More compeling, but still hazy. On what is this number based? What could it possibly mean in a situation where there are several listening positions? All of them cannot meet this requirement.
Last edited:
Wonder if anyone wants to get back to measurement of HOMs
Soongsc
I can tell you precisely how to do it and I would love to see the results! Let me know and I'll describe the procedure and you can get started:
Step 1) build an anechoic chamber. It should have one solid wall that can accept a waveguide, preferably of arbitrary size. The waveguide mouth must fit flush with this wall. This wall must be > 10 feet on any axis.
When you get that done let me know and I'll get you step 2.
This one is easy.
Dig a two foot deep hole in your backyard (away from any buildings). Point the horn up toward the sky. If you insist then lay some scrap plywood with a hole in the middle for the horn's mouth. The microphone can be supsended from a wire or boom.
Why not simply test outdoors on a windless day, waveguide flush with the ground, far distant from any boundaries?
Or on a baffle similar to it's intended use, elevated sufficiently?
Weltersys and Tarragon,
I agree that this is an effective test method and was pointed in this direction years ago. With some of the analysis software that I am aware of from the past it is even possible to remove the aperiodic background noise from the analysis through repetitive impulse response excitation and subtraction methods. The early SysId system worked this way and would do just that. Multiple log chirp impulse responses were used sweeping the bandwidth and using a matching sweep windowing function would take multiple sample and could subtract the added background noise or room reflections. I can't tell you much more than that but know I have seen it done that way.
I agree that this is an effective test method and was pointed in this direction years ago. With some of the analysis software that I am aware of from the past it is even possible to remove the aperiodic background noise from the analysis through repetitive impulse response excitation and subtraction methods. The early SysId system worked this way and would do just that. Multiple log chirp impulse responses were used sweeping the bandwidth and using a matching sweep windowing function would take multiple sample and could subtract the added background noise or room reflections. I can't tell you much more than that but know I have seen it done that way.
Great idea, I live in the woods, can't do that, but let me know when its done.
I can't do that either. Its not like I don't know any of these suggestions, I am not dumb. I don't have the capabilty to make the measurements that are required. I can tell others how to do it but I cannot do it myself.
That is extremely old technology. Today, just use HolmImpulse, no better system exists and its free. What's not to like.
But Holm still cannot do free field measurements when there is no free field. Nothing can.
At the University in Trondheim (where I currently work) we have a fairly large anechoic chamber, that could easily take a 10x10 ft wall (or floor, that would be the easiest).
Are the next steps something similar to what Makarski did; scanning the velocity profile of the waveguide mouth and decompose it into modes?
Regards,
Bjørn
Earl......don't know if this has been adressed prior or by yourself but Dayton has a new 1" CD that's supposedly an excellent clone of the B&C 250 that retails for less than half. I have no experience with it yet, but love the 250s and agree with your impressions of them. Have you tried/tested the Dayton d250p?
Dayton Audio D250P-8 1" Polyimide Compression Horn Driver 270-402
Dayton Audio D250P-8 1" Polyimide Compression Horn Driver 270-402
I don't know whether you have done it before. I am hesitant because it seems that it will not give me the results I am looking for.
It seems to me that taking measurements from the mouth to the throat should give me data that might make sense to me. I have measured some points like that a while back, but have not done it with all the waveguides I have yet.
Yes, I have, they are comparable, but there price is not low enough for me to switch. I don't pay retail for my drivers and parts express does not have very attractive discounts for manufacturers.
The price of that Dayton driver is incredibly low. I would say that the cost to a manufacturer if we are going to call speaker box builders manufacturers is incredibly low. The cost of just a beryllium diaphragm alone would probably exceed the cost of the wholesale price of that driver. I am still not buying that all compression drivers are interchangeable but those are compelling costs for using the driver if it works even reasonably in the intended frequency range. I still would like to see some comparisons of actual un-smoothed frequency response on a plain wave terminated tube and harmonic distortion measurements before I cast any judgements.
The Daytons can be sent back if you don't like them. I measured them and listened to them and although they measure and sound better then the B&C (sound better to me) I sent them back because I'm more of a large format driver guy and small format drivers to me are only used from 5K or higher up for treble. I have other drivers that perform better for that purpose. For 50 bucks get one and measure it, what do you have to loose?
Compression drivers surely are not 100 percent interchangeable.
Compression drivers surely are not 100 percent interchangeable.
Pooh,
I have used many 1" exit compression driver down into the 1.5Khz region many times and some old Altec drivers down to 500hz on my Barcelona's sitting in my living room after all these years. I personally have never like anything larger than a 1.4" exit Altec for anything I have done and never would use a 2" exit driver for anything. I just wonder how these polyimide plastic diaphragms can compare to a metallic diaphragm as the weight and stiffness is so much different from a metal diaphragm. I do remember the old Altec phenolic diaphragms and don't really like those. I just can's see how a low stiffness material such as the polyimide would not have many more breakup modes than a much higher stiffness material?
I have used many 1" exit compression driver down into the 1.5Khz region many times and some old Altec drivers down to 500hz on my Barcelona's sitting in my living room after all these years. I personally have never like anything larger than a 1.4" exit Altec for anything I have done and never would use a 2" exit driver for anything. I just wonder how these polyimide plastic diaphragms can compare to a metallic diaphragm as the weight and stiffness is so much different from a metal diaphragm. I do remember the old Altec phenolic diaphragms and don't really like those. I just can's see how a low stiffness material such as the polyimide would not have many more breakup modes than a much higher stiffness material?
They don't sound like phenolic, they sound like a smoother, gentler Altec 806A without loosing the detail - maybe a bit subdued for classical tastes. Probably a step up or sideways to the Radian aluminum with soft surround - I think they are better damped then most metal drivers. I do like the Dayton's and would recommend them in a system 1.2k -1.5k up - the SEOS waveguide Parts Express sells is really pretty sweet too and works well with the Parts Express driver.
I thought the task of research was to find the correct results, not "the results I am looking for."
It would be possible to do this, but not the easiest way. You would need a mic that was so small that it did not disturb the sound field and you would have to have a very good handle on the Oblate Spheroidal solutions, which are very complex. Doing this for an arbitrary waveguide shape would not be possible since you don't have a solution to fit the modes to.
When a wave guide generates HOMs, the amplitude reduces with distance, why would you measure so far away? Additionally, the purpose of measuring is to compare how certain phenomena can be brought to minimum demonstrated as close to the source as possible. Measuring at a further distance does not seem logical. Clear indication of reduction is more important than being totally accurate since I am not trying to match measurement with a mathematical solution, rather, I am trying to correlate measurement with listening perception and reasonable indication of lower HOMs with listening perception.