Re: Re: Re: my earlier readings...
Actually, I initially started my experiments with any VU levels of 17-25K (because the Unofficial SW Manual says so), then did them at 8-10K, and still continued to receive feedback on this thread that I need to lower the levels even more. So make up your mind. 😀soongsc said:
Thanks! Will try. I need something very simple, which will probably have to be battery-powered to reduce the messing around with xfmr-diodes-caps-regulators.
Let me recall what I've said on this topic on my earlier posts:
- My sound card is the Creative Soundblaster Digital Music USB
- My sound card operates till 48KHz. I tick-mark a check-box in SW which makes it check for 48KHz, and SW tells me I can go up to 48KHz.
- When I try to do a pulse response with SW and use a sampling rate above 24kHz, I get no pulse in the graph. At 24kHz, I get clean spikes (which you've seen). This is not affected by sample size: I get clean spikes at sample sizes of 64k to 256k
- When I try to do gated SPL measurements at 48kHz, it gives me some graph, sort of similar in appearance to what I get at 24kHz. (You can see both graphs in my earlier posts).
I wanted an actual design, something which would also supply phantom power to the mic. I know NE5532 would be fine for most opamp-based designs... the reason I mentioned the chip is because I didn't want to go discrete. I'm looking at the Eric Wallin design.
I guess I can use the RS SPL meter as a combined mic + mic-preamp, right? In that case, I can feed the signal into my line-input of my sound card, and I won't suffer the high-end roll-off that my sound card's mic input seems to enforce. Great. I can postpone making my mic preamp, in that case.tcpip said:
can't think of anything else for debugging, but it could possibly be that the mic input doesn't support sampling rates above 24Khz???? would make sense since voice over 12Khz is a non-issue.
soongsc did point to vikashes kit and also the url for the original design, it can be made on ic experimenter board if you don't want to outlay for the kit, I've done it on experimenter board and on Vikash's pcb 🙂 it works well, then only thing I would like is some more gain, but that is something I'll have to experiment with 🙂. Vikash I think has the pc boards on special due to a problem with the initial layout (easy to fix), a board will make your life a whole lot easier 🙂
There is also this one --> http://sound.westhost.com/project13.htm but you will need to build a power supply for it so it is not so suitable if you want to run off batteries.
Tony.
soongsc did point to vikashes kit and also the url for the original design, it can be made on ic experimenter board if you don't want to outlay for the kit, I've done it on experimenter board and on Vikash's pcb 🙂 it works well, then only thing I would like is some more gain, but that is something I'll have to experiment with 🙂. Vikash I think has the pc boards on special due to a problem with the initial layout (easy to fix), a board will make your life a whole lot easier 🙂
There is also this one --> http://sound.westhost.com/project13.htm but you will need to build a power supply for it so it is not so suitable if you want to run off batteries.
Tony.
See here (PDF, pages 3-5 of the document). It does. It also has one more feature which you taught me about: it has a ready-made threaded socket for mounting on a photo tripod.wintermute said:
Now I need to look somewhere for a calibration file for this mic. (I have read that the RS meter's mic SPL response is not very flat, but the piece-to-piece consistency in its SPL curve is very tight, thus allowing a calibration file for one of these meters to be used for correcting the non-linearities of any other sample of the same model.)
Re: Re: Re: Re: my earlier readings...
These values are just starting points. In reality, you want to use the highest signal without staturation and get good results. The larger the signal, the better the signal resolution when digitized.
