Transfer Function?

[speakerman39]

I asked this question over in the DIY section but not getting any replies. So, thought I would try it here and see what happens. I just recently downloaded WinISD Pro Alpha but have some questions. Wolud anybody here mind to take a few minutes and explain the following:
1. Transfer Function Magnitude
2. Transfer Function Phase
3. Group Delay
4. and 5. Max Power/Max SPL
6. Amp Apparent Load Power (VA)
7. SPL ( I know what this is)
8. Impedence ( Have a good idea of what it is but the graph confuses me)
9. Impedence Phase

Please explain to me what I should be looking for when modeling a specific driver. In other words, I do not need an overly thorough explanation BUT rather one that is straighforward and to the point w/o all the theory. Not looking for a physics lesson just what I need to know if I do attempt the DIY subwoofer project I have in mind. Thank you for your time.

 

[annunaki]

I will give you a basic look on the most important ones.

1. Transfer Function Magnitude: This is the anechoic response curve of any given woofer within the enclosure design specified in the program. Room interaction will change this when the woofer is placed in a room with this enclosure.

2. Transfer Function Phase: This is the phase response of the woofer in the enclosure specified. It will show you how far out of phase the response of the woofer will go at a said frequency

3. Group Delay: This will show any major detractions in transient response. Typically used when doing ported/bandpass/passive radiator enclosures, it will show if the woofer is slower to respond in a specified enclosre at any given frequency range.

4 & 5. Max Power/SPL: This is to show how much power is required in a specified enclosure to reach a given output level in an anechoic environment.

6. Amp Apparent Load Power: This is either is how much power you have available, or like above shows how much power is being used at a certain frequency.

8. Impedance: This just shows the impedance of the woofer at a given frequency withing the specified enclosure.

9. Impedance Phase: This shows the phase angle of the impedance at a said frequency.

What you want to look for is the smoothest/flatest response curve from 20hz-80hz when set to Transfer Function Magnitude. You do not want any large bumps above or dips below the reference point (listed as 0db). This will be the most important thing to look at. You will also want to check out the excursion level in a specified enclosure with specified amplifier power. You do not want the woofer to be over exerting itself within the useable range. You will also want to check the vent mach to make sure that you should not have vent noise issues. Impedance you can look into for any significant dips, but most good amplifiers should have no problem with most woofers.

Hopefully this helps.


*Disclaimer* I could be off a bit on some of my descriptions above as I am going from memory. A wilkpedia search is usually very beneficial for in-depth explanations.

 

[speakerman39]

I will give you a basic look on the most important ones.

1. Transfer Function Magnitude: This is the anechoic response curve of any given woofer within the enclosure design specified in the program. Room interaction will change this when the woofer is placed in a room with this enclosure.

2. Transfer Function Phase: This is the phase response of the woofer in the enclosure specified. It will show you how far out of phase the response of the woofer will go at a said frequency

3. Group Delay: This will show any major detractions in transient response. Typically used when doing ported/bandpass/passive radiator enclosures, it will show if the woofer is slower to respond in a specified enclosre at any given frequency range.

4 & 5. Max Power/SPL: This is to show how much power is required in a specified enclosure to reach a given output level in an anechoic environment.

6. Amp Apparent Load Power: This is either is how much power you have available, or like above shows how much power is being used at a certain frequency.

8. Impedance: This just shows the impedance of the woofer at a given frequency withing the specified enclosure.

9. Impedance Phase: This shows the phase angle of the impedance at a said frequency.

What you want to look for is the smoothest/flatest response curve from 20hz-80hz when set to Transfer Function Magnitude. You do not want any large bumps above or dips below the reference point (listed as 0db). This will be the most important thing to look at. You will also want to check out the excursion level in a specified enclosure with specified amplifier power. You do not want the woofer to be over exerting itself within the useable range. You will also want to check the vent mach to make sure that you should not have vent noise issues. Impedance you can look into for any significant dips, but most good amplifiers should have no problem with most woofers.

Hopefully this helps.


*Disclaimer* I could be off a bit on some of my descriptions above as I am going from memory. A wilkpedia search is usually very beneficial for in-depth explanations.

