TESTING Your Finished Project

[Nordo]

Is there an accepted way of testing your finished speaker(s)?

I know the industrial standard is Half Space testing with 2.83v at 1 metre.
But has there been a standard adopted on the BFM forum so that we can compare our speaker project to other peoples, as well as Bill's SPLs?
Having graphs that are easy to compare against each other would also make it easier to pin down any problems with your finished project.

Not only should we have a standard technique for testing that would be relatively easy for most builders to achieve, but also specifications for presentation of the results. e.g. axis range, smoothing, etc.

I know the techniques, presentation, etc, would change from speaker to speaker (e.g. sub or full range), but if we have a standard, then any graphs builders put up in the forum, will have more meaning, and can be easily compared to other builder's graphs.

Just saying.

[Grant Bunter]

Is there an accepted way of testing your finished speaker(s)?

I know the industrial standard is Half Space testing with 2.83v at 1 metre.
But has there been a standard adopted on the BFM forum so that we can compare our speaker project to other peoples, as well as Bill's SPLs?
Having graphs that are easy to compare against each other would also make it easier to pin down any problems with your finished project.

Not only should we have a standard technique for testing that would be relatively easy for most builders to achieve, but also specifications for presentation of the results. e.g. axis range, smoothing, etc.

I know the techniques, presentation, etc, would change from speaker to speaker (e.g. sub or full range), but if we have a standard, then any graphs builders put up in the forum, will have more meaning, and can be easily compared to other builder's graphs.

Just saying.

Since you started another thread lol.

Subs: ground plane testing ie microphone about an inch above the ground, cab bottom on the ground IIRC.
Results need to be equivalent to 2.83V/1m. You may not get a good reading at 1 metre.
So you might, for example, make it 28.3V for output, with the mike at 10m
Results should be smoothed, can't remember the value for that.

Sure it makes it easier to compare apples with apples, but really, if you can give your measurement criteria, then that can be enough to compare.

Always outdoors, with at least 17m of clear space before the nearest, well, anything, so that would include shed, fence, house, large tree etc...

[Nordo]

Since you started another thread lol.

Subs: ground plane testing ie microphone about an inch above the ground, cab bottom on the ground IIRC.
Results need to be equivalent to 2.83V/1m. You may not get a good reading at 1 metre.
So you might, for example, make it 28.3V for output, with the mike at 10m
Results should be smoothed, can't remember the value for that.

Sure it makes it easier to compare apples with apples, but really, if you can give your measurement criteria, then that can be enough to compare.

Always outdoors, with at least 17m of clear space before the nearest, well, anything, so that would include shed, fence, house, large tree etc...

Did a fair bit of ground plane testing many years ago, when I was building speakers for my home theatre.

I assume you are saying that the mic should be further than 1m away due to the size of the mouth of my T39s (and most horn designs).
Is the dB proportional to the voltage and inversely proportional to the distance - or is it logarithmic?
Or are they both logarithmic, therefore adjusting both by the same proportion would result in unity?
Not that it matters really. I'll can do different trials until I get a useable result.

Have reasonably good open areas around the house. Maybe not as good as you.

[Bill Fitzmaurice]

Is the dB proportional to the voltage and inversely proportional to the distance - or is it logarithmic?

SPL drops by 6dB with each doubling of distance. It also increases by 6dB with each doubling of voltage. Therefore SPL/1m/2.83v = SPL/2m/5.66v = SPL/4m/11.32v etc.

[ripNdeb]

I don't think there really is a 'standard' due to environmental conditions. If you measured exactly where Bill measured then you could compare but, if not then i think you have to estimate/compensate etc.
I'm finishing up a 2nd pair of T39s and no matter where I measure I just hope they are all, separately, the same; then I"ll be a happy camper because I'll be pretty sure that none of them is leaky or has other problems.

[BrentEvans]

Is the dB proportional to the voltage and inversely proportional to the distance - or is it logarithmic?

SPL drops by 6dB with each doubling of distance. It also increases by 6dB with each doubling of voltage. Therefore SPL/1m/2.83v = SPL/2m/5.66v = SPL/4m/11.32v etc.

You should drop practical snippets like this far more often.

I also second the motion of a recommended test / measurement procedure. I'd be willing to write it myself for the sticky section given the specs... I know the procedure I use but having it from the man himself would bee nice.

[Bill Fitzmaurice]

I also second the motion of a recommended test / measurement procedure. I'd be willing to write it myself for the sticky section given the specs... I know the procedure I use but having it from the man himself would bee nice.

I'm sure I've posted it before. Measure the speaker ground plane to find the 1/2 space measurement below the baffle step. In the case of a sub you're done. With a top take the ground plane first, then put the cab on the ground face up, with the mic suspended above it at least 2 meters high. Measure again. This gives the 1/2 space measurement above the baffle step. Splice the two measurements where they meet. You have to do the top this way, because a ground plane measurement will be 1/2 space below the baffle step, 1/4 space above it. You can't measure a top with the cab horizontal with the mic on axis either, as that will introduce a ground bounce response notch.

[Nordo]

I also second the motion of a recommended test / measurement procedure. I'd be willing to write it myself for the sticky section given the specs... I know the procedure I use but having it from the man himself would bee nice.

