Amplifier Setting Tutorial

[luvinthebass 5,000+ posts]

When setting the HPF on your speakers, how audible is audible distortion? Do you think it would hurt anything if my Frequency is set at about ~80-90 hz? Will it destory my speakers over time or anything?
But I finally went out and bought a DMM and wow it improved my sound so much, ty!

Pioneer AVIC-N3 (2v outputs

Phoenix gold R8.0:4

Alpine Type-R comps (60 RMS)

Set it at about 7.5 / 7.6 V, at 48 / 62 volume, and it really made a difference. I set my Frequency on the amp to about 85 ish hz, and I don't "think" I hear any distortion... Should that be ok?

you're fine, just don't go over 48 volume number.

 

[chartman18 10+ year member]

Heres my problem.

____________________________________________________________

Setting the subsonic filter (if available):

1) Determine the tuning frequency of your enclosure (if using a sealed enclosure, turn SSF off or to minimum frequency)

2) Calculate the desired SSF frequency by using the formula: SSF = Tuning*3/4

3) If number is not whole, round to the nearest integer

4) Write the SSF frequency down

5) Calculate the desired SSF voltage by using the formula: SSF_Voltage = 0.707*Sub_Voltage

6) Write the SSF voltage down

7) Turn the SSF to the maximum frequency

8) Set track to the SSF frequency and set to repeat track

9) Set to volume written down in speaker setting stage

10) Turn down the frequency on the SSF control until you reach the SSF Voltage from step 6

11) Turn the volume down to 0

_____________________________________________________________

Ok, I have a 4-ohm subwoofer, wiring into 2-ohm, how do I find my SSF Frequency, when it says "Tuning" *3/4 = SSF Frequency. Whats the "tuning"?

Then after that, I multiply my sub voltage by .707? So its .707 * 46.9 = 33.16

(for a Hifonics BXi 1606D at 2 ohms)

What do I need to do to set my subsonic? I tried to understand, and I didnt want to start a new thread, so I posted/asked on here.

Someone help me on the subsonic?

Thanks.

 

[Frankensuby 10+ year member]

are those value tables by Jmac and thch RMS wattages? If so, this is awesome info!

How do you figure out those voltages at 2.2v preouts? Just multiply them?

 

[energumeno]

Hi,

Well, this is only my humble newbie opinion but, from what I read, it is suggested to actually disconnect the speakers before taking the voltage measurement in the output of the amp.

Shouldn't the fact that the amp is not loaded with any speaker going to affect the voltage reading?

I mean, if you disconnect the speakers shouldn't you at least connect a 4 or 2 ohm resistor (depending on the type of speakers you are using) at the speaker output of the amp in order for it to actually put some current flowing out thru the resistor this way loading the amp and enabling you to measure the real voltage that will be feed to the speaker in a real setting?

 

[XTRProBoy 10+ year member]

^I said something similar in the other thread (the sticky on how to set your gain).

I don't get the calculation of the "target voltage" at all, particularly since you're measuring it with no load. Doesn't sense any make.

If my amp puts out a rated 450W RMS at 4-ohms and 1200W RMS at 1-ohm, that would mean if I were running it at 4-ohms I'd look for a target no-load voltage of 42.42V AC. If I were running it at 1-ohm I'd look for a target no-load voltage of 34.64V AC???

This doesn't sense any make at all if you ask me. Input stage gain setting is supposed to match the amplifier for the input signal and nothing else. How does this have anything to do with the load impedance you're driving (particularly when you're supposedly measuring the voltage on the speaker terminals of the amp with no load attached)?

You're telling me that if I drive my amp into 4-ohms instead of 1-ohm it magically becomes less sensitive on it's input stage requiring me to set the gain "higher"??? Again, how is this justified/explained? The RCA level from the HU did not change (same test tone, same HU volume level), the amp did not change (it's the same amp, the same gain knob, no load attached, etc.), so why if you plan on running the amp into a higher impedance (in this case) would you set the gain higher? That has nothing to do with matching the RCA level in that case, as you're arbitrarily setting the input stage sensitivity so the output terminals of the amp read some strangely calculated figure with no load attached.

