Rode NT1 okay for measuring?

[Mark DuBerry]

Hi groundie

Just saw your post after I put up my last one. I generated the cal file from the frequency response graph for the mic, which was hard to locate given the age of the mic. I made 3 attempts to get as close to the graph as possible (the 'b' in the file name is the 3rd attempt) so you may want to experiment a bit though I think the file is accurate enough. Here are both the graph and the cal file. The graph is in the bottom left corner of the 2nd page of the datasheet. I simply enlarged that part of the sheet and worked from there:

RODENT1.pdf

nt1b.cal

Good luck with it! Any probs opening or downloading those files, just post your email address and I'll send them to you.

Cheers,
Mark

[brucek]

Their stated fre. resp. is 50Hz - 21kHz. Could that be the reason for the problems from 60Hz down rather than modal issues in the room (I'm forever hopeful!)?

Yeah, I thought you had larger monitors and so with this new information, I would guess that the response is dropping off at 60Hz as a result of the speakers themselves, and the peak at 25Hz is a mode (although the room size doesn't really support that).

Would I be better to measure with each speaker separately?

No, you're doing the correct thing. Just be sure to never have any wacky soundfield or effects turned on. The mono signal from REW won't respond well.

I placed the mic as exactly as I could at that position, forming an equi. triangle with the monitors in order to simulate the setup when I am mixing etc. .........................
Due to the above, I am very restricted.

Yeah, seems correct and that's what I expected as far as movement..

Pretty ain't she?!

Looks pretty cool to me. Lots of treatment. I do think you have some wiggle room in your measurement / seating location in the forward-backward plane. Even if you weren't planning on moving it, it would be interesting to measure at least in different spots to reveal the differences. You need to work on the 100Hz problem.
Actually the next version of REW in beta test now has an RTA. This will allow you in real time to move the mic and watch the response dynamically on the screen (as opposed to measure once and observe after). It will be ideal for you.

brucek

[Mark DuBerry]

the peak at 25Hz is a mode (although the room size doesn't really support that).

Hmm.. I confess to not really understanding the ins and outs of modes, standing waves etc. Must read up some more. I wonder could that mode be due to some incorrect or over-treatment if it doesn't fit with the room size itself?

Just be sure to never have any wacky soundfield or effects turned on.

No, I kept it all as flat and neutral as possible, apart from any artifacts the mixer etc. itself might add.

Looks pretty cool to me. Lots of treatment.

Good to hear. It's not oppressive once you get used to it. Just cosy!

Even if you weren't planning on moving it, it would be interesting to measure at least in different spots to reveal the differences. You need to work on the 100Hz problem.

Yes. I think that is shaping up to be my main issue, which is a great thing to have learned. If I find a spot with a better response I may have to trade off in terms of the ideal theoretical position vs. the actual acoustic reality of the room. It might be difficult to choose between keeping the alleged sweet spot and losing that 100Hz dip.

It will be ideal for you.

Sounds cool. I presume that's not available yet?

Much appreciated brucek.

mark

[groundie]

mark
thanks so much for the files, i had no troubles reading them.
i will be measuring in the near future;
i was always interested in knowing what frequency response was
in the listening room, if there were any major issues.
as i stated elsewhere, after doing my initial measures using the
Radio Shack meter and setting a couple of filters per REW,
i went back to using no EQs at all.
so i'm curious if using the better mic will result in anything new;
i'm actually hoping not since i believe in minimum modfication to
the audio signal.

[Mark DuBerry]

You're welcome groundie. I agree with what you say about minimum modification. I guess it depends on what the room is used for. If it's for recording / mixing than I think it's always best to treat the room and leave the sound alone. That may not be so important for a home cinema or other type of purely listening room where some enhancement or modification of the sound itself may be desirable.

Good luck with it,
Mark

[Mark DuBerry]

Hi brucek. I took one of my speakers out into the garden today to test it along the lines you suggested. The garden is fairly small with quite a few bushes etc. but it is as neutral a space as I can test in, short of going up to the local park!

Here are the fre. resp. and waterfall graphs. They are quite revealing I think, though obviously I am only beginning to understand what I am looking at!

speaker test outside.jpg


speaker test outside waterfall.jpg

It looks pretty even to me with the predictable drop-offs at both ends. However, there is still a small peak at 25Hz and a small dip just after 100. I don't know how significant they are but it does look like the mic / speaker combination is contributing somewhat to those two areas. Obviously, the corresponding aspects of the room graph are much more pronounced so I know I do have a couple of room-related issues there. Still, I thought it might tell us something.

The other thing that strikes me is the huge dip and peak at 4.8kHz nd 6.5kHz respectively. What's going on there? That is pretty much the same pattern as appears in my room graph so again must be at least influenced by the mic / speaker in some way.

