Seeking advice on Extending Low-Frequency Response in 70's speakers

[Wafoe]

I recently acquired a pair of Philips 22AH487 speakers from the 70s, and I'm absolutely in love with their vintage aesthetic. However, I'd like to give them a modern upgrade while keeping their original look.

While they sound decent overall, the bass response is disappointing, especially considering the size of the enclosure. These are sealed three-way speakers with an 8-inch woofer and an 8-inch passive radiator, but the bass rolls off sharply at around 50 Hz (48 dB/octave), which is underwhelming for the type of music I enjoy. Ideally, I'd like to modify the speakers to extend their low-frequency response without altering the external box design or changing the original drivers.

In-room bass response (UMIK-1, 75 dBC):

My knowledge of speaker design is limited, but I'm eager to learn. From what I understand, achieving better low-end extension depends on factors like the woofer’s quality, its Thiele/Small parameters, and the tuning of both the enclosure and passive radiator. In the 70s, speaker design often relied on trial and error, with less advanced measurement techniques, which sometimes led to sub-optimal performance. Back then, increasing the enclosure size was a common solution to improve bass, but since I want to preserve the original box, that isn't an option.

I've considered a few potential strategies to improve the bass response, but I’m unsure if they will address the issue:

• Adding mass to the passive radiator
• Bracing the interior of the cabinet
• Sealing any small air leaks (such as around the speaker wire hole)

Below are the box measurements for reference:

• Width: 393mm
• Height: 588mm
• Depth: 225mm (210mm to the front baffle)
• MDF thickness: 13mm
• Calculated volume: 38 L

I’ve also attached the datasheets for the woofer (Philips AD8066/W4) and the passive radiator (Philips AD8000). I’d appreciate any suggestions or guidance on how to approach this project!

 

[solderdude]

EQ not an option or satisfying ?

The passive radiator is tuned to the woofer and in the correct closed volume. Lowering the resonance of the radiator is not going to extend the lower frequencies but rather change the response at the resonance frequency of the woofer.

Changing the the driver (and radiator) to 2 similar sized woofers could improve LF response but will also change the sound in other ways and needs to be at the exact sensitivity of the original woofer anyway.

Easiest solution... add a subwoofer

 

[ppataki]

The steep bass roll-off is normal if you have a PR in your cabinet - it is similar to a ported speaker but even steeper
EQ cannot help much here since there is nothing left to boost

There are two options in my opinion:
- remove the PR, seal the box: that will give you a nice gentle 12dB/octave bass roll-off that you can EQ and btw the transients will sound way better too compared to the PR design
- leave the speaker as-is and get a subwoofer and integrate it properly

 

[AnalogSteph]

Looks like the woofer isn't exactly happy playing loud below 50 Hz anyway (and it can only take 40 W, too), so a subwoofer really seems like the best option.

I would also really want to take some better measurements, maybe outside or groundplane or quasi-anechoic or whatever. These are more room modes than anything else. Some nearfield measurements of all drivers with appropriate windowing would also be interesting, I'd like to know e.g. whether the 3 kHz breakup mode is adequately suppressed.

 

[DVDdoug]

Yeah, it wouldn't be too hard to make a panel to temporarily replace the PR to see how it sounds as a "simple" sealed cabinet. But as you say, without the Thiele-Small parameters, trial-and-error isn't really practical for a ported design.

You could get some speaker design software, plug-in the cabinet volume, and the Thiele-Small parameters for 1 or 2 replacement woofers to virtually experiment with sealed and ported designs with various drivers. You'll probably have to get rid of the passive radiator (so you'll need to block the hole or use a 2nd woofer).

As a rule, with a given cabinet size you can get deeper bass with a single woofer (smaller total cone area) but more base with two woofers. The software will help with that as long as you have the Thiele-Small parameters for the woofer you're considering.

Or, there is a speaker measurement gizmo that can measure your existing woofer, so again you could virtually experiment with a completely sealed box or port variations..

 

[DanielT]

As already mentioned in the thread add subwoofers.And then make sure that your Philips 22AH487 does not play lower than down to 50-80 Hz, that is, add an HP filter. That so they don't have to work so hard with the lowest frequencies. That in itself can lead to audible distortion (depending on the level of volume/SPL)

A general note. A good integration with subwoofer may require some work but damn it can be good, when/if it is good.