Maybe I'm just not understanding something here, but just from what I can understand the whole thing seems like nonsense to me. :shrug:

Another note: The only way that target voltage would stay the same for all impedances of an amp is if the amp were completely unregulated on the output side meaning it would always double power into half impedance/halve power into double impedance. Some amps are like this, however many amps are not. (E.g. Calculating this target voltage would have netted the same value at any impedance on my old Orion XTRPRO 1000, but the same cannot be said in the case of my HCCA D2400.)

Still setting my amp like this either way (either 34.64V or 42.42V) ended up in the sub sounding ridiculously underpowered with music content. Perhaps that's the point of using the lower test tones/setting gain "overlap" but that still doesn't account for the strange voltage calculations which don't seem to make sense at all given that you're trying to set the amp for the RCA level.

 

[Blu84Rex 10+ year member]

Just a question out of curiosity. What would the difference be if you didn't unhook the speakers vs unhooking them? Would having the load make a difference considering that there would be a load when using. With that in mind wouldn't taking the reading with the load there be more accurate. Just a thought in my head.

 

[helotaxi 5,000+ posts]

I don't get the calculation of the "target voltage" at all, particularly since you're measuring it with no load. Doesn't sense any make.

OK Yoda, to you explain it I will.

If my amp puts out a rated 450W RMS at 4-ohms and 1200W RMS at 1-ohm, that would mean if I were running it at 4-ohms I'd look for a target no-load voltage of 42.42V AC. If I were running it at 1-ohm I'd look for a target no-load voltage of 34.64V AC???
This doesn't sense any make at all if you ask me. Input stage gain setting is supposed to match the amplifier for the input signal and nothing else. How does this have anything to do with the load impedance you're driving (particularly when you're supposedly measuring the voltage on the speaker terminals of the amp with no load attached)?

You're telling me that if I drive my amp into 4-ohms instead of 1-ohm it magically becomes less sensitive on it's input stage requiring me to set the gain "higher"??? Again, how is this justified/explained? The RCA level from the HU did not change (same test tone, same HU volume level), the amp did not change (it's the same amp, the same gain knob, no load attached, etc.), so why if you plan on running the amp into a higher impedance (in this case) would you set the gain higher? That has nothing to do with matching the RCA level in that case, as you're arbitrarily setting the input stage sensitivity so the output terminals of the amp read some strangely calculated figure with no load attached.

Your amp is not in a perfect lossless world. It doesn't double its power output each time you half the impedance. It's not that the amp is less sensitive at the lower impedance, its just that the rail voltage droops underthe high cerrnt load and it take less imput signal to clip the output because of the lower rail.

Maybe I'm just not understanding something here, but just from what I can understand the whole thing seems like nonsense to me. :shrug:
Another note: The only way that target voltage would stay the same for all impedances of an amp is if the amp were completely unregulated on the output side meaning it would always double power into half impedance/halve power into double impedance. Some amps are like this, however many amps are not. (E.g. Calculating this target voltage would have netted the same value at any impedance on my old Orion XTRPRO 1000, but the same cannot be said in the case of my HCCA D2400.)

This actaully answers your question from above and is exactly the issue. This whole tutorial is the basic way to do it if you don't have access to an oscilloscope and you have an amp whose rated output you can trust as legit.

Still setting my amp like this either way (either 34.64V or 42.42V) ended up in the sub sounding ridiculously underpowered with music content. Perhaps that's the point of using the lower test tones/setting gain "overlap" but that still doesn't account for the strange voltage calculations which don't seem to make sense at all given that you're trying to set the amp for the RCA level.

Setting the input gain to give you the correct output voltage is exactly what you are trying to do by setting the gain in the first place. If the output doesn't double each time you half the impedance, then you have to set it with a lower voltage in mind to account for power rail droop to avoid clipping. BTW, even on the older Orion and PPI amps that advertised 2x the powr into 1/2 the load, they didn't double their power into the lower load. If you were to set them into the higher load for max unclipped power instead of merely rated power you would find that they were comfortably underrated into the higher impedance in order to make their rating into the lower one. With the new amps there is no motivation to underrate the 4 ohm output os they rate it at what it will do and then apply reality to the lower imp. loads.

Most people that set their gains with the 0dB tones are quite disappointed because music just doesn't get to that output level. You are shorting yourself out of 50-75% of the potential of the amp. Use a -3 or -6dB tone and you will quickly find a lot more output.

