Hornresp

Hello David,

Could you give the exported "Hornresp record" for the data input of the le Cleac'h horn + adaptor you gave the example.

Best regards from Paris, France

Jean-Michel Le Cléac'h


Hi more10,

The best that you can do is to specify a chamber and throat adaptor using parameters Vtc, Atc, Ap1 and Lpt - see attached screenprint.

In the example shown, the throat adaptor half-angle has been made equal to the horn throat entry half-angle.
 
Could you give the exported "Hornresp record" for the data input of the le Cleac'h horn + adaptor you gave the example.

Hi Jean-Michel,

The exported record file is attached as requested. The Vtc, Atc, Ap1 and Lpt values were chosen simply to clearly show the adaptor - they are not necessarily very practical.

Kind regards,

David
 

Attachments

  • adaptor.txt
    421 bytes · Views: 59
Hi Lars,

I can't see how an optimum design can found using just f1 and T, but not f2.

Your subbashorn.xls spreadsheet appears to require all three parameters.

Am I missing something here?

Kind regards,

David

Hello David,

Off course parameter input is necessary as designers must take some decisions..

Note, we are talking about "System Design" now.

I can´t see how an optimum design can be found not deciding T. In my eyes T is far more important than f2. If the design using your choosen T and f1 doesn´t have a high enough f2 you have to find another driver.

So what I wanted to find in the SD is a possible the choose any two out of three parameters.
 
So what I wanted to find in the SD is a possibility to choose any two out of the three parameters f1, f2 and T.

Hi Lars,

While I don't think it is theoretically possible to arrive at an optimum solution using just f1 and T, I would be very happy for you to prove me wrong by developing a spreadsheet that achieves this. If the results calculated by your new spreadsheet proved to be satisfactory, then I could easily incorporate the method into the Hornresp System Design tool.

Kind regards,

David
 
Hey David,

If you use my present spreadsheet and use the method I recommend by deciding T and f1 and thereby achieve reactance annulling by changing f2 you will arrive at the same figures as in HR.

As both HR and the spreadsheet is based on Leach´s formulas they should.

I am not suggesting a change, I am suggesting an additional option for those who premieres T before f2 when designing.

I am not sure , but do you mean "System Design" arrives at the optimum f1 ,f2 and T in its initial suggestion?
 
I am not suggesting a change, I am suggesting an additional option for those who premieres T before f2 when designing.
I am not sure , but do you mean "System Design" arrives at the optimum f1 ,f2 and T in its initial suggestion?

Hi Lars,

I thought that what you were proposing was for the Hornresp 'System Design With Driver' tool to have f1 and T rather than f1 and f2 as user inputs. Assuming that Ang and the driver parameter values were correctly specified, the tool would then be expected to determine "optimum" values for S1, S2, L12, Vrc, Lrc, Vtc and Atc, without any knowledge of f2. At the moment, I can't see how this can be done.

Kind regards,

David
 
What should the difference be? We have three parameters that normally are adjusted for achieving for reactance annulling.

"System Design" finds T with the help of f1 and f2. Why couldn´t f1 and T be used to find f2 instead?

But as I have no knowledge at all about what´s inside HR this might as you say be impossible to achieve. After all one have to use trial&error in my spreadsheet to find f2 after inputting the two other parameters.
 
What should the difference be? We have three parameters that normally are adjusted for achieving for reactance annulling.

"System Design" finds T with the help of f1 and f2. Why couldn´t f1 and T be used to find f2 instead?

But as I have no knowledge at all about what´s inside HR this might as you say be impossible to achieve. After all one have to use trial&error in my spreadsheet to find f2 after inputting the two other parameters.

I believe reactance annulling is not reliable anyway, since it appears that anything that affects the acoustical impedance on the horn side will change the required rear chamber size. This means that if the horn is not in a perfect 1, 2 or 4 pi space, then the rear chamber is not going to be the perfect size. So, it seems to me that getting the rear chamber close to the right size is about as good as it gets. So, I'd rather just design from the driver parameters and pick the lower cutoff frequency as desired.
 
I believe reactance annulling is not reliable anyway, since it appears that anything that affects the acoustical impedance on the horn side will change the required rear chamber size. This means that if the horn is not in a perfect 1, 2 or 4 pi space, then the rear chamber is not going to be the perfect size. So, it seems to me that getting the rear chamber close to the right size is about as good as it gets. So, I'd rather just design from the driver parameters and pick the lower cutoff frequency as desired.

The point of proper reactance anulling is to get the Fs desired in the acoustical loading you are in. That is why it is not so easy to do in the first place. It can have very benifical results where properly implemented.
 
The point of proper reactance anulling is to get the Fs desired in the acoustical loading you are in. That is why it is not so easy to do in the first place. It can have very benifical results where properly implemented.

Based on simulations in Hornresp, it doesn't seem to have much of an effect on Fs. There is a point where Fs is lowest, and making the rear chamber smaller does raise the Fs, but making it larger beyond a certain point doesn't have any effect once the lowest Fs is achieved. Also, when the impedance curve is flat the Fs is not at it's lowest point, at least for the horn types I've played around with in simulation. Maybe real world tuning is different, I don't know.
 
Dirkwright,
I recommend you study Marshall Leach´s AES-paper to fully understand what happens. HR is based on this.

You might also have some use of my spreadsheet http://www.revintage.se/subbashorn.xls where you can see how the result of Leach´s equations interact.

As Mark points out you will always have to do compromises but the starting point for a design must always be the theoretically right one.
 
Last edited:
After all one have to use trial&error in my spreadsheet to find f2 after inputting the two other parameters.

Hi Lars,

Exactly - as you say, it is still necessary to specify a value for f2, otherwise a solution cannot be found. To illustrate, setting f2 to a blank field in your spreadsheet generates invalid results.

I think it is probably better not to change the Hornresp System Design tool so that it becomes necessary to use a trial and error process to find the answer - it sort of defeats the purpose of having the tool in the first place :).

Kind regards,

David
 
Hi Lars,

So is it a problem to input f1 and T, to get f2?

As far as I can see, it is impossible - but as I said in an earlier post, I would be happy to be proven wrong :).

Think I have a seen a quite similar spreadsheet in this thread that solves f2 w/o trial&error.

I would be grateful if you could track it down for me - I would be very interested in having a look at how it works.

Kind regards,

David