If you use 48K and can't see the impulse, you need to change the scale of the chart to minus time to see if it's there, or increase the plus side time scale to see if its there. USB devices have different latencies for different settings, so just lok forward or aft untill you find it. It's just playing hide and seek with you.😀
Just w word about SPL meters. Most of them are A or B weighed, thus cutting the High and Low frequencies. I beleived the C weighed is full pass, otherwise you will need to midify the SPL meter.
tcpip said:
These values are just starting points. In reality, you want to use the highest signal without staturation and get good results. The larger the signal, the better the signal resolution when digitized.
tcpip said:
If you use 48K and can't see the impulse, you need to change the scale of the chart to minus time to see if it's there, or increase the plus side time scale to see if its there. USB devices have different latencies for different settings, so just lok forward or aft untill you find it. It's just playing hide and seek with you.😀
tcpip said:
Just w word about SPL meters. Most of them are A or B weighed, thus cutting the High and Low frequencies. I beleived the C weighed is full pass, otherwise you will need to midify the SPL meter.
Re: Re: Re: Re: Re: my earlier readings...
How do I understand at what point the mic is saturating? I presume you've seen the earlier comments made by wintermute and others about "furry" SPL graphs?soongsc said:
Good point. My device is working currently with 20ms of latency setting in the Debug pane of SW. I'll increase and reduce this latency and see if I get anything. I've increased the X-axis of the graph right till the end to see if any peak is there anywhere on the extreme right. It's not there. Some more experimentation will be needed, I can see. Tonight will be so exciting... oooh! 😀
This one actually defaults to C weighted. So that part should be okay. If you see the manual (I've given a link to it in my earlier post), you'll see all these options and settings.
Hi TCP/IP what soongsc was saying was to change the X axis in the pulse response graph make it something like -100ms to 100 ms and look for the peak 🙂 you can then adjust the latency based on information rather than trial and error 🙂
on the saturation, I'd say turn it up until it starts getting fury then take it back a little, this is assuming your ears aren't bleeding 😉 I wear hearing protection even with 8K levels I hate to think how loud (with my preamp) it would be to get 15K levels!!!
Tony.
on the saturation, I'd say turn it up until it starts getting fury then take it back a little, this is assuming your ears aren't bleeding 😉 I wear hearing protection even with 8K levels I hate to think how loud (with my preamp) it would be to get 15K levels!!!
Tony.
Will do. In fact, I'm pretty sure something is there somewhere because the VU meter tells me the recorded signal strength is, say, 5k, but on the graph, using auto-minimax on the X1 axis, I see only +/-6. (Yes, not 6K, but 6). So I'm sure the peaks are there... I have to look more aggressively.wintermute said:
All I can think of is that your mic preamp has much lower gain than mine. My mic goes beyond 20K levels long before the sound reaches anything near hard-to-tolerate levels for my ears. And no, I'm not (yet) deaf. 🙂
possibly one reason is that I run the line in on 0db gain on the sound card, if I bumped that up I'm sure I'd get higher levels, but I figure I'd rather not do any more amplification than necessary 🙂 this is the setting that gives the best results in RMAA so I stick with it 🙂
Tony.
Tony.
Since you're talking about line-in in the context of a mic, are you using an external mic preamp? Which one?wintermute said:
Re: Re: Re: Re: Re: Re: my earlier readings...
Yes, teh furry graphs are a sign.tcpip said:
Actually I think Eric Wallin did a mod on the Radio Shack SPL meter, you might want to see if it's similar to the one you posted.tcpip said:
I think you have been taking some cues of soongsc 😉 I'm using the wallin preamp II (with one of Vikash's PCB's) I'm sure I mentioned it earlier in the thread 😉
http://www.madaboutsound.com/prod_det.asp?product_id=6
Tony.
http://www.madaboutsound.com/prod_det.asp?product_id=6
Tony.
wintermute said:
Yep.
Also the volume to turn up would be the mic input volume in the Windows volume control. It is quite unlikely that this will saturate if this control is just the software portion. If it controls the hardware section of the sound card, then there might be a point where saturation occurs.
Re: Re: Re: Re: Re: Re: Re: my earlier readings...
I spent one more night trying out experiments, and have been able to conclude the following:
Therefore, at least one thing I'm now sure of: I don't need to look for an alternate mic preamp: my mic input is already doing clean 10KHz recordings.