And as AnalogSteph pointed out, Philips AD8066/W4 bass driver only has 40 watts of power handling. It comes down to not stressing such a driver too hard.

For a sealed box, Philips recommends 25 liters for this 8 inch HIGH POWER WOOFER LOUDSPEAKER.
In its day maybe it was like that, but that time is long gone. That 40 watts is considered a high power bass drives that is.

Philips AD8066/W4:

Philips AD8066/W4

 

[sergeauckland]

As others have said, you'll need to add a subwoofer. With some loudspeakers, you can extend the LF by equalising, but in order to do that, you need to have the power handling and excursion capabilities your loudspeakers just don't have. They were designed to be used as they are, within their limitations.

Adding a subwoofer can still be done within the mid-century aesthetic if you're happy to buy used, or into DIY. The J Rogers subwoofer which was sold to accompany the JR149s, looks like a coffee table. The Bassett subwoofer was a DIY project in HiFi News and doubles as a window seat, so subs can be disguised to fit the decor. Tapped horn subs can again be made to look like a table or seating.

There's plenty of free loudspeaker design software available that you can use to design a sub in almost whatever shape you like.

The crossover and amplification is a little more complicated, but not by much, as small amps and DSP crossovers are easy to source.

S.

 

[DanielT]

I recently acquired a pair of Philips 22AH487 speakers from the 70s, and I'm absolutely in love with their vintage aesthetic. However, I'd like to give them a modern upgrade while keeping their original look.

View attachment 395800

...I'm absolutely in love with their vintage aesthetic...I can agree with that.

Speakers around 50 years old, played on through the decades. It can take its toll. Measure both speakers and see if there is much difference between them.
IF , I say if, that is the case, I would, if I were you, consider whether something could be done about it.

 

[Killingbeans]

I’d appreciate any suggestions or guidance on how to approach this project!

Have the surrounds on the woofers been replaced lately?

When the rubber surrounds on a driver stiffens/harden over time, the Fs shifts.

 

[Flaesh]

its Thiele/Small parameters

You can measure the parameters of your drivers; see REW manual for how to do this. No scientific apparatus is required, just some wire and a couple of resistors.
I cannot confirm or deny the data of the venerable Kees, who wrote 20 years ago:
Ik heb net even snel de T/S paramters van de eerste de beste AD 8066/W4 opgemeten en de speaker voldoet geheel aan mijn verwachtingen.
Dc Resistance = 4.1 Ohm
Resonance frequency = 46.2 Hz
L = 0,5 mH (Handig voor Zobel netwerkjes).
Qms = 4.343
Qes = 0.751
Qts = 0.641
Mmd = 16.06 gram
Vas = 51.56 Liter
BL factor = 5.05
Rendement = 90.3 dB/W/m
Let's assume Sd of 200 no, of 220 sq. cm and put this driver in a simple sealed 38 liter box in half-space:

(gray is 2 drivers in series)
PR or BR will not add any significant benefit:

Put the cover on the PR (inside the box) and then deal with the sub.

 

[kemmler3D]

Relative difficulty based on my observations:

Just get a sub and integrate it adequately: 1/10

Get a sub and integrate it seamlessly with the existing speakers: 2/10

Design a highly respectable 2-way from scratch that rivals Genelec or Neumann: 6/10

Re-tuning a 3-way and redoing the crossover to extend bass, but keep everything else the same: 9 / 10

Design a cardioid speaker that rivals TOTL designs: 10/10

 

[617]

Get or build a sub, visible or not. Those are very pretty speakers and I would not modify them unless they were malfunctioning.

 

[Doodski]

Get or build a sub, visible or not. Those are very pretty speakers and I would not modify them unless they were malfunctioning.

I would be leery of putting money into a speaker of this vintage. How much money do you think could be put into them to bring them up to a appreciable higher standard?

 

[617]

I would be leery of putting money into a speaker of this vintage. How much money do you think could be put into them to bring them up to a appreciable higher standard?

The drivers have inherent limitations, getting output below 50hz would cause output to drop precipitously. Sealing airholes won't do anything and adding bracing will if anything reduce bass by reducing cabinet volume.