 

[XTRProBoy 10+ year member]

OK Yoda, to you explain it I will.

LOL. Thanks... Actually I was waiting for someone to reply to this but no one did for a while.

The thing is you did explain some thing all of which makes perfect sense, I agree. //content.invisioncic.com/y282845/emoticons/smile.gif.1ebc41e1811405b213edfc4622c41e27.gif

However you didn't explain one vital flaw in the formula here.

And that is this:

The principle of setting input stage gain is to match the input level of the amp with the RCA output level from the source unit. We all agree with that, right?

Soooo given the same head unit and the same amp that input stage gain setting should ALWAYS BE THE SAME AND SHOULD NEVER CHANGE, no? If not then gain is simply a volume control! To me all it takes is matching that input sensitivity to the output level of the source unit--that's what we're trying to do, not use it as a volume control to take it easier on the amp in certain situations (driving different loads).

So let's do that. Let's match the input sensitivity to the output of the head unit and be done with it. But this guide does NOT tell you how to do that.

Why doesn't it? Because if you're using the same amp but driving a different impedance this guide tells you to set a completely different input sensitivity than if you were driving some other impedance. But regardless of what impedance the amp is driving, the input sensitivity should be matched to the source unit. Shouldn't it? Turning a knob to match input stage gain to the output of the source unit has nothing to do with what load you drive from what I thought. Regardless of what load the amp is powering, that level/setting on the "gain knob" if properly matched should never change given the same source unit and the same amp.

The knob on the amp never changes--4V or 2V or .25V on that "gain" knob is always 4V or 2V or .25V isn't it? Though I understand there are not always settings shown on the knob what I mean is the location to properly match that level for the input stage never changes. And there are many amps that do have markings on them. So there's no reason to set it differently by measuring a no-load output and dreaming about what load you're going to drive later.

You gave the explanation of the rail dipping and the amp starting to clip, which I understand. However, then what that amounts to is saying "yes the gain knob is a volume control". Because then you're saying oh well you match the amp to 4V for a 4V output if you're not going to drive the amp hard. But if you drive the amp hard (and let's keep in mind we're still powering a load the amp is *designed* to drive) then we gotta take it easy on the amp and not match the input to the HU output but instead set the sensitivity to something else--say 6V when we have a 4V output from the HU. That may make sense from the point of view of not clipping the amp but it does NOT make sense from the point of view that "gain is not a volume control". What that says is "forget matching the input to the output but instead match the input such that the amp doesn't get driven too hard". I don't buy it. The amp is designed and engineered to produce a given output power level into a given load impedance, when the input sensitivity is *matched* to the input level the amp gets. Not to some sissy level because the amp can't handle it because the amp is working harder. That's not my problem. The amp has to be able to produce it's rated output into rated impedance with the proper signal level.

Never have I once read in any amp manual that input sensitivity changes based on the load you drive. Sure enough there are many amps that actually have voltage markings on the gain knob. None of them have two or three points for the same voltage saying "here's 4V for driving 1 ohm, here's 4V for driving 2 ohms, etc.". Because if it is genuinely the way you say and we have to set the input differently because the amp is working harder, then there ought to be some mention of that in at least one amp manual out there, and that I have never seen. Note I'm not saying you're either right or wrong I'm just saying that if this tutorial is somehow actually right and what you say is indeed right, then gain is a volume control and we set it lower (to a higher-than-actual voltage given) to prevent overdriving the amp. If gain truly is not a volume control what I said is true meaning you always match the input to the HU output and ask no questions about it.

Edit: Yet one more reason I don't buy this whole gain is a volume control because the amp is a sissy business, lol:

Say you buy an amp from any manufacturer. They give the rated input sensitity range as whatever it is. Say it's .25V to 8V. Fine that's the spec. Does it ever say anywhere that the amp will only take an 8V input when you load it with 4-ohms instead of 1-ohm? Or going the other way does it ever say it will only produce full power with a .25V input if you load it with a 1-ohm load instead of a 4-ohm load? NO. It never does. But what this method says is exactly that. It implies that if you load the amp heavily then you can never use an 8V input to the amp; if you don't load it heavily you can never give it .25V and get full output. That's simply nonsense (or the manufacturer giving a rated range is nonsense, in absence of telling you when you can and can't use those voltages).

