I haven't taken the photo yet. Been enjoying my Friday night. Will be away most of the day tomorrow as well. Perhaps Sunday.
But yes, it is pretty simple when it is done. I'm just new at reading circuit diagrams, so it puzzled me for a bit initially.
Since we're on the subject, there is also a way to determine T/S parameters using an oscilloscope. The process is outlined in the Audio Engineer's Handbook. I'll see if I can find the details later, but obviously that's only of use if you have an oscilloscope lying around. 🙂
I actually used to have an O-scope lying around before I got into DIY but sold it. Wish I had one now, comes in handy for looking at input vs output for amps and seeing rms power and distortion.
An O-scope would be nice to have, but for TSP this method is pretty easy.
Here is how I made my "TSP Cable":
Just plug the jacks into your sound card appropriately, and the spade connectors to the driver...
...flip the driver over and start testing:
Here is how I made my "TSP Cable":
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Just plug the jacks into your sound card appropriately, and the spade connectors to the driver...
An externally hosted image should be here but it was not working when we last tested it.
...flip the driver over and start testing:
An externally hosted image should be here but it was not working when we last tested it.
whats the advantage?
I fail to see an advantage. What software do you use to get the final result? How do you calibrate it? And it all kinda depends on your soundcard quality. I still find the old Weems type method satisfactory, and dependable. As tuxedocivic says added mass Vas method is horrendous.
I fail to see an advantage. What software do you use to get the final result? How do you calibrate it? And it all kinda depends on your soundcard quality. I still find the old Weems type method satisfactory, and dependable. As tuxedocivic says added mass Vas method is horrendous.
The advantage is that it makes TSP measurement accessible to anyone.
Go back and actually read the first post to find the answer.
You don't need to. It simultaneously sends the left channel output direct to the left channel input and to the right channel input via the resistor and driver. It then does realtime analysis to determine the delta in the signals received on the two channels and computes the impedance curve.
Nearly any relatively modern "HD" onboard audio is perfectly sufficient.
Fantastic. Continue to use that method, rather than thread crapping here. People have different needs, and plenty of people find this particular method both easy and effective.
Last edited:
Perhaps I haven't been clear. The method I have reported on in the first post is easy, has very little hardware requirements, and is effective. Did I inadvertently indicate that I was looking for a more complicated method? Did someone else?
I was hoping that you might appreciate the extra info/links i provided ? Often good info/knowledge can be obtained by reading more ! I like it when other members offer extra info/links etc ! Many times i've discovered/learnt interesting things, even though i knew most of it !
Also if you noticed,i included et al 😉
Sorry for getting a bit snarky. mondogenerator's response put me off a bit and I reacted to your post simultaneously.
woah! I didnt post to cause offence, check your attitude. It just doesnt seem any easier to me at all (and i have read the thread). The linked article points to the need to calibrate, and implies the current method is simplest and I agree. I merely thought id missed a piece of info and perhaps the software performed a sweep for you (which would make life easier). Having a valid opinion certainly isnt 'thread crapping'. And FWIW the extra info is always interesting, and it is why i came and read the thread in the first place, and NOT to cause offence to those clearly suffering with localised paranoia.
I come here to learn, same as everybody else...
I come here to learn, same as everybody else...
Last edited:
Your approach seems to be hindering that process.
I would suggest you (again) to go back and read the first post. The logical, effective approach would be to:
1) Observe the name of the software mentioned there.
2) Copy/paste that into google, then click the most obvious link.
3) Briefly familiarize yourself with the features of the software listed on the main page.
4) Download the PDF file, then open it and find the link on page one which takes you directly to the section on measuring TSP.
5) Read that section.
I am confident (or perhaps hopeful is a better word) that if you had done that to begin with, your first post would not have been necessary.
I reacted the way I did because instead of taking the time to actually read my first post and do a bit of investigation, you submitted a post asking questions which either have already been answered or can easily be answered by a few clicks and a few seconds of reading. Not only this, but before knowing anything about the software and/or method being discussed, you stated that the method you know is better.
That is thread crapping, and I won't apologize for my reaction to it.
OK, let's get back on track here. Misunderstandings are just that so please, let's move on.
Thank you.
Cal,
I've tested those 12" woofers that I am using as sub drivers in my "console sub".
