71Hz, dagnabbit!

Brian Cheney · « **on:** 18 Oct 2006, 04:50 pm »

71Hz is the vertical room mode for an 8ft ceiling, the most common in residences. You can't address it by moving your speakers within the room (although getting them as far as possible from room boundaries will tame many late arrivals) or with corner traps. Even in my fully treated 13x31'ft LEDE environment this mode is a lot more audible than I thought.

For years John Casler has urged me to add bass control/active amplification, and as it turns out, the PE sub amp (1000W rms in 4 Ohms) can be modded for our 280Hz hinge point and, best of all, has a single-band parametric EQ. The "parametric" part is important as you want to zone in on a narrow range, preferably a single frequency. I set the EQ frequency control for 71Hz and adjusted level and Q (I settled on .25) pots by ear. In 20 minutes I had much better sound top-to-bottom.

As an added bonus, the big bass amp took a burden off my Son of Ampzilla monoblocks and dynamics improved considerably.

Samash sells an analog parametric EQ for $149 (with tube output yet) and I urge all VMPS'ers to investigate. Digital EQ's like the RDES should also work well.

Next: DEQX, or who needs a speaker designer?

Housteau · « **Reply #1 on:** 18 Oct 2006, 07:22 pm »

Thanks for posting this Brian. Your statements may have corrected a misunderstanding I have had on passive crossovers. If I understand you correctly, your SOA amps are amplifying the full range signal and the bass is passively being filtered out at the speakers crossover, so that duty can be handles by your built-in bass amp. My question is, when a signal is being filtered out (to ground), does that part of the signal still place a load on the SOA amps? My thoughts have been that it is still a load, but one of converted energy to heat instead of one to drive a speaker. Your statement that: "...the big bass amp took a burden off my Son of Ampzilla monoblocks and dynamics improved considerably.", would indicate that I am wrong about the load being present, or that there is still a load, but it is less than it would be if it was driving the entire speaker system. Which one is it?

Brian Cheney · « **Reply #2 on:** 18 Oct 2006, 10:13 pm »

The voltage waveform delivered to the SOA in this circumstance is unchanged. However, the amp sees a high impedance below about 280Hz and passes no current at bass frequencies. Basically, it is disconnected in the LF range and generates power only at mid and high frequencies.

The voltage waveform may even clip in the LF's when the amp is driven hard, but no power is generated. So there is definitely less of a burden on the SOA.

PLMONROE · « **Reply #3 on:** 18 Oct 2006, 10:58 pm »

DEQX -- YEA! Do away with speaker designers -- NAY! (At least not all of them)

warnerwh · « **Reply #4 on:** 19 Oct 2006, 04:19 am »

So I read into this that the load the amp sees when passive biamping would actually let it run cooler as it is passing less current. This is what I thought but want to confirm because at one point I thought as Housteau did that the energy was dissipated in heat.

BobRex · « **Reply #5 on:** 19 Oct 2006, 02:51 pm »

Hey Brian,

I've been thinking about the plate amp and EQ since you announced the amp so many moons ago.... The RDES appears to be a good idea, except that you can't fully defeat the inboard crossover (it only high passes to 200Hz, too low for RM30s and 40s to use.)
Any chance you could convince Mark to develop a "VMPS" version that eliminates the crossover?

The VMPS SMS-1 may be the better solution (albeit twice the cost) since you can fully defeat the xover.

James Romeyn · « **Reply #6 on:** 19 Oct 2006, 10:14 pm »

I understand the suggestion to be:

Use one PE sub amp per speaker, only for the 6.5s on my RM30C. Is this correct? Did Brian mount one amp inside each speaker, or is each amp sited externally? Too bad these amps look strange sited externally.

I also understand that the sub amp must be modded to allow the low-pass xo pole to be set high enough (around 280 Hz) to x to the ribbons. Please describe the mod. FYI my buddy just bought this sub amp; he said a PE tech told him the slope is definitely 4th order. OEM passive is 1st order, but both 1st & 4th order are in-phase, so that's great. Obviously the oem passive coil must be by-passed.

This sounds sexy. I talked to B about this exact scenario several months ago. Maybe even the smaller 500W amp would suffice. I don't know the price difference.

