Mullard speaker box strange impedance resonances

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I designed speaker box for two Philips 9710 fullrange speakers, following completely Mullard engineers specification from 1959
Result is fullrange massive speaker box, very nice and bright highs, middles are wide and realistic, bass.. also good level
First I heard some specific resonance, which I suppose comes, because enclosure is not right damped or tight enough. When I red specification as in the picture - there are 2 impedance tops - at 30Hz and 100Hz. First one is excellent and nice, but second.. 100Hz resonance sound like barrel or cartoon box. Why it is there and can it be removed :confused:

Other speaker box I tried to put bassreflex which is adjustable. In this case there is only one resonace about 81Hz and it is not moved even if I change bassreflex depth ?! These drivers have a good resonance about 45Hz and with 93 litter enclosure can I pull it down more or at least keep it, but please not more than 50Hz ?
 

Attachments

  • mull1.png
    mull1.png
    225.4 KB · Views: 228
  • mull2.png
    mull2.png
    90.5 KB · Views: 231
  • rsz_img_20190313_211620.jpg
    rsz_img_20190313_211620.jpg
    26.6 KB · Views: 215
  • rsz_img_20190313_211613.jpg
    rsz_img_20190313_211613.jpg
    20.8 KB · Views: 205
Last edited:
Okay, I was pretty fast I think and went to a wrong direction. Box resonance is in a result of box impedance not equal:

http://images.audiojudgement.com/2016/08/impedance-bass-reflex.jpg

Depending on how the 2 peaks compare to each other, in terms of height, we can consider 3 particular cases :

The two peaks match in height. This means that the resonant frequency of the driver in free air (fs) matches the resonant frequency of the box (fb).
The first peak is higher than the second. This means that fb > fs.
The second peak is higher than the first peak. This means that fb < fs.

In my case fb < fs and according Mullard manual F = 35Hz. Why I hear higher resonace and how to measure and improve the boxes - damping ?
 
Okay,
I was pretty fast, box resonace is not equal to box impedance, but result of it as in the picture. According Mullard f = 35Hz, which is pretty low? Can I calulated it f1=30Hz and U=140mV, f2=98Hz and U=177mV
 

Attachments

  • impedance-bass-reflex.jpg
    impedance-bass-reflex.jpg
    43.7 KB · Views: 200
Bass is too louder in original vent and makes U-u-u-u time to time

In bassreflex bass is light "uuuu" situation is not observed but sounds like suppressed a bit. I removed there also all damping material (empty box)

Here is frequence response
 

Attachments

  • mull3.png
    mull3.png
    104.5 KB · Views: 61
Last edited:
There appears to be some confusion here.

The graph of impedance against frequency in post #1 does not indicate the frequency response of the speaker. If you are hearing resonances, they are not related to the humps in the impedance curve.

The frequency response graph in post #4 appears to be that of a pickup cartridge and not that of the loudspeaker.
 
did you try just the "hole" in the vent's cover by itself without a port tube?

Freddy, you mean something interesting, but please let me know your thought!

Yes, it was interesting I continued with tries and found that as plastic vent goes shorter, the same sound goes better, which is in opposite of the bassreflex theory (long vent makes bass more low). Later on I left only round hole, removed plastic bassreflex

@Galu- thanks for clarifications, firstly I mess impedance with frequency resonances

@turk_182 - my enclosure has absolutely same resonances and responses like in the copied graphs. People use 9710 in different cases, but if you are putting it in crossover project, this driver is going to be useless. His advantage is fullrange projects
 
Finalizing my current experience:
1. Original box has too much bass, which is not only sub-basa, but also includes low mid frequences causes to "U-u-u-uh" sometimes. Can I remove "curtains" damping in original project and use appropriate damping? My expectation is to allow only low bass, but also dont want to lose perfect middle freqs on this driver

2. Second bassreflex plastic obviously didn't gave more good result - plastic tube sound very light in bass (but it is low) no matter of the vent length, even removing the tube and left a hole is better

I have to found decision somewhere in the middle ...
 
To help analyse your problem, could you explain what you mean by "U-u-u-uh"?

The curtains are intended to damp resonances in the mid to high frequency range, but do not effect the bass.

The quality of the bass depends only on the enclosure design.

If there is too much bass, try closing the reflex opening to see what the fullrangers sound like in a sealed box.
 
well if the original size of the box is too big as in too much bass reduce the cubic volume (either cut the box size down or start filling space with solid objects to reduce the volume)

without a proper understanding of what U-U-U-U is i'll have to assume that it's either resonances or breakup modes that are occurring because the driver is not adequately damped at low frequencies in general (forget driver resonance for the moment) what your looking for at the moment is to find a volume of air that acts like a spring against the back of the driver to prevent it from flying to the end of it's travel at low frequencies and yes the resonant point of the driver does generally turn out to be the first place it occurs but if the "air" spring doesn't increase as the frequency decreases it will cause driver damage or go active and high pass it.
 
frugal-phile™
Joined 2001
Paid Member
If what i remember of the 9710 their high Qt means they are not suited to reflex enclosure — which is what the impedance curve indicates. THis box is a poorly designed attempt at periodic but they did not go far enuff (and as a vented box i expect poorly designed (ie from before T/S made it a lot easier). You need to stuff the crap out of that box to push it towards or make it aperiodic.

dave
 
An aperiodic enclosure would produce a single impedance hump of low amplitude.

In accord with Dave's suggestion, but instead of stuffing the c**p out of the box, you could convert the large open vent in the original enclosure design into a resistive vent.

This can be done by stapling a thick sheet of acoustic fibreglass over the inside of the rectangular opening. If found to improve the bass, a more substantial resistive vent can be fashioned by sandwiching the fibreglass between metal mesh sheets.

Having said that, my recommendation to block off the vent entirely to see how the speakers behave in a sealed box still stands - that's worth trying first as we're fumbling in the dark here!
 
Member
Joined 2010
Paid Member
Firstly, the Mullard/Philips enclosure designs are from a period when T & S parameters were basically unknown.
Secondly, the Philips 9710, whilst being an exceptionally good speaker design for its day, will have aged and possibly be in need of some remedial attention. This would require new braided tinsel wires, checking that the corrugated wave spider has not collapsed inwards and painting the outer paper surround with a suitable latex coating.
Based on the parameters from the Troels' site , the enclosure internal volume should be in the vicinity of 125 litre, for two speakers in the one enclosure and this should be 'tuned' to around 40 Hz.
As far as adding damping material to overcome the high Qts figure, the safest way to do this is to fix a wad of Tontine fibre (25mm thick) directly behind the speaker frames, just like a teapot cosy. This will alter Qm slightly and on the downside will add to the box losses; however this is less of a tradeoff than killing the vent output.
The 9710 can be a little forward in its' sound characteristic at the higher frequencies and a slot diffuser of a modified Kolster-Brandes type as described by Gilbert Briggs
in his book ''Loudspeakers" is a useful tweak.
 
Last edited:
If it were perfect it would almost completely flatten the impedance bumps.
Absolutely - nice illustration!

For Powerpz's benefit I'll explain that 'aperiodic' means 'without period' - i.e. 'without frequency' of resonance. In practice, aperiodic loading reduces the height or 'Q' of the impedance hump exhibited by a driver when it resonates at its natural frequency. True aperiodic loading would eliminate the resonance hump entirely.

Powerpz may care to experiment with the suggested methods of reducing the 'Q' in his enclosure - i.e. either by employing a resistive vent or, as helpfully suggested by VaNarn, by stapling a pad of wadding tightly over the rear basket of the drivers.

Basically, he has to fool the drivers into thinking they are in a larger enclosure.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.