Just out of interest, here is the graph of a measurement I took earlier inside in the middle of our living room which is pretty big; same setup as outside. I did it again outside because I suspected the room was affecting things in some way. I noticed the huge dip at 55Hz and the lack of any peak at 25. Hmm..

speaker test 1.jpg

I realise every environment is going to yield somewhat different results but I thought this big room vs. garden might be interesting. The next thing I'll do is the backward / forward testing you suggested re. trying to nail that 100hz dip. Will report back. (Oh no, not more graphs! )

Cheers,
Mark

[brucek]

there is still a small peak at 25Hz and a small dip just after 100

huge dip and peak at 4.8kHz nd 6.5kHz respectively

Interesting. The garden graph shows that you're in good shape below 3KHz when a room isn't involved. You can use that graph as a standard against problems that show up in the studio.

The above 3KHz problems still need to be sorted out. It's hard to say if they're caused by the mic or the speaker. You'll need to determine this. It would be easy enough if you had another model speaker to test outside. If it was smooth above 3KHz, that would reveal the problem (assuming a sample size of two speakers was suffice).
Your speakers appear to be two-way, so the wonkiness can't likely be attributed to a crossover. It has to be the mic or the speaker. I would concentrate on that......

We haven't discussed it, but I'm assuming you've done a proper soundcard calibration and created the file and did a cable loopback test and got a perfect flat response with it before you began.........

brucek

[Mark DuBerry]

I'm assuming you've done a proper soundcard calibration and created the file and did a cable loopback test and got a perfect flat response with it before you began

Yes. I went through all the setup steps as carefully as I could. Here is the cal and the response from the loopback test:

soundcard.jpg

I left the sample rate at 48kHz.

It would be easy enough if you had another model speaker to test outside.

I do in fact still have my old monitors. They're not as good quality but they do go from 78Hz up to 20kHz so I will definitely test one of them outside to get a comparison.

Here is the properly scaled graph for the original measurment I took in the studio using the RS meter. I know it's not reliable in the upper frequencies but it does show a smoother response between 3 and 8kHz despite that 'typical rise' at 5.

room measured with rs meter.jpg

Could this indicate that the problem is more likely to do with the mic than the speaker? Rode did some subtle enhancing to the upper frequencies when designing this mic but the cal file I created should compensate for that I'd imagine. If the mic is causing this problem, maybe it's time to invest in an ECM8000? If I do, can I use it with my own pre-amp (it's a Metric Halo ULN-2, very high quality)?

the wonkiness can't likely be attributed to a crossover.

The crossover frequency of my Dynaudios is stated as 1500Hz with a crossover slope of 6 dB/oct.

I'll measure that other speaker next.

Thanks,
Mark

[brucek]

Could this indicate that the problem is more likely to do with the mic than the speaker?

I just don't trust the RS meter, but yeah the graph says it's the mic. But better to also try the NT1 with a different speaker. I guess the NT1 could be damaged somehow - sure is quite a dip at 5KHz.

I do, can I use it with my own pre-amp (it's a Metric Halo ULN-2, very high quality)?

Yeah, any standard mic preamp with phantom power will do. You can compensate for any non-linearity by including it in the soundcard cal loop when you make the file....

brucek

[Mark DuBerry]

You can compensate for any non-linearity by including it in the soundcard cal loop when you make the file

Sorry brucek.. Not sure what you mean by that. Would you mind explaining a bit? Do you mean to include the preamp in the line-out to line-in loop? I didn't do that with the NT1 setup.

Thanks,
Mark

[brucek]

Do you mean to include the preamp in the line-out to line-in loop?

Yep. Since the preamp is part of the microphones response, then to compensate for its response imperfections, it is a good idea to offset it by including it in the soundcard cals correction file. You have to use the line-in and line-out of the preamp and do a bit of fussing with the levels to get the soundcard cal to setup correctly, but it's worth it if it has a poor upper or lower frequency response. Once you've created the soundcard cal file, you remove it from the loop and use it as normal.

brucek

[Mark DuBerry]

Okay that makes sense. Sounds like it would be a good idea to re-do the room measurements with the pre-amp included this way, even with the NT1. The other thing is I did all measurements with both the RS meter and the mic pointing directly (horizontally) towards the speakers. I have just read some other posts recommending that the mic/meter should point upwards toward the ceiling. I'm not sure what difference that would make but I'll do some measurements that way and see.

I tested one of my other monitors today and got pretty much exactly the same result for the area above 3kHz. So it looks like the problems up there are due to something to do with the mic.