Practically free ultra-cheapo active high-pass 1st-order xo for the 6.5s: Last time I used a sub, I utlilized small non-polarized speaker caps at the inputs of the seperate amp driving the 6.5s. The caps are now packed for moving, but I believe the values that were appropriate for a high-pass first-order xo pole of about 100 Hz (I had some strange modes I was working around) was about .047uf, give or take (doubling should move the pole down one octave, halving moves it up one octave). Obviously these caps interact w/ the amp input impedance, meaning the values are'nt universal. Change input z = changed filter pole. But it works. You just add/subtract small cap values in parallel to alter the pole. I think the lower the filter pole the less critical is the cap quality.

warnerwh · « **Reply #7 on:** 19 Oct 2006, 10:35 pm »

Here's a room mode calculator for Windows that's easy to use: http://www.realtraps.com/modecalc.htm

You can just enter the dimensions of your room to the nearest inch and hit enter. For 300 bucks the Behringer DEQ 2496 with the ECM 8000 microphone will tame alot of room modes as well as work as a realtime spectrum analyzer and other functions as well.

This piece easy to learn to use properly. It also takes time to get things set perfectly although major improvements can be finished in an hour. This is especially important for people who have no way to put in real room treatments. It's no substitute for bass traps or room treatments but can help in a major way even if you have treatments in your room.

Brian Cheney · « **Reply #8 on:** 19 Oct 2006, 10:36 pm »

I'll have John Curl check the filter slopes when he measures the amp.

If you remove the stock frequency pot and substitute double the value the marked frequencies double. I have a value that gives you 280Hz with the pot all the way up but I ain't talkin about it.

We mount the plate amp in the OXO. The same amp is available with chassis as a standalone for about the same price, call PE for details, it's new.

Brian Cheney · « **Reply #9 on:** 19 Oct 2006, 10:46 pm »

Rex:

mls is putty in my hands and will do anything for me. I'll pass along your demand. 300Hz/24dB will do the job.

John Casler · « **Reply #10 on:** 20 Oct 2006, 12:40 am »

Yes, as B, states, the 1000w amp is now available in 2 versions.

1) is the Plate amp which pretty much limited our application to those speakers which had a large enough cabinet to allow mounting (only RM40 and RM2)

2) The new version is a "fully cased" component just like any amp.

This has expanded our application ability to include "ANY" of the line that has 2 pair of binding posts (one for the bass system, and one for the mid/hf's)

So now we have 3 options of deployment to use this amp.

1) As a Plate amp for any and all the subs

2) The component version can drive any sub, and still stay in your rack or stack

3) As a "Modded" plate amp for the RM40 or RM 2, mounted in the cabinet

4) As a "Modded" component amp for any of the VMPS line with two pair of binding posts.

The Sub version of the amp sells for $398 each MSRP

The "modded" PBS system for the Mains sells for $599 each MSRP

The advantages are numerous for all the applications, in that there is a lot of flexibility in the system.

Those who want a simple method of using lower power tube amps with the VMPS line can now do so without sacrificing Majestic and Dynamic Bass.

Already purchased your VMPS "without" PBS?

NO PROBLEM, just order the component version and "Fire when ready Gridley".

In fact, for most applications the "outboard" version will yield the more flexible application.

But the unmistakable outcome to this application is 1000!!!!!!!! Freaking WATTS that grab those woofers and spank

(OK maybe this isn't a spanking emoticon, but it is funny) them into producing some incredibly solid and dynamic bass while allowing the amps for the mids and tweets, to not have to carry the whole load.

gongos · « **Reply #11 on:** 20 Oct 2006, 02:05 am »

How much is option #4 and will 1 of them power both woofers on my RM2s?

John Casler · « **Reply #12 on:** 20 Oct 2006, 02:12 am »

Quote from: gongos on 20 Oct 2006, 02:05 am

How much is option #4 and will 1 of them power both woofers on my RM2s?

The amps are monoblocs so it takes two amps per pair of speakers.

However, the RM2's you have might not need the "mod" since they likely x-over lower (155-166Hz or so, instead of 200Hz+).

Then you would be looking at the $398 version.

James Romeyn · « **Reply #13 on:** 20 Oct 2006, 03:26 am »

Quote from: BobRex on 19 Oct 2006, 02:51 pm

Hey Brian,

...The RDES appears to be a good idea, except that you can't fully defeat the inboard crossover (it only high passes to 200Hz, too low for RM30s and 40s to use.)
Any chance you could convince Mark to develop a "VMPS" version that eliminates the crossover?

