to another. And make it repeatable.
Things such as Gain from one Makers device to another can be different from brands etc. This is important as when you are trying to get the S/n (Signal to noise ratio, and Noise floor, & Dynamic range, this all plays into how you measure power.

And of course, there is another thing. Bass Frequencies from an electronic perspective require more power reproduce then High Frequencies. (Generally speaking) AKA why Avionics use 400hz power vs 60hz. like you have at your wall outlet.

Here is a good Visual on that from the Big-Dummy Lord Supreme Mystro of OldSkool !


As you can see from these tests, the amp gives you different power levels of output on the dyno resistive load.
in practice? Unless you run magnepans or Ribbons? You will never see that power. And there is the whole Amp stability thing as well. (second modern reference here) And if you want to get your AES Inviite one day? Look here.



With all these variables, you start to see the reason and logic of the 1 KHz Standard used for gain and power testing. It just makes things simple & repeatable. Not to mention it is a heck of a lot simpler for any human to hear it, and see it on an O-scope.

Remember, computers were not really common place in the 90's. Even worse still was that the know how on how to get your computer to take in a signal and then write the software on looking for a Distortion or un-lineratity in just the 20-20Khz bandwidth for just one second Stereo would be about 288 kB (kilobytes) for a 24 bit 48 khz sample rate. But say we just want to look at what they had back in the day. PCM @ 16 Bits @44.1 Khz---That would be 176.4 kB (kilobytes) a second sample with a data Bit rate of Bit Rate: 1411.2 kbps (CD RedBook audio)

So, with that in mind, we know have a screaming fast 286 CPU. Well. Hold on here. It still can't PLAY BACK PCM by itself. Not in real time with everything else it has to do.

So you are in the Pentium Area. That was the late 90's. If you got the hardware to do it, then you could. But then again, just doing the Data calculations would have taken the machine at least a gigabyte of data. More then the CD at the time to do a Full analysis. Let alone the RAM needed. And Custom hardware.

It would have taken you for just that ONE second of audio? about a day or two if you wrote that program in C. More like a Week if we are being honest.

If you did do this kind of work? You had one of these.
263859


Or an SGI workstation.
263860



And a few of these:
263861


But most likely, you had this to work with:
263863



And just Writing the program to do this would have cost you upwards of maybe a Million dollars. Lets not forget the hardware you need to make it work all custom built.

Or you Went to HP and got a turn Key solution. If you were lucky, You could get a GREAT deal on a Vacuum Tube Frequency counter.



And you need all this too.
263866




For a Magazine? Or for anyone else? Using the 1Khz method sure is a LOT simpler and a O scope. :)


The other thing to take into account is that what you get from the LAB is not going to be what the End user gets.
This is more of a Problem with Car Audio. As you can tell, we have all sorts of Standards that are used for gain. Its my opinion that its the reason why the Optical Output or Digital output is the perfered method and why it sounds so good as you get bar none the FULL output of any gear. Its all to a standard already (SPDIF/Toslink)

So from your source, (Car Radio) to your Amplifiers, if you can take it all the way to the last item in the chain, then you get that full dynamic rage possible. Hence, at that point, you don't even need to bother with gain settings. Its all done for you. (See Rockford Fosgate's/ Stillwater Designs Link Above and other links to point out gain vs Dynamic range and Loudness levels)


When you do all the Math, and calculations of all the variables? You get the magic 1 Khz to set gains that works across all gear. And Anyone with Basic tools can do this and get the same results. (Mostly)

Some Reference Points.


















WAY more then you need to know. But Hey, I had time on my hands. I hope you find all this Info Useful.
 
#13 ·
I have a dd1. When I am using it, say I’m using 1000hz tone. The 1000hz led comes on, then as I increase the gain on the amp or output power from the helix there is a point where he 1000hz light goes out. About two clicks on my B2 audio amps. Then the distortion led will come on and sometimes with no 1000hz led lit. Sometimes the 1000hz led and distortion led will light.
Is my dd1 faulty?
 
#14 ·
I have to admit. That is a weird one. Steve can run a test on the meter. But man, I would love to see a video of that.

Without more into I can't give you a good answer. But here is my take:


1. Battery voltage sagging?
-----This may cause the amp to go into distortion somewhere in the chain.

2. EMI/RF
------This may indice noise in the signal. It would mean maybe noisy cables, or a power supply that is not filtered.

3. Mechanical issue.
------Such as a gain knob being corroded or out of spec. As you turn it, it could be non linear as most knobs are just simple resistive devices.

4. Bad signal connection
----- with the DD-1, the plug may be off? Or not getting contact? MAYBE? Or maybe the battery is low. Try a fresh new Name Brand Alkaline cell.
 
