Peerless HDS-P830875 Dayton ES25NdXP-4

ThumperTom

New project for IowaDIY using the Peerless HDS-P830875 and Dayton ES25NdXP-4.  The box is going to be about .5 cf tuned to 50 hz and matching the knockdown box from PE that is 8.5 x 16 x 11.  The center of the tweeter is 4 inches from the top and the woofer is 10.5 inches from the top.  The sim XO is about 1.5k to 1.8k with 3rd order on the woofer and 4th order on the tweeter.  



My OmniMic has been having software issues with windows 10 but I believe I was able to capture some good files.  I am going to attach the ZMAs, FRDs, and PCD files so if other would like to play around.  The FRDs have the blended response and were measured at 96 kbits from 24 inches on a mock up baffle matching my final measurements. 

ThumperTom

How do you go about attaching files?  I did just the individual files and I just tried a zip file.

ThumperTom

I have not been able to upload files, I guess I am going to have to use dropbox or some other form but here are some more screen shots of the crossover.

jhollander

imo I'd add a bit more baffle step

ThumperTom

I am thinking that too, I am hoping to have the boxes complete this weekend or early next week to see if it raises the bottom end.

ThumperTom

Added a little more baffle step.

ThumperTom

When calculating tuning and port length, do you trust UniBox or WinISD Beta?  UniBox is wanting to tune lower and show a flatter response but WinISD is showing more of a downward slope before dropping off.  Most of the time they follow very close to each other.  I don't think I can split the difference like I normally would because the tuning is 45hz to 50hz and my 2" pipe is goes from 4.5" long to 6" long.  The volume of the box is .53 cf or 15l, it will be lined, and the exit will have a round over.

Here are the measured parameters.
Peerless HDS-P830875

* f(s)= 41.72 Hz

* R(e)= 6.36 Ohms

* Z(max)= 54.33 Ohms

* Q(ms)= 2.911

* Q(es)= 0.386

* Q(ts)= 0.341

* V(as)= 21.700 liters     (0.766 cubic feet)

* L(e)= 1.07 mH

* n(0)= 0.39 %

* SPL= 88.00 1W/1m

* M(ms)= 20.19 grams

* C(ms)= 0.72 mm/N

* BL= 9.34 

JasonP

I remember reading that WinISD models box losses somewhat differently than unibox. My experiences have shown results closer to Unibox than WinISD, in limited testing.

On the crossover: Listen to that as it is, and then listen to it with a 1-2 ohm padding resistor on the tweeter shunt inductor. I'm seeing ringing at ~1.5k in that filter. It may or not be audible.

PWRRYD

JasonP said:

I remember reading that WinISD models box losses somewhat differently than unibox. My experiences have shown results closer to Unibox than WinISD, in limited testing.

Oh boy Jason... don't let Grumpy Chris Romer hear you say that!  I swear he is on WinISD's marketing payroll.  No seriously.

jr@mac

The software all make different assumptions. WinISD makes no assumptions, so it requires  digging into the advanced modeling features. 

jhollander

I like uni box but the port end corrections need tweaking every now and then.  Seems my ports get closer to the walls than I've anticipated.

ThumperTom

After a few reworks, I got one box put together. 

ThumperTom

I might have something worth listening to.  I think the polar plot looks good but would like some feed back.  Don't trust the stuff in the lower but more concerned about the xo region and tweeter.

dcibel

If you want your polar plot to accurately represent the change in directivity, normalize everything to the 0 deg response. The polar will then represent only the change in response, it will be much easier to see the directivity pattern.

ThumperTom

dcibel

That doesn't look right. If it was normalized to 0 deg, your 0deg measurement should be a flat line, and it should look similar to this.

dcibel

To do this, it's just the normalize button under the "Add curves" menu. Then pick your 0deg measurement.

ThumperTom

I think I did it right this time