the audible problems
Hi RDV , I searched around to try to find a frequency response plot for the Goldwood woofer you bought , as no plot on the Parts Express site . Have a look at :- http://members.aol.com/_ht_a/Debertin/gold.htm , and which also appears at :- http://members.aol.com/Debertin/gold.htm . There you will see he shows a plot for the 4ohm version of that Goldwood driver . The 8ohm version will be similar but not identical - cheap drivers often have different upper-mid responses between 4ohm and 8ohm versions , BUT , the 8ohm if different will likely have less of that sharp peak at about 4.4kHz - and if so , good ! ___ What you bought is really a bass driver only , and not good for midrange - look at that drop from about 1.5kHz - that is part of what you are hearing . ___ OK , so $10- is a bargain , but you got somthing you will have to try to be clever with to make work - a good challenge to start speaker designing with ! ___ It can be got to work - it will not be ideal , but it can be better than you have got so far with it . ____ Do you have any way of measuring its frequency response - at least from about 1kHz to 8kHz {or slightly less} ? , if not then we'll guess its similar to the 4ohm version . If you have an Audio Oscillater , then connect it through your amplifier and slowly sweep up from about 2k and through the 4 to 5k region to beyond and listening at at moderate volume level because these are screech on the ear frequencies . Listen directly on-axis to the woofer . The human ear does NOT hear in flat response like a measuring microphone , but you should be able to hear a peak that is of much higher amplitude than the area on either side . Listen outdoors or in a dead sounding room , because resonances in a live room will confuse the issue . Find the high point of the upper mid peak by ear - this is where to cross-over that woofer - exactly on its peak . Use a 2nd order electrical - or a first order electrical {inductor only} plus a Zobel in parallel with the woofer . If you cannot find data to help and do not have an AC Voltage measuring meter that works accurately to at least 5kHz , the for a start point Zobel use an 8.2ohm , 10watt , resistor in series with a 10uF film capacitor . It may need larger capacitor - you can try adding caps in parallel with the 10uF to try 12.2uF and 14.7uF , etc .... Resistor should not be lower than the DC resistance of the Goldwood - so measure that . A 6.8ohm resistor may work better than 8.2ohm . Next connect the inductor - probably a .33mH or a .27mH - depending on the mid-range impedance you have achieved with your Zobel , and the exact location of that Peak . If the DC resistance of your Goldwood is greater than 6.8ohm , use 8.2ohm in the Zobel and .33mH ; if DC is less than 6.8 , use 6.8ohm and .27mH . This is electrical 1st order , but with the roll-off of the driver above the mid peak you will be getting approximately a 3rd order . To combine with tweeter you will have to experiment and listen - try Linkwitz-Riley 2nd order {-6dB at the crossover frequency} and basic 3rd order from charts {assume the impedance at crossover freq. of your tweeter is about 10% larger than its DC resistance}. Try both polarities connction with the tweeter . If you do not get OK sound then leave the Zobel on the woofer and try Linkwitz-Riley 2nd order electrical there also , and you can increase on tweeter from 3rd order to 4th order Link-Riley . ____ Post you results here if you would like further comments .____ Better would have been to buy a woofer after seeing its Response Plot . Draw the graphs of the frequency response for the two Dayton 8 ohm woofers using the data given as numerals under the picture of each on Parts Express site . ____ What is the internal volume of your cabinet ? , at least 11 litre {0.4 cubic feet} I hope , ? ____ That Audax tweeter is good and worth keeping if you go to another woofer - and you can go back to your Pioneer tweeter if you are crossing over above 4kHz with the Goldwood woofer . Best hopes , Alan .
Hi RDV , I searched around to try to find a frequency response plot for the Goldwood woofer you bought , as no plot on the Parts Express site . Have a look at :- http://members.aol.com/_ht_a/Debertin/gold.htm , and which also appears at :- http://members.aol.com/Debertin/gold.htm . There you will see he shows a plot for the 4ohm version of that Goldwood driver . The 8ohm version will be similar but not identical - cheap drivers often have different upper-mid responses between 4ohm and 8ohm versions , BUT , the 8ohm if different will likely have less of that sharp peak at about 4.4kHz - and if so , good ! ___ What you bought is really a bass driver only , and not good for midrange - look at that drop from about 1.5kHz - that is part of what you are hearing . ___ OK , so $10- is a bargain , but you got somthing you will have to try to be clever with to make work - a good challenge to start speaker designing with ! ___ It can be got to work - it will not be ideal , but it can be better than you have got so far with it . ____ Do you have any way of measuring its frequency response - at least from about 1kHz to 8kHz {or slightly less} ? , if not then we'll guess its similar to the 4ohm version . If you have an Audio Oscillater , then connect it through your amplifier and slowly sweep up from about 2k and through the 4 to 5k region to beyond and listening at at moderate volume level because these are screech on the ear frequencies . Listen directly on-axis to the woofer . The human ear does NOT hear in flat response like a measuring microphone , but you should be able to hear a peak that is of much higher amplitude than the area on either side . Listen outdoors or in a dead sounding room , because resonances in a live room will confuse the issue . Find the high point of the upper mid peak by ear - this is where to cross-over that woofer - exactly on its peak . Use a 2nd order electrical - or a first order electrical {inductor only} plus a Zobel in parallel with the woofer . If you cannot find data to help and do not