The VMPS SMS-1 may be the better solution (albeit twice the cost) since you can fully defeat the xover.

What are the above "RDES" & "SMS-1"? Thanks all.

Warner your link is a gold mine. Thanks!

James Romeyn · « **Reply #14 on:** 20 Oct 2006, 03:55 am »

I have often thought, using one amp per bass system, it would be ideal to set each of the parametric eq's to seperate & different poles. Why not flatten 2 seperate room modes rather than one mode w/ both channels? Your thoughts fellow fanatics?

Even if an integrated is used w/o preamp outputs, full-range speaker outputs can drive a 10k pot which in turn can drive the bass amp inputs. The pots act as level controls. I'm using this scenario now

As I mentioned earlier in this thread, small speaker cap values have been employed at bass amp inputs. This 1st order high-pass filter could be set above 71 Hz. This would free the parametric eq to flatten some higher frequency room mode, but I'm not sure of the highest pole setting on the plate/component amp. The excursion of the 6.5s is dramatically minimized, w/ a concomittant decrease in thd & increase in dynamics & power handling. This scenario coupled w/ huge subs is the best I've heard for hair-raising slam & impact.

Even if preamp outputs are available the above speaker to pot to bass amp could yield better results. A 10k load is absolutely invisible to a speaker output source; there is often a far worse source to load ratio between preamp outputs & power amp inputs, sometimes causing audible interaction & degradation.

warnerwh · « **Reply #15 on:** 20 Oct 2006, 04:46 am »

Jim: The Rdes unit is one very nice parametric equalizer. It has Windows based software that has a wonderful interface that makes it very easy to use. If you go to this page: http://www.av123.com/products_product.php?section=processors&product=31.1
you'll get a better idea on what this piece can do.

It's really an outstanding piece that anybody with half a brian can use. It should be a law that something like this is required for any decent stereo system. This will add 5 more bands of parametric equalization as well as an adjustable crossover. I'm amazed nobody has come out with a product like this before.

Housteau · « **Reply #16 on:** 20 Oct 2006, 06:11 am »

I see that the new fully cased version has an extra control for phase adjustment. That should help adding more placement ideas for separate subs.

John Casler · « **Reply #17 on:** 20 Oct 2006, 06:41 am »

Quote from: Housteau on 20 Oct 2006, 06:11 am

I see that the new fully cased version has an extra control for phase adjustment. That should help adding more placement ideas for separate subs.

Good eye Dave.

That additional feature makes it more flexible for sub placement and orientation experiments.

It is really a pretty remarkable unit for the price. 500-1000w for $398 is pretty good value to begin with, but add in the control and the equalizer, and it is a winner.

BobRex · « **Reply #18 on:** 20 Oct 2006, 02:13 pm »

Damn, one of these days I'll learn to type.....

Warner has explained the RDES. The SMS-1 is by Velodyne (not VMPS as I originally (mis)typed). The SMS is another digital parametric EQ that has more automation than the RDES. With the RDES, you must manually plot out the FR of your room and then enter the values into a supplied computer program to determine your room curve. You then set the RDES to compendate. The SMS is almost fully automated (I say almost because you can manually modify the parameters if you wish.)

One thing to keep in mind, with both units expect to buy 2 of them and use them in mono for best results. They (the RDES for sure, I'm not 100% on the SMS - which I think is mono) are stereo in that there are right and left jacks, but the EQ curve is applied to both channels. Since these units were designed to be used with subwoofers, not complete speaker systems, they assume mono bass or at least similar bass from multiple woofers. With the higher VMPS xover points, you can't run them as designed.

BobRex · « **Reply #19 on:** 20 Oct 2006, 02:31 pm »

Jim you are correct that in most cases stereo correction of bass is better. The problem is unless you can have truly symmetrical placement, each woofer channel will excite differently. This can result in some really weird eq being applied to the "wrong" channel in an effort to obtain smooth mono bass. Added to the fact that the crossovers on both RM40s and RM30s cutoff the bass section above middle C, and you've got too much going on to use a mono EQ. 80Hz my be fine (questionable), but that far up the spectrum will be a major issue with mono.

At least that's the way I'm looking at it in my application (PBS on RM30s.) With your installation and the lower crossover to the subs (I think you stayed uned 100Hz if memory served), you might get away with the mono EQ. But I'll bet stereo correction would still be better.