Can someone comment on my hornresp simulations on the MCM8?

blamus

Member
2009-02-25 3:06 am
Just started to learn how to use hornresp. can the TH experts give me some comments just for educational purposes? the aim is to get as high SPL as possible in as small an enclosure between 35-80Hz, in a vehicle powered by a 70W amp at 2 ohms.

best regards
 

Attachments

  • THdual1.txt
    407 bytes · Views: 58
  • THdual2.txt
    412 bytes · Views: 18

MaVo

Member
2006-01-07 12:40 pm
Number 1 looks fine to me. Its interesting how you got that design to work by using such a long L12. I usually try to make L12 as short as possible, but here it seems to be better your way.
Number 2 has this deep notch directly above the passband. If the real box is just slightly off (which is very likely to happen - happened to me all the time), it could jump right into the passband.
Try to make your designs a little "safe" or "stable" concerning parameter variation, because what you build will never be a 100% representation of what you design.

All that said, i would go with no.1.
 

blamus

Member
2009-02-25 3:06 am
I made some improvements - i think!

Thanks Mavo for the comments!

Thats a very valid point about the huge notch may get into the passband when built, I never even thought about it that way.

I made some adjustments/improvements - smaller and louder:

Now all my simulations are done in 2pi and 2.83v, so I can compare to other box programs such as winisd

3. Flat from 40-100Hz at 99db, at 75L - This is a variation from TH1 above, smaller, higher SPL, flatter.

4. 30-80Hz, 93-99db, rising response at about 6db/oct. 46L. This I think I like. I believe the response shape here is perfect for cabin gain. Larger than TH2 above, the notch is still there but at a higher freq, way above the low pass would be used.

displacement seems fine at 12V ([email protected]) if I high pass at 25Hz.

This time I have included both screenshots and the honresp export, so people can just look at it or import it if they want to play with it (please!)


What do people think? Are these designs worth pursuing? As in, can it be improved much more?
 

Attachments

  • TH dual 3.jpg
    TH dual 3.jpg
    156.7 KB · Views: 300
  • TH dual 4.jpg
    TH dual 4.jpg
    168.1 KB · Views: 296
  • thdual3.txt
    409 bytes · Views: 11
  • thdual4.txt
    413 bytes · Views: 18
What do people think? Are these designs worth pursuing? As in, can it be improved much more?

Caution: I don't think HornResp includes losses in its modelling. What looks flat in the same might not end up being flat when you measure it. Rolloff might end up being higher and faster than predicted, for example. On the plus side, the response is likely to be smoother than predicted.
 

blamus

Member
2009-02-25 3:06 am
Compression (area of horn at throat entry vs Sd) looks a little high. I tend to aim for values lower than 4 with these drivers.

You mean the ratio hornresp shows when I put the mouse over S2? I have done a bit research there and apparently jbell has tried using a compression ratio up to 8:1 and its fine for these particular drivers. But I maybe wrong.


Caution: I don't think HornResp includes losses in its modelling. What looks flat in the same might not end up being flat when you measure it. Rolloff might end up being higher and faster than predicted, for example. On the plus side, the response is likely to be smoother than predicted.

So do you think sim 3 would work better even with cabin gain? Which one do you think would work best in the average car? Or do you think none of them are any good? Am I doing the right thing/heading in the right direction? I really have no idea!
 
You mean the ratio hornresp shows when I put the mouse over S2? I have done a bit research there and apparently jbell has tried using a compression ratio up to 8:1 and its fine for these particular drivers. But I maybe wrong.

4 to 1 is usually safe. 8 to 1 is asking for some blown drivers if they are not tough enough. Remember that the ratio is describing the area of the cone to the area of the throat. That means that what ever pressure is being created by cone movement is being amplified by roughly eight times.

Mark
 

jbell

Member
2003-04-02 4:34 am
.
yea, 8:1 with a compression chamber works great -- I've yet to rip a cone in that configuration -- and I've tried. 8:1 FLH with chamber is the most efficient config I've found. I've yet to find a TH config for the mcm that will beat the FLH's I've done.

I'm not sure why you are using 2.83 volts and 2pi with a pair of 4ohm drivers in parallel?

