Oops - just realized that (now deleted) sim above was for the FE206EN
Here's how W8-1772 would behave in a 4 liter box with and without a passive highpass vs FE206EN
4 liters might be a bit tight for W8-1772 - depending upon what one wants to do with
the crossover to the woofer
Here's how W8-1772 would behave in a 4 liter box with and without a passive highpass vs FE206EN
4 liters might be a bit tight for W8-1772 - depending upon what one wants to do with
the crossover to the woofer
Last edited:
Another question about the 1772's - how good is the pair matching of the frequency response, especially at high frequencies ?
If you measure two drivers in an identical test rig and overlay the results how close are they ?
And do they vary much from batch to batch ? The measurements I've seen online seem to differ a lot, especially from 2-5Khz, but it's hard to tell whether it's batch variations, or just very different measurement conditions, or some bad measurements.
My two Coral drivers are hand assembled new old stock, and as such the pair matching of them is not very good. (Although on the other hand, two brand new factory assembled Fostex FE207's I had were not very well matched at high frequencies either!)
They have a sensitivity difference of 0.87dB at mid/low frequencies, they have a significant difference in the depth of the surround notch at 1Khz, and the response from 2Khz upwards differs quite significantly, and while they follow the same general trend up to about 6Khz the small details differ significantly and they diverge quite a bit beyond 6Khz.
I'm using them in a 2 way system with ribbon tweeters and because of the significant difference between the two drivers I've had to custom design the crossovers and equalisation network of each of the two Coral drivers independently and although the crossovers are broadly the same in design, almost every component value differs between them, in an attempt to get the overall result as close a match as possible between left and right channels.
Quite an ordeal in hindsight so for my next project I'd really like the drivers (hopefully the 1772) to be a lot closer pair matched so I can use the exact same network design for both!
But that is only possible if they're well matched and consistent in manufacture, and I'm not sure how often that's the case with large full range drivers where the tiniest manufacturing tolerances can affect the high frequency response significantly.
If you measure two drivers in an identical test rig and overlay the results how close are they ?
And do they vary much from batch to batch ? The measurements I've seen online seem to differ a lot, especially from 2-5Khz, but it's hard to tell whether it's batch variations, or just very different measurement conditions, or some bad measurements.
My two Coral drivers are hand assembled new old stock, and as such the pair matching of them is not very good. (Although on the other hand, two brand new factory assembled Fostex FE207's I had were not very well matched at high frequencies either!)
They have a sensitivity difference of 0.87dB at mid/low frequencies, they have a significant difference in the depth of the surround notch at 1Khz, and the response from 2Khz upwards differs quite significantly, and while they follow the same general trend up to about 6Khz the small details differ significantly and they diverge quite a bit beyond 6Khz.
I'm using them in a 2 way system with ribbon tweeters and because of the significant difference between the two drivers I've had to custom design the crossovers and equalisation network of each of the two Coral drivers independently and although the crossovers are broadly the same in design, almost every component value differs between them, in an attempt to get the overall result as close a match as possible between left and right channels.
Quite an ordeal in hindsight so for my next project I'd really like the drivers (hopefully the 1772) to be a lot closer pair matched so I can use the exact same network design for both!
But that is only possible if they're well matched and consistent in manufacture, and I'm not sure how often that's the case with large full range drivers where the tiniest manufacturing tolerances can affect the high frequency response significantly.
Last edited:
Good morning,
I have built a TQWT speaker using the famous W8 1772. It sounds for me a little brilliance and this fatiguing harshness is, I have to admit, a disappointment... of course, the speaker is brand new and need certainly some burning timel. Do you think a Zobel circuit will correct this acidity? If yes, which resistor and capacitor references are recommandante? Thank you for your advices. Musicaly, yours. TYM
I have built a TQWT speaker using the famous W8 1772. It sounds for me a little brilliance and this fatiguing harshness is, I have to admit, a disappointment... of course, the speaker is brand new and need certainly some burning timel. Do you think a Zobel circuit will correct this acidity? If yes, which resistor and capacitor references are recommandante? Thank you for your advices. Musicaly, yours. TYM
Pass DIY Addict
Joined 2000
Paid Member
I had good success with a simple notch filter composed of a 0.56uH inductor in parallel with a 15R resistor. Place this in series with the speaker to tame the top end.
I got the idea from the Transcendent Sound webpage. Their implementation uses a variable resistor so you can dial in the amount of treble that you like best. I've posted some plots just above in post #561. Ignore the absolute dB scale on the left (I didn't adjust it properly), the point was to plot the relative differences among networks.
I've been contemplating building a TQWT for mine, right now they reside in a 40L bass-reflex box. Would you mind sharing your design?
I got the idea from the Transcendent Sound webpage. Their implementation uses a variable resistor so you can dial in the amount of treble that you like best. I've posted some plots just above in post #561. Ignore the absolute dB scale on the left (I didn't adjust it properly), the point was to plot the relative differences among networks.
I've been contemplating building a TQWT for mine, right now they reside in a 40L bass-reflex box. Would you mind sharing your design?
Last edited:
one blog suggested a BSC of 1.1mH and 2.7 ohms (a Zobel network had no sonic effect for that experimenter) - sometimes it doesn't take much to work wonders (one leg of an L-pad can be used to tweak things by "ear" - measure then replace with a fixed
resistor)
Google Translate
resistor)
Google Translate
Last edited:
Pass DIY Addict
Joined 2000
Paid Member
Pass DIY Addict
Joined 2000
Paid Member
How's this?
Roughly speaking: The signal from your amp must pass through both the inductor and resistor at the SAME time (thus, the inductor and resistor are in parallel). The signal must pass through the inductor+resistor combination BEFORE it gets to the speaker (inductor+resistor combination is in series with the speaker).
Roughly speaking: The signal from your amp must pass through both the inductor and resistor at the SAME time (thus, the inductor and resistor are in parallel). The signal must pass through the inductor+resistor combination BEFORE it gets to the speaker (inductor+resistor combination is in series with the speaker).
Attachments
Good morning,
Finally, I built your baffle step corrector recommandation ( coil 3,9 mH- 1,4 mm/ resistor 10 Ohm-10w) and I have to say after several minutes of careful listening: it works very well. I am pretty glad and thank you very much for this great advice. The result is a global better balanced spectrum and no more acidity and fatiguing effect... I don’t have the feeling of losing details or deepness.
I have, please, another question: shall I add a Zobel network in order to achieve the optimisation?
Warmest regards,
TYM
Finally, I built your baffle step corrector recommandation ( coil 3,9 mH- 1,4 mm/ resistor 10 Ohm-10w) and I have to say after several minutes of careful listening: it works very well. I am pretty glad and thank you very much for this great advice. The result is a global better balanced spectrum and no more acidity and fatiguing effect...
I don’t have the feeling of losing details or deepness. I have, please, another question: shall I add a Zobel network in order to achieve the optimisation?
Warmest regards,
TYM
Pass DIY Addict
Joined 2000
Paid Member
TYM - great to hear that you are happier with your speakers now! As you can see above, there are a number of combinations of coil and resistor values that people have used to solve this problem. Each will provide a slightly different result and I suspect that each works better for different room sizes/configurations and speaker placement. You might want to experiment a little with different coil sizes or resistors sizes and see which you like best.
I don't have any real insight on adding a Zobel, perhaps others can help here. I'm of the general opinion that the fewer things between your amp and driver, the better.
I don't have any real insight on adding a Zobel, perhaps others can help here. I'm of the general opinion that the fewer things between your amp and driver, the better.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.