Yes. Welcome to the wonderful world of modes.
Speaker placement and dB differences
So I’m pretty anal about speaker placement. I use tape measures, along with laser lights and levels to assure proper and equal placement from walls. I was using my new Omni mic and testing software to run some white noise frequency graphs vs db range to see where I stand. After overlaying the R/L graphs I noticed that there is about a -1db difference in the right channel generally across the entire frequency range and about -5 dB at about 250 hz.
Also my room is 12’ x 13’.
I remeasured my speaker distance from the seating position wall and I noticed the left speaker was 96” from the front the chair and the right speaker was 98.5” from the chair. By my calcs that would only account for a .22 dB difference. Not enough to account for the 1 dB drop in the right speaker and certainly not the -5 dB at 250 hz.
Question is, is a 1 dB difference between two speakers normal. My speakers are Thiel CS 2.4’s and I’m using an older ARC tube amp - which can certainly account for 1db.
the bigger questIon is can a 2.5” difference in room placement cause a 5 dB difference at 250 hz?
Also my room is 12’ x 13’.
I remeasured my speaker distance from the seating position wall and I noticed the left speaker was 96” from the front the chair and the right speaker was 98.5” from the chair. By my calcs that would only account for a .22 dB difference. Not enough to account for the 1 dB drop in the right speaker and certainly not the -5 dB at 250 hz.
Question is, is a 1 dB difference between two speakers normal. My speakers are Thiel CS 2.4’s and I’m using an older ARC tube amp - which can certainly account for 1db.
the bigger questIon is can a 2.5” difference in room placement cause a 5 dB difference at 250 hz?
15 responses Add your response
If you swap left to right does it stay the same? That should narrow it down to JUST speakers. If it moves left to right it’s before the speaker including IC. Check all connections, you know. BUT swap the cable at the amp first. If it moves left to right. It’s the Amp back. If not swap at the speaker end, make sure the cable is not the culprit... But just so you know, a lot of OX are way off 10% in not uncommon. That is why you will see different values, Left to Right in matched speakers, (rare) not production speakers. 10-15,000.00 usually aren’t matched just so you know... there is a big difference. Building an XO is pretty straight forward, matching.....Left to right is not..You really want to nit pic.. LOL you sure can... How does it sound? Amazing that it doesn't sound so different... 5 db at 250 at what db? It probably won't... Regards... |
You would not have had that problem had you followed my advice which is to use the speaker placement track on the XLO Test CD or similar TEST CD or LP. Trying to find the absolute best locations for speakers by lasers, tape measures, trial and error are bound to fail. The speaker placement track allows you to find the absolute best locations for ANY speaker in ANY room and with ANY level of room treatments applied. All other techniques are like trying to solve x simultaneous equations in x + n unknowns, the best they can do is find local maximums. Hint: most speakers are way too far apart. Yes, I know what you’re thinking, but that’s how you get a wider soundstage, right? |
Was too tired when I answered this. If you are using OmniMic, use the swept signal with gating/blended results. This will help you see if the problem is consistently across all frequencies (which would imply an equipment issue) or if it's in the bass for instance, which would imply more room issues...or narrow across 1 driver's range. |
Unless you are using time-gated curves, windowed to avoid room reflections, what you are picking up is speakers + room. In that case, unless your room is acoustically symmetrical across the spectrum (which never happens), some measured difference between the two speakers is virtually inevitable. And also totally okay. What matters is that the first-arrival sound be the same from both speakers, and you can't measure that without time gating, and even then you can't measure it down very low in frequency. In particular, that 5 dB difference at 250 Hz is not the result of "a 2.5 inch difference in room placement". Your two speakers are much farther apart than that. The room starts to dominate below about 500 Hz, and from there on down each speaker interacts with the room somewhat differently. None of this PRECLUDES the possibility of other factors being the cause of the issues you see, but they could just be normal room interactions. Duke |
Sounds like you measured things the way I did. Call Soundings in Denver. They specialize in Master Setting. This is the best way to set things up. After doing so, you will enjoy the sound in different sitting positions in the room. Nice when having people over to listen over a few glasses of wine. The more you drink, the better the sound. |