If you want a 1 watt reading @ 2 ohm, you need to be using 1.41volts.
 

jbell

Member
2003-04-02 4:34 am
.
Have you ever tried the 8:1 without a compression chamber? And what size do you use?

The compression chamber is THE KEY to flat response with the mcm and FLH design. Depending on the design -- anywhere from 4-20 liters. It's amazing how changing the compression chamber size really affects response. One of my test cabinets, I intentionally left the chamber too big, and played with closed cell polystyrene foam to test/measure the response differences. That was a real learning experience.

One thing I did find out, that at 8:1 compression ratio, a more 'square' aspect throat sounds better than a high aspect ratio throat -- aka, 2x2 sounds better than 0.5 x 8. That I don't have a real good explanation for....
 

tb46

Member
2006-01-09 7:04 pm
Texas
Hornresp version problems

Hi blamus,

I just tried to import your thdual4.txt, and it imports fine, but there seem to be some version problems. What version of Hornresp did you use? I'm using Version 28.10 (2810-101129).

To make the woofers fit you have to either add a chamber between the woofers and the horn, or provide at least one woofer diameter as L12, the same hold true for L34. Ap1 and Lpt are the port dimensions from a throat chamber (Vtc / Atc) to the horn at S2, you don't need them unless you add a throat chamber.

Regards,
 
Last edited:

blamus

Member
2009-02-25 3:06 am
Hi guys,

Thank you very much for your responses.

tb46:

I have increased L12 and L34 to 25cm (about 9") and tweaked the model. See attached. Good that you pointed that out. I really didnt even think "physically" in real life space when I was playing with the sliders. The numbers in AP1 and lpt are just copied from another hornresp record.

I have now removed it.

Turns out I am using version 28.00 I hope there isn't much that would change my simulations in the different version. I shall update soon.

Jbell:

Thanks for the heads up about the compression ratio significance between THs and FLHs and compression chambers. I honestly didn't know the differences until now.

I used 2.83V and 2pi for all my simulations just for comparison. I started with just one MCM driver so thats why it was 2.83V. All that matters is that I know I will have 12V ([email protected]) in 0.5pi (in car) at the end of the day. It doesn't matter as long as I make sure I have the same input voltage for all my simulations right?

Mark:

you have mail :) and thanks again for all your patience!

I have my MCM woofers in the mail. No turning back now!

Best regards

Bern
 

Attachments

  • dual MCM TH.jpg
    dual MCM TH.jpg
    158.4 KB · Views: 196

blamus

Member
2009-02-25 3:06 am
I do have some other queries.

1. In some designs an inductor is used to tame the response. If I do that, do I just add the inductance of the coil to the Le of the driver specs in hornresp? Because that does seem to affect the response alot.

2. When EQ is used on bass horns, be it TH or FLH, is the effect of the EQ altered by the horn as well? For example, if I have a 3db dip at 60Hz, and I "EQ-up" at 60Hz 3db, would that actually work? Or would that boost also affect other freq in strange ways and at different levels (e.g boost 6db instead)?
 
Bern Driver.JPG

In the Horn Wizard there is a drop down menu on the bottom right. One of the things on the list is the driver. In the driver parameters you can add in inductance. This will create a lowpass filter effect.

As for EQ you can shape the response of a horn in it's designed passband. If you try boosting it below it's passband you will have to buy drivers fairly regularly. The effects on the EQ on the adjacent areas of the frequency response depend on the Q of the filter used. High Q means a narrow filter width. Low Q a broader filter width.

In a car sometimes the best thing to do is move the sub around a bit. It will change the response quite dramatically. Same goes for a house actually.

Trying to tame low frequencies any other way gets increasingly complex and ineffective. Location and number of points being excited in the room are the most important points to work with.

Mark
 

blamus

Member
2009-02-25 3:06 am
ok cool, so eq works just the same way as with sealed boxes. And when I add a coil, I'd also have to add the resistance.

Now that brings me to wonder, if I was to use a passive high pass filter - as with most of these bass horns a HP is needed - at say, 30Hz, would the caps and coils do anything unexpected to the response? Or would it just apply the designed high pass effect just the same way as it would if used on a sealed box?

That might be a stupid question with a simple answer, but I have to ask, since horns are sooooo different than a sealed box, and I have never used a horn before.