Home Theater Forum and Systems banner

1 - 15 of 15 Posts

·
Registered
Joined
·
7 Posts
Discussion Starter #1
I'm trying to measure a pair of new Goldwood drivers I received for Christmas that I plan to use for outdoor speakers. I'm following the instructions provided online.

The calibration worked just fine. I'm using a USB Soundblaster XFi soundcard switched to headphone out. I have a 100 ohm resistor from parts express hooked up as the sense resistor that measures 99.8 ohms on my multimeter. However, every measurement I try does not come out as anything viewable on the graph even though there's a thumbnail. I tried using a lead test and connected the two ends without a load. Here's the result of that test, which is nowhere close to flat based on the thumbnail. The numbers that come up when I click around in the graph window are in the Mohm range, which cannot be correct. Phase is all messed up too.

I am attaching measurements. Please help me figure out what is wrong.
 

Attachments

·
Premium Member
Joined
·
6,711 Posts
Looks like a measurement of an unconnected input. The input device and input have not been selected in the REW soundcard preferences, so REW will be getting the defaults for Windows which may not be the XFi - it is best to explicitly select the device and input you are using. There can also be an additional setting for Soundblaster cards to select the line input rather than the mic input.
 

·
Registered
Joined
·
43 Posts
Hello John and everybody,

No matter which sound card is used, the built-in Realtek or the external SB HD X-Fi that I also own, the results are disastrous and random.
Connections have been checked and rechecked, preferences have been deleted at each new measurement and a non inductive resistor is used.
The curves are less horrible with Asio4all but just as weird as with Java.:wink2:
I would like to know which sound card was used to obtain the results published in the REW Help.
Best regards.
 

·
Premium Member
Joined
·
6,711 Posts
The images in the help were generated with a cheap two channel soundcard and an old subwoofer driver. The soundcard doesn't need any special features, but the inputs and outputs must be on the same card. Try measuring a resistor to minimise the variables.
 

·
Registered
Joined
·
43 Posts
Well, I succeeded but not as expected.
I did a test on a newer PC (a 3 years old Shuttle DS61) with the jig wired according to the schematics of REW Help and it did not work.
I modified the jig connections by taking the left channel (white cable) as the main channel and it worked.
I tried this inverted jig with the laptop (a 7 years old Asus B43E Pro) and it didn't work.
The integrated sound cards are different on both devices.
I'll keep the modified jig.
Measurements were made with a 3.3 ohm Mox resistor measured at 3.275 ohms with a multimeter. The curve gives a value of 3.26 ohms. It's a very good result.
Now, in a few days, I'll measure my old Tannoys to get the FDD advanced TS parameters.
The advanced FDD settings interest me because I often use the Scanspeak Excel spreadsheet that can be downloaded from their site in the Toolbox tab.

The resolution of the problem must be done step by step according to the computer equipment used. It's not so obvious.

Thank you John for the work done since the beginning of REW.
 

Attachments

·
Premium Member
Joined
·
6,711 Posts
You can change which channel is treated as the main by changing the input channel selection in the REW soundcard preferences. The usual effect if the channels are swapped is to offset the impedance measurement by the value of the sense resistor.
 

·
Registered
Joined
·
43 Posts
I tried all possible combinations and the impedance sometimes varied from several kohms to a few mohms. Only the physical inversion of the right / left wiring of the jig allowed me to obtain a correct result with the Shuttle. The same jig and the same preferences on the Asus produced the curve of the attached file of the previous message.
Why these differences in behavior between machines? I don't know and that doesn't bother me.
You know, 35 years of work in computer science was enough for me. I do not try to understand the why of how when it exceeds my understanding. I try to spend the last quarter of my life in peace.
This is not a problem because now I will be able to compare the T&S parameters of REW, Arta, and Woofertester 2. Clio Pocket is way too expensive for me but this is the guarantee of peace of mind.
Best regards.
 

·
Registered
Joined
·
43 Posts
@bjaurelio

A burst of pride ! We shall never surrender.
You may know that the Frenchman Nicolas Boileau wrote in 1674 : ' Make haste slowly; do not be discouraged, but return to the work frequently. '
That's what I did and I did well. At this moment, the impedance measurement works on my laptop according to the REW scheme with the Creative X-Fi HD sound card. You have to fight with a gas plant as Windows 10 and it's not easy to overtake on it.

More details tomorrow. I used a 3R3 Mox resistor instead of a loudspeaker.
 

Attachments

·
Registered
Joined
·
7 Posts
Discussion Starter #10
I've been busy since the replies have come in. I plan to try getting measurements again this weekend. Life has been quite busy the past couple of weeks. I'll update on how it goes.
 

·
Registered
Joined
·
7 Posts
Discussion Starter #14
I made sure my soundblaster was both the output and input device. I confirmed this with changing level checks by adjusting the volume and microphone on the external soundcard in Windows. Both regular and inverted input measurements are still all over the place like my first post. I have no idea what's going on.
 

Attachments

·
Registered
Joined
·
43 Posts
Hello,
Can you post screenshots of Windows sound properties ?
This is not a failure of REW, rather a jig connection problem or Windows settings.
My bad curves were due to a ?&$xx! switch that I had installed for calibration.
 
1 - 15 of 15 Posts
Top