I appreciate your explanations BUT what I am looking for is how to interpret the graphs. In other words what are the thresholds I should be looking for both good and bad???? No graph is perfectly smooth BUT what is considered good or what is considered bad. More specifically, what is considerd real good and real bad???? I have read several definitions of the terms BUT in the real world how does this apply??? Once again I appreciate your help.

 

[annunaki]

On the transfer function magnitude graph, you will not want any deviations of more than +/-3db. However, in most circumstances you should be able to figure the enclosure size for response within +/-1db. Most of the enclosures I have designed with the program are all flat as in they do not deviate from the reference plane (0db) until they approach the tuning frequency (or Fs in the case of sealed) or reach the tuning frequency (or Fs). As I stated before this is going to be the most important thing to look at. If the response is not there, it just won't perform. Only once the response is satisfactory do I begin checking the other parameters to make sure they are in line.

Once the frequency response is the best I can get it, I then check excursion levels to make sure it is not exceeding the limits of the sub. Remember that you can usually exceed linear xmax by 10% before audible distortion. If I am using a vented enclosure, I also check the air speed and/or vent mach to make sure noise should not be a problem. I also then check the impedance curves to see if the sub is a real reactive load or not.

Remember that your results are only as good as what you put in. The more accurate your input parameters, the more accurate the results. If you are fortunate enough to have all of the driver parameters, you will want to shut off the "auto calculate unknowns" feature. Most subs I have seen do not include ALL parameters.

 

[speakerman39]

On the transfer function magnitude graph, you will not want any deviations of more than +/-3db. However, in most circumstances you should be able to figure the enclosure size for response within +/-1db. Most of the enclosures I have designed with the program are all flat as in they do not deviate from the reference plane (0db) until they approach the tuning frequency (or Fs in the case of sealed) or reach the tuning frequency (or Fs). As I stated before this is going to be the most important thing to look at. If the response is not there, it just won't perform. Only once the response is satisfactory do I begin checking the other parameters to make sure they are in line.

Once the frequency response is the best I can get it, I then check excursion levels to make sure it is not exceeding the limits of the sub. Remember that you can usually exceed linear xmax by 10% before audible distortion. If I am using a vented enclosure, I also check the air speed and/or vent mach to make sure noise should not be a problem. I also then check the impedance curves to see if the sub is a real reactive load or not.

Remember that your results are only as good as what you put in. The more accurate your input parameters, the more accurate the results. If you are fortunate enough to have all of the driver parameters, you will want to shut off the "auto calculate unknowns" feature. Most subs I have seen do not include ALL parameters.

Hello, once again I appreciate your help. Would you mind to take a sub and use it as an example? Perhaps something like the Dayton 12" Reference HF driver and model it in a sealed enclosure AND a vented enclosure?? Maybe post some of the graphs and include your interpretations??? If you can it would be very useful for me as well as others out there. If this is too much work or takes too much time I understand. But, it never hurts to ask right...????? Have a good evening my friend.

 

[Warpdrv]

What are dangerous mach numbers for the vent...?

I haven't been able to figure that out... How high is too high...?

 

[annunaki]

Hello, once again I appreciate your help. Would you mind to take a sub and use it as an example? Perhaps something like the Dayton 12" Reference HF driver and model it in a sealed enclosure AND a vented enclosure?? Maybe post some of the graphs and include your interpretations??? If you can it would be very useful for me as well as others out there. If this is too much work or takes too much time I understand. But, it never hurts to ask right...????? Have a good evening my friend.

I will when I get some time. It may be a few days though. Sorry.

 

[speakerman39]

What are dangerous mach numbers for the vent...?

I haven't been able to figure that out... How high is too high...?

Very goos questions. Not really sure either. Maybe annunaki will shed some light on this for us.

 

[speakerman39]

I will when I get some time. It may be a few days though. Sorry.

Hey man I understand as work has been taking up a lot of my time. Stress is stress and no matter what it sucks. Just when you have the time my friend. We will be awaiting.

 

[annunaki]

What are dangerous mach numbers for the vent...?

I haven't been able to figure that out... How high is too high...?

One threshold I have seen is 64ft./sec. This should be well below any audible port noise.