I'm sure I've posted it before. Measure the speaker ground plane to find the 1/2 space measurement below the baffle step. In the case of a sub you're done. With a top take the ground plane first, then put the cab on the ground face up, with the mic suspended above it at least 2 meters high. Measure again. This gives the 1/2 space measurement above the baffle step. Splice the two measurements where they meet. You have to do the top this way, because a ground plane measurement will be 1/2 space below the baffle step, 1/4 space above it. You can't measure a top with the cab horizontal with the mic on axis either, as that will introduce a ground bounce response notch.

I'm getting a bit lost here.
I think you're saying that conducting Ground Plane testing for subs like the T39, "baffle step" can be ignored because :-
1. We would only be interested in the results up to say, 200Hz.
2. Most horn loaded cabinets have no real baffle.
However, due to the size of the horn mouth, would you tilt the cabinet forward to aim the axis at the mic? Or wouldn't it matter due to the low frequencies being used?
Also, for subs only, would Near Field testing be an option? This would make testing easier, and help provide uniformity to individual builder's submitted results.

I understand that testing mains (or any speaker that has useful response above 200Hz) would need to take into account the baffle step.
When I was building my home theatre main speakers about 20 years ago, I can't remember doing a half space test, but I do remember adding a compensation circuit in the crossovers to compensate for the step.

Twenty years on and still going strong.
The speakers were a closed two way cabinet. I built the speakers, carried out the testing, then designed the passive electronics (mounted externally) to provide the crossover and smoothen the response, whilst trying to keep the phase in control. Very challenging, but very satisfying result.

[Bill Fitzmaurice]

I think you're saying that conducting Ground Plane testing for subs like the T39, "baffle step" can be ignored because :-
1. We would only be interested in the results up to say, 200Hz.

That. The baffle step takes place where the baffle is a wavelength in dimension. At 200Hz that's 5.6 feet.

However, due to the size of the horn mouth, would you tilt the cabinet forward to aim the axis at the mic? Or wouldn't it matter due to the low frequencies being used?

Below the baffle step frequency radiation is omni-directional. That's why you get a half-space result with it on the ground.

Also, for subs only, would Near Field testing be an option? This would make testing easier, and help provide uniformity to individual builder's submitted results

With a large speaker the result will be corrupted by the presence of the cabinet itself. That's why measuring from at least two meters out is preferable.

[Nordo]

OK.

Can someone with a bit more knowledge than me, write a sticky (in plain English "for dummies") which can recommend testing procedure for our BFM projects?

e.g. For Subs - Ground Plane on a non-vibrating surface (concrete, bitumen, min area needed,etc); 2m mic distance; 5.66v; 1/6 smoothing; suggested testing equipment/software; etc
And perhaps suggested presentation format - axis limits, etc.

Then a separate protocol for Mains including detailed procedure for ground plane and half space testing and how to combine them, as well as the rest of the procedure covered in Sub testing.

Edit : I guess there should also be standard methods of testing individual drivers/piezos, as I came across a thread where Bill was trying to help someone who was having a problem with their piezos and filter, and Bill (and others) were spending a lot of time recommending varying testing procedures to try and narrow down the problem.
The whole saga could have been simplified with an instruction of "Go here and carry out the recommended test(s) and report back".

This shouldn't be left up to Bill, but obviously should have Bill's input and endorsement.
This would save Bill repeatedly quoting the same information over and over again.
I initially looked at topics throughout the forum that addressed "testing", but found nothing. I have since done a search for "ground plane", and now realise how many times Bill has commented on testing within other threads.

[67baja]

I would like to suggest also a quick and easy sticky for using a Driverack and other brand equivalents. I know there have been some threads explaining how to do it it in the past, but it would be nice to have a sticky so people can find it easy.

[Nordo]

I would like to suggest also a quick and easy sticky for using a Driverack and other brand equivalents. I know there have been some threads explaining how to do it it in the past, but it would be nice to have a sticky so people can find it easy.

Using rack equipment is fairly well covered in forums such as "Mixers and Processors", "EQ Settings" and "Pro Sound Setup". But I agree, Setup/Tips stickies for popular rack equipment within the BFM forum could be added into the sub-forums mentioned above.

My thread was aimed at having a standard form of testing of your finished project WITHOUT room or other equipment influences.
This should help in confirming if your finished project compares with the BFM design; or pinpointing what the problem is, if there is a problem.
Also such testing will let you know the deficiencies and limits of the gear you are introducing to the rest of your setup, which should help to give you a heads-up before you start to EQ.

[Grant Bunter]

I would like to suggest also a quick and easy sticky for using a Driverack and other brand equivalents. I know there have been some threads explaining how to do it it in the past, but it would be nice to have a sticky so people can find it easy.

The pretty comprehensive one I wrote for the DCX2496 is still on page 1 of "Mixers and Processors"...

[Nordo]

The pretty comprehensive one I wrote for the DCX2496 is still on page 1 of "Mixers and Processors"...

Thanks Grant.
I'll definitely be making use of your tutorial.
(Where's your tutorial on the DEQ2496?????)

[Grant Bunter]

(Where's your tutorial on the DEQ2496?????)

I don't know how to use a DEQ2496

edit: I have a few, but I don't know how to use them, I just make it up as I go along...