 

[617]

Relative difficulty based on my observations:

Just get a sub and integrate it adequately: 1/10

Get a sub and integrate it seamlessly with the existing speakers: 2/10

Design a highly respectable 2-way from scratch that rivals Genelec or Neumann: 6/10

Re-tuning a 3-way and redoing the crossover to extend bass, but keep everything else the same: 9 / 10

Design a cardioid speaker that rivals TOTL designs: 10/10

It is fun to think about. The woofers probably don't have any inductor on them, just a big cap running first order into the mid. Just my guess but that's how old speakers are. So you'd have to use that inductor to depress the entire passband of the woofers to extend the bass.

By the time you're done the cabinet would be so full of crossover components it would throw the bass tuning off.

OP I don't mean to dissuade you, but as a lover of old speakers, respect the intelligence of the people who put these machines together without the aid of computers, and just get a sub. If you high passed them at 150hz they'd probably play quite loud.

 

[Flaesh]

https://elektrotanya.com/philips_22ah487_hifi_box_sm.pdf/download.html - free download claimed

 

[DanielT]

Adding a subwoofer can still be done within the mid-century aesthetic if you're happy to buy used, or into DIY. The J Rogers subwoofer which was sold to accompany the JR149s, looks like a coffee table. The Bassett subwoofer was a DIY project in HiFi News and doubles as a window seat, so subs can be disguised to fit the decor. Tapped horn subs can again be made to look like a table or seating.

If TS has the interest and desire to build a pair of bass modules, an aesthetic solution could be to build two and use these as speaker stands. Place the Philips 22AH487 on the bass modules. Then it would have been nice if all units had the same color, veneer and the same speaker protection fabric. Bass modules that look like they are from the 1970s but have killer good modern bass capability.
Bass modules with the same width, 393mm, like the Philips 22AH487 speakers. Height suitable to get the tweeters at ear listening height perhaps. Depth, hopefully some flexibility there.

It's a matter of taste, but I had thought about something like that anyway. In practice, it will then be a pair of four-way speakers. With two bass modules, you can also set the crossover frequency quite high, say 200-250 Hz, because they play in pairs/stereo. It increases the flexibility of finding a suitable crossover point.

____

Generally. What the OP can do with his Philips 22AH487 speakers, hm. Now I throw out what I come up with here and now. It is not at all certain that it is needed or that TS wants to tinker with this:
Unscrew the drivers and measure them. Make sure they are matched, i.e. measure the same.

Measure the FR of both speakers and see that they measure reasonably similar (acceptable similar).

Maybe take out existing damping material and put in new (maybe unnecessary but that doesn't cost many $).If you are going to be picky, you should follow up with measurements after it is done.

A 50 year old passive crossover, how is it? I know, don't replace something that works, but even if it works, maybe it could be better? With new capacitors that measure the same, it perhaps increases the possibility that the speakers sound more the same, have fairly the same FR for example, observe perhaps.
I know the discussion here on ASR about the pointlessness of overkill modifying passive crossovers as for example Danny GR Research do BUT it can hardly hurt to replace a few capacitors in a 50 year old crossover, IF those new capacitors don't cost that much $.
But okay, TS can measure the FR for each speaker before the replacement of parts in the crossover takes place.

What else to do, hmm. Lubricate the rubber edges of the Philips AD8066/W4 bass drivers.

If TS thinks it is necessary, tidy up the surface finish on the speakers.

 

[Pareto Pragmatic]

I would be leery of putting money into a speaker of this vintage. How much money do you think could be put into them to bring them up to a appreciable higher standard?

Assuming the surrounds are good, I would just recap them. $20 with half of that being shipping for 6 caps is what I paid. More for fancy caps of course, if you want them.

I have done that, and then put $1000 of new drivers into vintage speakers. The effect of that $20 seemed pretty close to the $1000 (but I did the drivers one at a time, so who knows). The $1000 was worth it because my wife loves those things and won't let them be moved much less sold.

I have seen projects that add ports to vintage sealed speakers. Never saw that as a good idea, personally. After looking at some results and thinking about it, still don't.

Let vintage be vintage, recap. That's my take on it.

(and yes, another vote for adding sub(s) as the way to go).