 

[thch 10+ year member]

the maximum voltage gain the amp can support without artifacts is based upon supply regulation and output impendace of the amplifier.

if the supply can be reduced, then to garuntee no clipping (if such is the goal), the gain must be set based upon these low line conditions, giving a slightly lower gain setting.

if the output impedance is high, then at full scale output, the load will pull down the signal a bit, reducing the gain a bit.

the penelty for such a small amount of clipping is generally low.

the DMM method does match input sensitivity indirectly. the input is assumed to be "the desired input", and the output is adjusted (by gains) to give "the desired output". if done, this means full-scale on the input gives full scale on the output.

which is not to say there are not other uses for gains. level matching can be accomplished. a user might even get an SPL meter and adjust it such that a certain volume setting gives 110dB, thus setting the volume's range based upon what they want to listen to, and not the capabilities of the equipment.

 

[helotaxi 5,000+ posts]

The amps I have seen a legit spec for the input voltage range on specify the load into which the range is valid. As with all amps specs, input voltage range should be taken with a grain of salt because without supporting numbers to go along with it, it really means nothing.

Matching levels is still what you're doing. You are matching the input level of the amp to the output at that impedance. I think that your misunderstanding comes from trying to apply theory to reality without accounting for the losses that separate the two. In a perfect world you would set the gain based solely on the voltage range markings on the input sensitivity control and that would be that regardless of the load that the amp was driving. You would also be able to get twice the power by halfing the load on every amp. We know the latter isn't the case so why should we expect the former. The amp simply isn't that smart, most are quite dumb in fact. They take an input signal and try to multiply it by a certain factor and you have to tell it the desired factor. Even if it has no prayer of pulling it off it will still try.

As I said above, an amp that is rated to double its output every time the load is halved, is underrated at the higher impedances. If you set the gain on one of those amps with a 4 ohm load to give you rated power, the setting should be pretty close all the way down to the lowest rated impedance of the amp. Most amps aren't rated that way though. They are rated with the higher high impedance power that they had to make in order to make their low impedance power. They are not more or less sensitive at any given impedance they are simply not rated on any kind of scale. Take our example from before that doubles powr as the load drops and set it to produce not rated, but maximum power at the higher impedance. You will discover two things: 1)it makes a lot more than rated power and 2)if you tried to run it into a lower impedance load without resetting the gain you will probably clip the hell out of it. Since a low 4 ohm rated power is no longer advantageous in competition, amp makers rate their amps with the actual power it can produce rather than way underrating them at 4 ohms so they will make their low impedance rating when the higher current on the output starts to sag the voltage rails.

Honestly the perfect analog amp would have a button on the end. You would use that button to set the amp up once and then you wouldn't have to worry about it. The way the button would work is that you would turn the source volume all the way down and press the button to mark the zero point. Then you would turn the source volume all the way up and press the button again setting the max position. The preamp would then be smart enough to set its own gain based on that range giving you full unclipped power at max volume and silence at min volume and the full range in between. The other option would be a standard digital signal and data bus protocol that would leave the signal digital until the amp and make gain setting a thing of the past all together.

 

[Kimb0 10+ year member]

Excellent Write up deserves a sticky ?

 

[helotaxi 5,000+ posts]

It should be linked to in the sticky at the top of this forum.

 

[thch 10+ year member]

helotaxi, i take it that multi-paragraph post wasn't in reply to mine, i didn't really see anything that disagreed with what I had posted.

it really shouldn't be too hard to make a clipping detector on the amplifier side. during normal operation the desired signal can be compared to some fraction of the amps output (and is). this difference should be 0V or very close. significant clipping occurs when this difference deviates from 0V by a good amount.

of course outside of the amp is another issue, since you don't know what the amp was trying to do, nor what it was capable of it.

 

[helotaxi 5,000+ posts]

Not in reply, no. More a translation than anything.

I actually have several old amps with clipping indicators built in. Auto setting gain would be much easier.

 

[thch 10+ year member]

well, yes. but controlled amplifiers aren't too popular in hifi. plus, you'd need memory to keep this setting. this pretty much limits the options to either a motorized potentiometer, or digital conversion (probably ADC-DSP-DAC).

so who knows, maybe one day auto gain setting will be done. of course by then a digital interface would be available and there would be no issues.