I spent one more night trying out experiments, and have been able to conclude the following:
- There's no rolloff at the upper freq in the mic or the mic preamp. I did a loop record of sine waves at 8KHz, 10KHz, 4KHz and 1KHz. With the same volume setting of my amp, I got the same amplitude in the recorded signal on the computer, and I got clean sine waves, no clipping, no sharp sawtooth edges. So I can't blame the hardware or the audio device drivers any more.
- I am still failing to get any pulse at 44.1KHz. (I'm not sure I want to try at 48KHz, but I'm sure the card hardware supports 44.1KHz correctly.) I get sine waves cleanly at that frequency, but I don't get square waves, and when I do "Measure->pulse response", I don't see any pulse. The VU meter shows me high amplitude signals are being received and recorded, but I just can't see it in the graph even after I increase the X-axis from -100msec to +200msec.
- I've learned about the mic input volume control in Windows. I've used sine wave loop recordings to set the gains now, and I am pretty sure there is no "furry" sine wave or clipping of any kind.
Therefore, at least one thing I'm now sure of: I don't need to look for an alternate mic preamp: my mic input is already doing clean 10KHz recordings.
Right now, I've not yet reached the stage where I'll need to look into the finer points of the RS SPL meter.soongsc said:
Started a new thread with this stuff
I've tested the patience of the two of you enough, and I wanted to see if there are others in this forum who may want to chip in. I don't think anyone other than the two of you are reading this thread, and I can't believe my problems are all that unique that I need to struggle so much.
Let's see what other responses we get. 😕
I've tested the patience of the two of you enough, and I wanted to see if there are others in this forum who may want to chip in. I don't think anyone other than the two of you are reading this thread, and I can't believe my problems are all that unique that I need to struggle so much.
Let's see what other responses we get. 😕
is one more response ok?? 😉
I just thought of something. do you have the use pre-emphasis box ticked??? this will cause hf rollof if you do (especially if you haven't done an amp calibration and don't have the ref channel. If it is ticked turn it off!!!
with the pulse, did you try setting the x axis to start at -200ms?? It must be in there somewhere if the vu meters are showing a reading 😉
If no one chimes in, it might be time to start a new thread with a different title 😉
Tony.
I just thought of something. do you have the use pre-emphasis box ticked??? this will cause hf rollof if you do (especially if you haven't done an amp calibration and don't have the ref channel. If it is ticked turn it off!!!
with the pulse, did you try setting the x axis to start at -200ms?? It must be in there somewhere if the vu meters are showing a reading 😉
If no one chimes in, it might be time to start a new thread with a different title 😉
Tony.
Re: Started a new thread with this stuff
What kind of mic do you have? How long is the capble from the mic to you sound card mic input? Most mics start decaying around 5K. At one time I did get some that did that. Very frustrating when you think the mic meets spec and it doesn't. The MadAboutSound mic caps are quite up to spec. Also, if the mic cap cables are too long, you will also see some effect due to capacitance.
tcpip said:
What kind of mic do you have? How long is the capble from the mic to you sound card mic input? Most mics start decaying around 5K. At one time I did get some that did that. Very frustrating when you think the mic meets spec and it doesn't. The MadAboutSound mic caps are quite up to spec. Also, if the mic cap cables are too long, you will also see some effect due to capacitance.
Re: Re: Started a new thread with this stuff
Hey, any number are welcome. Always. 🙂wintermute said:
No, no way. I know this setting, and I've never switched it on. Moreover, see what I've posted about 8KHz and 10KHz.... I'm getting absolutely uniform amplitude at all those frequencies. From 1KHz to 10KHz, the same amp volume gets me the same recorded amplitude.
No, I didn't try 200ms, but I did try -100ms. I am now desperate enough to take an exported dump of the file and look at the actual text values to see where the high readings go. In software development, we say, "When all else fails, look at the source." 🙂
A Panasonic WM61, purchased from Digikey.soongsc said:
About two metres, I think.
Yes, but then how do I get clean sine waves at 10KHz?
- Status
- Not open for further replies.