I don't have the measurements with me right now, but I remember you were interested so I'll post them later.
The drivers tested significantly differently from each other, but I am not really too surprised considering their age and condition (in fact, with that in mind maybe they tested surprisingly close - I'll let you be the judge of that). It was my first time doing TSP measurements, so I made sure to run the tests repeatedly, until I saw a pattern of similar results. For one driver, it only required 3 runs. For the other, it took 5. I then made a speadsheet and averaged all the results, which I plugged into WinISD to get an idea how to tune the console enclosures.
Turns out there's not a whole lot more that I could do to get more LF extension out of the drivers in the given volume, but it did allow me to at least optimize the tuning of each enclosure for the respective drivers. I was able to get about 2db more at around 45Hz (simulated) from one of the cabinets and I smoothed the response of the other a bit. Since they both port to the bottom of the console and push out the bottom/back and up the rear wall, I wasn't too worried about getting them "matched".
Data later...
I've tested those 12" woofers that I am using as sub drivers in my "console sub".
I don't have the measurements with me right now, but I remember you were interested so I'll post them later.
The drivers tested significantly differently from each other, but I am not really too surprised considering their age and condition (in fact, with that in mind maybe they tested surprisingly close - I'll let you be the judge of that). It was my first time doing TSP measurements, so I made sure to run the tests repeatedly, until I saw a pattern of similar results. For one driver, it only required 3 runs. For the other, it took 5. I then made a speadsheet and averaged all the results, which I plugged into WinISD to get an idea how to tune the console enclosures.
Turns out there's not a whole lot more that I could do to get more LF extension out of the drivers in the given volume, but it did allow me to at least optimize the tuning of each enclosure for the respective drivers. I was able to get about 2db more at around 45Hz (simulated) from one of the cabinets and I smoothed the response of the other a bit. Since they both port to the bottom of the console and push out the bottom/back and up the rear wall, I wasn't too worried about getting them "matched".
Data later...
Last edited:
After having read the PDF properly...(or better at least)
Hi again.
Having read the TSP part of the pdf manual. I would agree with ZeroDs' first comment, that 10R series resistance is a bit low. (I blew the line out on my last card, with a much higher load impedance than 10R.....) My current soundcard is PCIe and a lot better, it states a 20-20k response, but -3dB is actually at ~40 Hz. Whether that would have an effect on measurements....well I dunno. Its possible though.
I would worry about loading my soundcard a little too much with only 10R. Software looks reasonable though, especially given the option of Vas by Vb option for calculating Vas. Ive been meaning to get a WT2/3 but its just a little too dear for what it is (probably very similar, no idea about the burden resistance though)
Hi again.
Having read the TSP part of the pdf manual. I would agree with ZeroDs' first comment, that 10R series resistance is a bit low. (I blew the line out on my last card, with a much higher load impedance than 10R.....) My current soundcard is PCIe and a lot better, it states a 20-20k response, but -3dB is actually at ~40 Hz. Whether that would have an effect on measurements....well I dunno. Its possible though.
I would worry about loading my soundcard a little too much with only 10R. Software looks reasonable though, especially given the option of Vas by Vb option for calculating Vas. Ive been meaning to get a WT2/3 but its just a little too dear for what it is (probably very similar, no idea about the burden resistance though)
Last edited:
Ulrich does suggest a voltage reduction network in there to avoid frying the sound card.
I simply started with the input gain set extremely low and the output set extremely low, then played with each until I got good results and no magic smoke.
But that's the way I go about things. Fry the onboard sound of a $50 MB? No big deal. I've done much worse. 😀
I simply started with the input gain set extremely low and the output set extremely low, then played with each until I got good results and no magic smoke.
But that's the way I go about things. Fry the onboard sound of a $50 MB? No big deal. I've done much worse. 😀
Ah. I didnt see that bit. I thought it was a bit strange....
Sensible precaution 😀 Obviously a higher signal level and a higher burden R would be better, less noise affecting measurements. (Stating the obvious, I know. Habits die hard)
Ha! my motherboard was £200 5 years ago lol.
Hang on... You're using it with with on board sound? You're lucky you have a 'proper' monitor channel. My last on-board sound didn't.
- Status
- Not open for further replies.