#17 ·
Yeah that sounds interesting.
where canI find a silent track to set the tweeters gains with?

to the other- I tried a brandnew battery in the dd1. It has done this since the first day I had it. With a completely different system.

he first system I had skar amps and JL Twk88 same kenwood 9905s. The dd1did what I described.

second system is two B2 audio amps, helix dsp.3 Kenwood9905. You mentioned the gain knobs coild cause it. But I can replicate this by raising the helix dsp.3 signal up or down also.

I can not replicate it by maxing my kenwood unit. 39 out of 40 either the 40hz or 1000hz light is on and and on volume 40 the clipping light comes on.
I’ve tried every possible combo of gain structure. Low helix out high amp gains or the opposite. No matter what the dd1 will usually do that where the frequency light goes out for a few clicks then only distortion comes on.

I forgot to add also when I got my second system I added a 370 amp alternator and rear battery. My voltage never drops under 14v.
I also replaced all of the rca cables for the second system with stinger 8000 series.
I honestly think it’s the B2 amps. I have the 2500.1 and the 1200.6 and had to have the 2500.1 warranteed for a bad rca input and the 1200.6 has way too much hiss for my liking and it makes a crazy noise from channel 6 when it’s hot. (There is Another thread where I posted this sound)

as far as I’m concerned that B2 1200.6 is total garbage. I can’t wait to take it out and blow it up or shoot it full of holes.
 
#18 ·
I get what you are saying with the DD-1. Be sure you are setting max clean output following the 'gain chain'. Use it for head - LOC - DSP - amp, in that order. It could very well be the cheap amps but what's the chances of them both being bad ? My DD-1 does wierd blinking similar to what you describe, at times. I think the distortion light blinking before full on is normal. I usually go by when it just barely blinks dimly, if that makes sense. I use mostly Alpine amps.
 
#26 ·
So after reading that it says if you have the input level control set to 0db or 100% in some you have no room for any boost. The Helix go to +10. So if I keep it at 0 do I have room for boost?
 
#28 ·
And what is this crossover filter q? How do I know if it’s higher than the .707? What makes it higher or lower? Or is this a setting in the helix I’ve always overlooked?
 
#29 ·
Makes me wonder bavause I did some slight adjustments the last couple days and today I got nasty distortion and I’ve never had it before. Now I gotta go figure out wtf happened Maybe I changed the Q and don’t know it? Or maybe setting the helix at 0 in the input side and having a couple eq boosts caused it. If that’s the case then why does helix go to +10
 
#34 ·
So on the input in the I/O tab of the helix is 0 considered attenuated since it goes -10 to +10?
If not and it’s bad to raise it then why is there even an option to raise it 10db?
 
#37 ·
I believe you are looking at the dB level for each input source on the IO tab. This is usually used to volume match different sources to avoid the situation of one source playing way louder than another source at the same volume level (as set by your volume knob or HU).

From Andy’s advice he is saying to turn this down. He doesn’t say by how much but it should be enough so that the levels/volume is low enough so that you clearly are not getting any audible distortion. He recommends also turning down the output gains (which are in the main tab and are channel specific. You want to be careful to link all the channels when adjusting this downward as if you’ve level matched your speakers already this will be messed up if the channels aren’t linked).

He then recommends doing your tuning in the DSP after both your input level and output levels have been lowered (I would think it would make sense to lower these at least as many dBs as the max boost you plan on using in your tune).

Then, once your tune is set, gradually increase the input and output levels proportionally (again making sure the channel levels are linked) until you notice the first bit of distortion and then back it off a bit.

The levels from -10 to 10 on the inputs just provide flexibility to volume match your sources. The exact position of 3 or -4 isn’t that important as long as you are happy with how they are volume matched and if in the above process you have enough room to lower the input to account for your maximum boost that you did in EQ and end up with no distortion.


Sent from my iPhone using Tapatalk
 
#38 ·
I try to not boost at all but sometimes it’s needed. Not sure what the max recommended boost is that people go by. I keep the helix usually 0 to -2.5db in the I/O page If I put it into the +dB it causes some hiss. Trying to find a good balance is hard between the dsp and the B2 1200.6 amp because it most definitely causes hiss! I hate that amp and need to research a good 6 channel amp that doesn’t have much hiss. The B2 has an snr of >90db and I have questioned B2 directly how they came to this spec and if they did it cea 2006 compliant or how and the responsed by avoiding the answer. Saying they did it properly..... blah blah blah. They did. It like being called out about their specs and the amp hiss issue.
 
#39 ·
On the major streaming services you can search for silent tracks. Most that I’ve used have them.
but if it’s streaming service who is to say it’s silent. The service has a noise floor (hiss) of their own
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top