I guess the NT1 could be damaged somehow

I hope not! I've never dropped or knocked it and I've always kept it in its hard protective case. Also, I haven't used it all that much and love the sound I get from it; I've certainly never noticed any problems with it. So unless it was damaged in the shop or was faulty to begin with.. I'm going to call Rode and run those test results by them and see what they say. Perhaps it could be a characteristic rather than a fault, or maybe that wonkiness is just too bad for that..

Do you think it would be worthwhile going out and buying a Behringer to re-do the measuring? I'm going to hang on to the NT1 unless it becomes clear it is faulty. Also, is it not the lower frequencies that are the most relevant in terms of the room treatment and I already know that the 100Hz dip is a problem there? I suppose the above 3kHz problems could bring that into question, i.e. is my mic really reliable across the rest of the spectrum?

I am conscious this has been going on a while brucek. I do really appreciate your on-going help with this and I hope you don't mind all the posts and replies. I think I'm getting there, once these anomalies can be sorted out. When I'm done, perhaps there is a way that I can show my appreciation (a donation of some kind or whatever?)

Thanks again,
Mark

[JohnM]

It can be quite tricky measuring a typical multiple drive unit loudspeaker as there will usually be dips in the response where the outputs from the various drivers overlap but are out of phase with one another, caused by the path length differences from the drive units' acoustic centres to your meaurement mic position and the phase shifts through the crossover network. With those kinds of cancellations you typically find that moving the mic up or down shifts the frequencies of the dips.

[Mark DuBerry]

Yes that makes sense. However, the problems I discovered occurred with almost identical dips and peaks with two different makes of speaker on two separate occasions. That being said, I did have the mic positioned in exactly the same way with both speakers, i.e. pointing towards the tweeter. So I will definitely try some tests with the mic at different orientations as you suggest and see if those dips and peaks shift at all.

Thank you for your input John,
Mark

Edit: I just spoke to someone in Rode who said that the behaviour above 3kHz was most likely caused by a faulty capsule and that it needs to be replaced. The good news is the mic is still under warranty so I should be able to get it repaired for free. I never would have thought there was a fault but hopefully that will explain the wonkiness. Perhaps it was also behind the 100Hz problem(?)

Only one way to find out: I think I'll pick up an ECM8000 tomorrow and see what results that yields. Should be interesting..

[brucek]

I think I'll pick up an ECM8000 tomorrow and see what results that yields. Should be interesting..

Yeah it will be interesting. It's actually a pretty good mic for the price. We have the cal file on the download page...

brucek

[Mark DuBerry]

Hi brucek

Well I bought the ECM8000 today and have been experimenting with it. Overall, it seems to produce a more even graph across the whole spectrum. There is no real difference between the graphs with the existing soundcard.cal file and the one made after including the pre-amp in the loop so I think the pre-amp is doing a good job.

Here are the fre. resp. and waterfall graphs for the measurement at the 38% listening position:


room measured with ecm.jpg


ecm waterfall.jpg


The fre. resp. of the speakers seem to be represented accurately enough, i.e. 50Hz-20kHz. The 100Hz dip is back along with one at 350. The ones at 700Hz, 5 and 10kHz are also striking, though perhaps less severe(?)

Moving the mic backwards and forwards as you suggested in an earlier post makes some slight difference to those dips but they are still there. Here is the graph for the measurement a foot back from the listening position, i.e. further away from the speakers. It shows the biggest change.

room measured with ecm back.jpg

I can try a few more variations, e.g. moving the speakers further apart etc but as I said before, I am limited as to how much flexibility I have re. the final setup. However, I'll do whatever's necessary to deal with any problem frequencies.

Another option I've just thought of is to place the listening position at 38% from the back wall and to set up the speakers accordingly. That would give me a bigger triangle and may 'spread' the sound some more. I may do that and measure the response again. Could be interesting..

I haven’t tested the speaker outside with the new mic as yet. Is this necessary?

I'd be grateful for any comments or thoughts.

Thanks,
Mark

[brucek]

so I think the pre-amp is doing a good job.

If you make a soundcard cal with just a cable, then put the preamp (or any line level device) in the loop and do a measure, it will reveal its frequency response. It's a useful trick to try on all sorts of line level devices to see what their bottom and top ends are. Your preamp is probably so good it simply doesn't need to be in the cal file. Anyway, that's another useful thing REW can do - taking frequency response measures of electronic devices.

Is this necessary?

I think not. You now have a reliable method of measuring your response, so the only job left is to get the best response in your room. I honestly have no suggestions. This is where acoustic guys in our other forum sections may be able to help. It's getting out of my league. You know what you have to do now anyway.

brucek

[Mark DuBerry]

You know what you have to do now anyway.

Yes I do, probably enough for my purposes anyway. That is mostly thanks to your help brucek so once again, many thanks for that.

I will look for some advice on what to do next from the acoustics experts.

All the best,
Mark