 

[DanielT]

https://elektrotanya.com/philips_22ah487_hifi_box_sm.pdf/download.html - free download claimed
View attachment 396029

Here are some pictures of the components of the Philips 22AH487s crossover:

Vintage Audio Sweden

Philips 22AH487 filter capacitor replacement. On the 3rd picture you can clearly see the 8" passive radiator. I'm quite surprised by the sonic quality of those speakers.

www.facebook.com

Philips 22AH487 - Le forum Audiovintage

 

[Wafoe]

I would like to thank you all for your responses. They have been incredibly helpful in kicking off my research of speaker design. As my time for hobbies is quite limited at the moment, I haven't had the time to respond earlier. But I did some experiments and developed new ideas for this project. I’d like to share where I’m at and get your thoughts.

First, responding to some of your comments:

• Many of you suggested adding a subwoofer. I agree the Philips AD8066/W4 woofer struggles with low frequencies, and there’s not much to improve there. While adding a subwoofer is the most straightforward fix, this project is also a learning experience for me. I’m excited about the idea of squeezing better bass from these classics without external help.
• Building my own subwoofer to match the style of the existing setup is a cool idea, but since I lack woodworking tools (or skills, really), that’s not a feasible option for me right now.
• Some suggested sealing the boxes for smoother bass roll-off. To test this, I picked up the smaller Philips 22AH486, which is in the same series as the 22AH487 and uses the same drivers but without the PRs. The 486s sound tighter and smoother in the bass, which I prefer. If paired with a subwoofer, this would definitely be a solid option. Yet, they don’t look as striking as the 487s!
• The woofer surrounds are surprisingly soft and pliable—made of rubber—so I doubt they’ve degraded much over time. That said, I can’t compare them to brand-new ones, so it’s hard to say for sure.
• I measured the individual drivers in the 487s up close and noticed the crossover points differ, likely because capacitor values in the crossovers have drifted over time.
• Oh, and a small correction to my earlier notes: the box volume of the 487 is actually 35L, not 38L, because the front baffle and rear walls are 20mm thick, not 13mm.

A new approach, combining elements from Philips' 487 and 488 speakers
In my research, I discovered another step up in the same Philips range—the 22AH488 (also marketed as the 22AH468, which has a different look). These use the same 35L box as the 487s but feature two Philips AD8067/W4E woofers instead of the single AD8066/W4 woofer and a PR. The AD8067s are optically similar to the AD8066 from the front but have larger magnets, bigger voice coils, and are generally regarded as "real woofers" instead of broadband drivers like the AD8066.

To test this theory, I got my hands on a pair of 488s. Though I don’t have T/S measuring tools (like DATS v3) at the moment, I found the specs for a similar model, the AD8067/W8 (the 8-ohm version of my 4-ohm AD8067/W4E, which is also about 3 dB less efficient):

• Fs: 39 Hz
• Vas: 40.8 L
• Re: 7.7 Ω
• Qms: 5.1
• Qes: 0.42
• Qts: 0.388
• Mms: 25.2 g
• BL: 10.6
• Sd: 210 cm²
• Le: 0.1 mH
• Sensitivity: 89.8 dB/2.83V/m
• Xmax: ±4.0 mm

These woofers seem better suited for bass than the AD8066, but I find it odd that Philips put two of them in the same 35L box. More bass = better bass? Another interesting detail about the 488s: they have car lamps in their crossovers to handle high-power amps and a recessed midrange to prevent it from exploding under heavy loads.

My idea is to combine the best parts of the 487 and 488 designs:

1. Replace the AD8066 woofers in the 487s with the AD8067s from the 488s. However, these are more efficient and would require crossover tweaks. Since I’ve been meaning to experiment with DSP anyway, I could bypass the internal crossovers entirely and try to design my own using a DSP board and a 3-way amplifier per speaker.
2. Use the PR design from the 487s for better low-frequency extension. The AD8000 passive radiator is reportedly tuned to 40Hz, which I believed can be lowered by increasing the Mms (i.e., sticking weights to the metal dust cap). That said, I think its Xmax is identical to the AD8066/W4 and AD8067/W4E woofers (as the surrounds appear to be the same thickness), which could be an issue (and if so, was already an issue in the original design of the 487s). I believe a PR should ideally have twice the excursion capability of the main driver. Maybe I’ll just keep the volume low—or add another AD8000 radiator in the back, lol.

Unrelated, but one thing that bugs me about this lineup of speakers is the lack of dedicated left/right speakers. The driver layout is identical on both sides, which surely doesn’t help sound quality.

What do y'all think?