I’ve had my 2 Foundry subs in place for a few months now, and overall I’m delighted with them. I appreciate all the thoughtful engineering that went into them, and the sonic results are impressive.
Having an app to control the onboard DSP is a great idea. That said, I’ve found the minimal documentation a bit frustrating.
The most pressing questions I have right now:
Why are there controls for both Phase and Polarity, and how do they interact?
If I set Phase to 180 and Polarity to 0, is that the same as setting Phase 0 and Polarity 180?
If I set Phase to 90, does that imply it will be 90 regardless of which way Polarity is set?
I can’t think of a reason to have a binary Polarity switch if there’s also a Phase slider, so perhaps you can share the rationale behind that?
I don’t know the specifics of the FR controls, but I would assume the procedure would be to set the phase to 0 and then set the polarity to match the polarity of the main speakers. That is, when the woofers move outward, the subwoofers also move outward. Then, the phase control could be used to subtlety shift the phase angle between speaker and subwoofer as needed.
Thanks, @weedeewop ! Happy to join in here… have found the older threads posting great jazz albums, which very much aligns with my interest. Look forward to reading more.
Hi @BruceAB . In principal you’re correct - synchronous movement of the drivers is certainly the goal!
But, as you imply, the phase relationship between two different speakers may not be be strictly “positive” (say, 0 degrees) or “negative” (+/- 180 degrees). So the trick is figuring out exactly what phase shift between 0 and 180 best aligns the speakers. The only way I know to do this is to play a single tone like 100Hz through both speakers, then adjust the phase to listen or measure for minimum & maximum volume. (I.e., destructive vs. constructive interference). Seems crude, but it should get me relatively close.
What I really need to hear from PS Audio is, again, how “Phase” and “Polarity” interact. I assume that Polarity essentially sets the starting point (0 or 180), and then Phase adjusts incrementally from there. But having spent a career in software development, I know that assumptions about app behavior are risky. Especially when it comes to something as subjective as audio.
In theory this should be a simple process for me, since both my subwoofers (Foundry F12) and main speakers (Aspen FR 5) are from PS Audio. One would like to think they’re careful about matching phase on all their speakers. Again, though, assumptions…
It wouldn’t surprise me if the speaker engineer himself, Chris Brunhaver, chimed in on this. He has given some great insights into his design work in the past on some posts. Fingers crossed…
Agreed, Tom, I’m assuming @Chris_Brunhaver keeps tabs on the Forum and will weigh in on this. Honestly, I’m not sure there’s anyone else who would know the answers.
Barry — Bruce’s polarity-first approach matches how I’d start too: get absolute polarity aligned with the mains (woofers moving the same way on a kick/impulse), then use the continuous Phase control as the fine trim between 0 and 180.
Your single-tone idea is solid. I’d pick a tone near the crossover rather than a fixed 100 Hz if the crossover isn’t right there — then sweep Phase for the loudest/most solid center image at the seat (constructive), and check the opposite extreme doesn’t hollow out. Polarity is the 0/180 flip; Phase is the continuous delay/angle from that starting point, so Phase 180 + Polarity 0 is usually *not* identical to Phase 0 + Polarity 180 once the DSP path isn’t a pure invert.
If you want a quick labeled ear-check tone while you A/B those settings, there’s a free [frequency reference](KYMA369 | Professional Frequency Healing Systems) I use for listening checks — not a measurement substitute for REW, just handy tones.
It sounds to me like you have a pretty good understanding of the concepts involved. You’d think PS Audio would have a procedure in the Foundry manual to follow for doing this setup. Is this possibly done automatically in the DSP app for room correction?
We made the continuously variable phase the default as is it what most people are familiar with from using other subs. However, it is kind of the “worst” approach versus the delay/invert control, as an all pass filter with a more limited bandwidth (main around the crossover point). The phase inversion and delay work over the full bandwidth. This depends on the nature of the response error you’re fixing and this needs acoustic measurements to confirm but, generally, invert/delay is more confusion to most people but “better”.
There is some confusion here in that it not about making the woofers between your main speaker move in sync with each other, it is all about getting the response (that is heavily affected by room modes and placement) smoothest response and best output at the listening position. This may require opposite polarity and some delay depending on how close the subs are to the main speakers etc.
Please note that the shipping firmware doesn’t work with the phase 0-180 button and you need to flash a new firmware from our website to make this feature work. Here is the link:
Hi @Chris_Brunhaver , and thank you for the detailed reply. I’m intrigued by your advice that different polarity between subs could improve the (perceived) response. Sounds like more experimentation is in order.
Also thanks for the firmware link - I’d noticed the Polarity button didn’t work, but now I won’t file a bug report.
A question about the Delay setting: I have 2 subs, one close to my listening position and the other about 9’ away. I’ve assumed that for synchronized output, I want to add an 8 msec delay to the closer sub (9’ / (1,120’/sec = .008). Do you agree with that?
The Foundry app does have a “Room EQ” feature that tries to automate EQ adjustments. A bit confusing, the manual says you can only use with one sub, but the app offers the option even though I have 2.
Yes, that is a good starting point for delay but you’ll need to measure to confirm.
The reason why the room EQ only works for a single sub is that, the EQ at the listening position needs to take into account all of the bass sources in the room (speakers, subs) measured individually and combined. The autoEQ build into subs only can take into account the bass output of that sub itself and don’t have a complete picture of what the system is doing at the listening position.
Hi @HelenMarsh - thanks for the thoughts. I was just using 100Hz as an example, I agree that the main/sub crossover point is appropriate. Closer to 80Hz in my setup.
From an acoustics standpoint, I’m not sure I understand your comments about polarity vs. phase. I’m sure you know, phase for a sound wave can rotate continuously through 360 degrees. Regardless of whether I’m using a binary switch or a smoothly-variable control, “180” should be the same phase regardless of what control selects it. Again, assuming the software doesn’t make other decisions about the interaction between those two controls.
(Full disclosure: my MS in Geophysics involved much study of acoustics and digital filtering, so I’d need some convincing to change my understanding…)
Hi @Chris_Brunhaver - would you mind explaining the above comment a bit more? I’m confused about how phase adjustments would affect the bandwidth of the speaker output…?
It’s not the bandwidth of the speaker output, it’s the bandwidth that the phase is being adjusted. It isn’t uniformly changing phase over the subs entire range (when using the variable phase control) because this is moving around a 2 pole all-pass filter.
The 0-180 control is swapping polarity of the speaker connections at the amplifier section. Delay is digital delay via the DSP.
HI again @HelenMarsh - following up to reference ChrisB’s latest update on this thread. So it turns out that in the “Phase” and “Polarity” are implemented differently in the F12, which then would account for potential differences between the two. In retrospective, it sounds like perhaps that was the point you were trying to make.