I previously owned Thiel 2 2 speakers. These speakers are time and phase coherent and thus needs proper placement and seating to get the best out of them. I now own Thiel 3.6 and you can refer to one of my previous posts for details here:https://forum.audiogon.com/discussions/how-far-away-from-your-speakers-are-you/post?postid=1917078#1... Thiel speakers use first-order cross over networks and drivers are vertically aligned so that direct sound waves from the three drivers arrive at listening position at the same time. Thiel 3.6 user manual says chair/sweet spot should be at least 8’ from the speakers for this to happen.I would assume 2.4s would require similar set up. Once you move your speakers 2 ft from front wall, your seating position is almost close to the back wall assuming you are seated at least 8 ft from the speakers. So you are going to get back wall reflections in addition to the side-wall reflections. Depending on room height 8 to 10 ft, there may be even reflections from the top. I suggest you look for room treatment. Another option is a toe-in if you have not done already. I suggest you keep both speakers same distance from your seating position. Seating height is another factor to consider. Your Thiel 2.4 manual should give you a pretty good description of how to set this up. These speakers need room to breath and muscle amplifiers to drive to their capacity. My final recommendation is to try a good solid state amp with plenty of current capacity at 4 ohms. From Stereophile review: its impedance plot (fig.1) indicates that the CS2.4 demands a lot of current from amplifiers. Not only does its impedance drop to 2.73 ohms at 600Hz, but it stays significantly below 4 ohms from 100Hz to 50kHz, and there is a difficult combination of 4.5 ohms magnitude and –45° electrical phase angle at 80Hz. Thiel CS2.4 owners should make sure they have a good 4 ohm–rated amplifier to drive this speaker. that the CS2.4 demands a lot of current from amplifiers. Not only does its impedance drop to 2.73 ohms at 600Hz, but it stays significantly below 4 ohms from 100Hz to 50kHz, and there is a difficult combination of 4.5 ohms magnitude and –45° electrical phase angle at 80Hz. Thiel CS2.4 owners should make sure they have a good 4 ohm–rated amplifier to drive this speaker. Good luck! |
sorry to respond late to some of the comments. thanks for all of them btw. The speaker are now within a +/-1/4" from front and side walls from one another and they are toed in a bit so that their axis crosses behind my head a foot or two. I have sound treatments on all walls and even the ceiling. Large wall of diffusers/absorbers behind my head and on side walls for reflections. I have diffusers on front wall between speakers and on ceiling where main reflections would come from as well. I also have a few large GIK wall panel absorbers as on three walls. The floor is a large plush rug. As for amplification, I had a Parasound A21 driving them when I had the speakers in a larger room, plenty of power, but the room had a bass suck out right where I sat that I couldn't do anything about. I had to use a REL sub so compensate, and it work a bit, but the sound was still anemic in the mid bass and mid base region. Now I use an ARC CL 60 tube amp and it sounds clean and full sounding with PLENTY of bass and power in the smaller room. I feel zero need for a sub. Based on my Omni mic im getting very good bass down to 32Hz. I do realize the Thiels require time alignment and a min. 8' from listener position to integrate and I have just over 8' now. Just as a point of reference when i set speakers up I try all types of placement and then once I'm satisfied about a general location I tweak further until the last inch or so and then fine tune to within a quarter inch. It has been my experience with these speakers that they tend to sound best further apart than most with only little toe in. As for the suggestion for time gated cures, I have no idea how I would generate such a signal. I have an audio analysis app on my I pad and the calibrated mic is attached to to receive sound from the source - in this case a test CD from my HSU sub I once used, but nothing with a time gate on it. Not even sure what a time gate is really, though I have a thought based on the name. |
Time gating is a technique for measuring the speaker's output without including any room reflections. A specialized measurement system is required. The microphone is switched on for just long enough to capture the direct sound from the speaker and then switched off before the first reflection arrives. So all room reflections are excluded from the data. But the ability to collect good data at low frequencies is limited by the shortest reflection path length in the room (typically the floor bounce), because the microphone is off before it can adequately capture longer wavelengths. Time gating is one of the techniques John Atkinson of Stereophile uses when generating "quasi-anechoic" frequency response measurements. He uses a different technique called "close micing" at low frequencies, which is where the microphone is placed very close to the woofer(s) and then very close to the port(s) and those curves are combined with appropriate weighting. The result is then spliced with the time-gated curve. Duke |