[Mijknarf]

I've been using pink noise test tones and a RS SPL meter to make some freq plots of my OB speakers.  As I've moved the crossover point down from 175Hz to 100Hz and increased the output level on my plate amps to get the best plot, the bass in the sweet spot is just OK, but as I step into any off axis position it is much more pronounced.  Just stepping two feet to the left or right the bass is significantly stronger.  The off axis is so strong that when the rig is playing in the basement, the bass upstairs rattles the china in the dining room cabinets.  I've lost my way here, any help will be appreciated

[Poultrygeist]

Hi M,

I always admire your great looking gear. This may not be helpful but I pull my H-frames way out from the corners and toe them in so they are pointed towards me at the apex of a triangle. I never listen to them any other way.

FWIW my main drivers are on the opposite side from yours. I cross my Alphas at 80hz but with two Daytoin APA150's which I never drive above half volume. It's hard to imagine H-frame Alphas rattling china at least mine never have. Could those powerful looking plate amps be the culprit?

[Rudolf]

M,
room dimensions?
speaker positions?
listener position?
Without such information from your side it is guessing in the dark.

Almost no toe-in of the H frames is a big NO if you listen in the standard stereo triangle.

Rudolf

[Mijknarf]

Room is 14' wide, 24' long
Speakers are 5' out from front wall and 2' out from the side walls
My chair is 10' back from the speakers
They're toed in to point at my shoulders

I've had these for 6 months w/o this condition,
So I'm assuming it's not the set up.

It wasn't until I lowered the cutoff And boosted the level.

I just tried switching the phase on the plate amps and
That has helped.

Poultry:  I've run the Alphas with ss amps and lower wattage
Plate amps and until I tried these larger plate amps, I didn't
Know what they were capable of.

Many thanks so far.  Can my layout be improved?  Any other
Info needed?

[richidoo]

I think 100Hz is too low for the Vistaton top driver, but you can see where it rolls off when you measure it by itself. Raising the crossover point will make the sound more full and better dynamics. Might also fill in some floor bounce.  If you are boosting bass driver to fill in floor bounce then the low bass will be too loud.

China rattling upstairs means the low bass is boosted too loud. If the bass driver that you are using has highish Qts (as often recommended for OB applications) then it may not need much boost at all. 

The off axis rise in bass SPL implies that you have a null at the listening position and you are trying to fill it in to achieve flat response in the null. This makes the rest of the room not in the null too loud. Using just the SPL meter you can research this using modecalc.exe to determine what frequencies will ring in your room, then excite those freqs with realtrap's 1Hz test tones to verify the null freq to know which axis is creating it, then move the listening spot out of the null. Sometimes moving the speakers only a couple inches or less can move the null a large distance away from the listening spot. 

Also, don't believe the myth that good audio must have a flat frequency response. You were happier when you had the null instead of filling it in flat. Also, not everyone likes flat bass going too low. I don't. It is fun on dance music and organs, but unnatural on classical and good acoustic recordings. Adjust it by ear to what you like. Don't be slave to the SPL meter.

Check out some speaker measuring software for the computer to help you really dial it in. RoomEQWizard, ARTA, HolmImpulse, OmniMic, etc, will let you see much more of what's happening. If you want to stick with SPL meter, try using Ethan Winer's (realtraps.com) sinewave test tones instead of or in addition to your pink noise. This will allow you to plot the dB every 10Hz or finer. Liberty Instruments also offers free SynRTA which is OK for a rough view of FR. The RS SPL meter's bass accuracy is not great. A small diaphragm condensor mic (like omnimic) is important tool for bass adjustment.

Your system is truly beautiful...  Nice work!   Did you DIY the electronics too? Is that an active crossover circuit on bottom shelf between the plate amps?

[Mijknarf]

Richidoo, your response makes a lot of sense to me.  I'm guilty of trying to achieve a ruler flat response and that is where I went wrong.  In the past I had the low pass set at 175 Hz and it sounded good, but I wanted more bass punch, so I lowered the low pass to 100Hz and boosted the output in an attempt to reach the same output level as the rest of the freq. spectrum.  It gave me a little dip in the 175Hz range (between the low pass of the woofer and the point where the Visaton+baffle start to contribute) but the low end 40Hz-100Hz was boosted significantly (off axis of course).   The response graph looks good, but there is a large null at the sitting position.  I will experiment with your suggestions and see what happens.  For now, I'm returning to the higher low pass until I get get some resolution.

My WE91A 2A3 SET amp and the JE Labs Linestage were made by Min at Tube Audio Lab.  I built the cherry bases and Min did the rest.  The device on the bottom shelf between the plate amps is a USB NOS DAC I've had for many years built by Peter Daniels at Audiosector.  The Visatons run full range straight from the amp.  I use the EQ in J.River to cut the freq response between 3-7KHz where the Visatons seem to spike.  I use a simple a 4.7mH coil on the woofers.

Thanks again for everyone's contributions, I'll follow through with the suggestions until I dial in the right sound.

[*Scotty*]

Deleted

[*Scotty*]

Mijknarf, at exactly what frequency is the large null you measured at the listening position occurring at? And, where in the room, relative to your listening position, are you measuring the large off axis boost between 40Hz and 100Hz?
Scotty

[Mijknarf]

I'm not sure if this will help, but here are the freq plots of the individual drivers and the combined:





I have to go back and measure the excessive bass off axis to see where it measures.  I'll get back to you later today with those measurements. 

I have tweaked the speaker positioning and backed down some of the low end boost as well as changed the phase on plate amps and all this has helped. 

[Rudolf]

Mijknarf,
my guess:
You are sitting in the null of a ~80 Hz lateral mode and trying to fill that up. When your H frame was reaching 175 Hz, the booming next lateral modes at ~120 and 160 Hz would have kept you from boosting the H frame level. But now even 120 Hz are rolled off a bit and you are still trying to fill that 80 Hz null.

As richidoo suggested, you will probably need better measurement capabilities to achieve control of whats happening in your room.

Rudolf

[Mijknarf]

In response to Scotty's question above and to give a bit more clarity... below is the freq response at roughly 20 degrees off axis (green line), it demonstrates the much higher bass I'm hearing.  I have no experience with room treatments, so I'm not sure where to start, perhaps corner traps?  Not sure?

Do I have a null at the listening position, or a boost at the off axis position?