have an AC Voltage measuring meter that works accurately to at least 5kHz , the for a start point Zobel use an 8.2ohm , 10watt , resistor in series with a 10uF film capacitor . It may need larger capacitor - you can try adding caps in parallel with the 10uF to try 12.2uF and 14.7uF , etc .... Resistor should not be lower than the DC resistance of the Goldwood - so measure that . A 6.8ohm resistor may work better than 8.2ohm . Next connect the inductor - probably a .33mH or a .27mH - depending on the mid-range impedance you have achieved with your Zobel , and the exact location of that Peak . If the DC resistance of your Goldwood is greater than 6.8ohm , use 8.2ohm in the Zobel and .33mH ; if DC is less than 6.8 , use 6.8ohm and .27mH . This is electrical 1st order , but with the roll-off of the driver above the mid peak you will be getting approximately a 3rd order . To combine with tweeter you will have to experiment and listen - try Linkwitz-Riley 2nd order {-6dB at the crossover frequency} and basic 3rd order from charts {assume the impedance at crossover freq. of your tweeter is about 10% larger than its DC resistance}. Try both polarities connction with the tweeter . If you do not get OK sound then leave the Zobel on the woofer and try Linkwitz-Riley 2nd order electrical there also , and you can increase on tweeter from 3rd order to 4th order Link-Riley . ____ Post you results here if you would like further comments .____ Better would have been to buy a woofer after seeing its Response Plot . Draw the graphs of the frequency response for the two Dayton 8 ohm woofers using the data given as numerals under the picture of each on Parts Express site . ____ What is the internal volume of your cabinet ? , at least 11 litre {0.4 cubic feet} I hope , ? ____ That Audax tweeter is good and worth keeping if you go to another woofer - and you can go back to your Pioneer tweeter if you are crossing over above 4kHz with the Goldwood woofer . Best hopes , Alan .
The Audax tweeters were a "night & day" change. The speakers are very listenable now. Of course the woofers are still quite crappy but for now are going to have to suffice.
I suppose I could use some suggestions on a better woofer to match the Audax tweeters, perhaps the 6.5" Audax Aerogels? If I do that, I'll definitely buy crossover parts as well.
For now the speakers are sounding quite good though I'm having to lower the mids & highs on my mixer a great deal as there are no pad resistors and the tweeters are much better than the woofers now. It's not a big problem dialing in a good sound with my Mackie 1202VLZ.
The cabinets are around the same size as BR-1s, but the port is in front rather than the rear.
RDV
I suppose I could use some suggestions on a better woofer to match the Audax tweeters, perhaps the 6.5" Audax Aerogels? If I do that, I'll definitely buy crossover parts as well.
For now the speakers are sounding quite good though I'm having to lower the mids & highs on my mixer a great deal as there are no pad resistors and the tweeters are much better than the woofers now. It's not a big problem dialing in a good sound with my Mackie 1202VLZ.
The cabinets are around the same size as BR-1s, but the port is in front rather than the rear.
RDV
An "L" network - parallel plus series - 2 resistors around the Audax tweeter will achieve two things :- {1}- pad to match the woofer ;{2}- part damp down the Impedance peak at the tweeters Fs - which is interacting with the series capacitor-only crossover and giving audible output too high there {about 1.2kHz it looks to be on the published Audax plot} . Measure yiour Tweeters DC resistance and post here AND post how many dB - approximately - you want to reduce the tweeter by . I'll calculate a start point L-pad for you , and describe how to do it if you like , though you will need to understand Mathematics . ____ You may have to change the capacitor to the tweeter to allow for the impedance at the cross-over frequency , so post what crossover frequency you want to use , plus the capacitor you are currently using . Part of the excess mids is the peak in the woofer's response , and part is the capacitor effect at the tweeters Fs . Do you still have the .53mH and 10uF 2nd order on the woofer ?
Cabinet volume .
Oh , the BR-1 - is this the .55cuFt that Debertin mentions on his page ? , if so , that is reasonable as a start point for that Goldwood woofer if your cabinets are tightly sealed - no vents or ports {?}. If a region of the low bass is too prominent , that can be improved by making the cabinet behave as if it was larger . This is done by stuffing it - full is better than the 2/3 that Debertin mentions , but it must not be packed tight as then it becomes smaller to the woofer . Loose but physically stable packing is required . Wool batts cut to size or loose wool fill is best . Fibreglass or Rockwool is 2nd best , and can be good enough . Polyfill can be got to work to lesser effect . Post if you want to try this and I'll describe how to lay the filling - this is important as it will not work properly and can give worse sound if done flat side facing the backs of the woofer .
Oh , the BR-1 - is this the .55cuFt that Debertin mentions on his page ? , if so , that is reasonable as a start point for that Goldwood woofer if your cabinets are tightly sealed - no vents or ports {?}. If a region of the low bass is too prominent , that can be improved by making the cabinet behave as if it was larger . This is done by stuffing it - full is better than the 2/3 that Debertin mentions , but it must not be packed tight as then it becomes smaller to the woofer . Loose but physically stable packing is required . Wool batts cut to size or loose wool fill is best . Fibreglass or Rockwool is 2nd best , and can be good enough . Polyfill can be got to work to lesser effect . Post if you want to try this and I'll describe how to lay the filling - this is important as it will not work properly and can give worse sound if done flat side facing the backs of the woofer .
Yes. It's a mystery to me where that would have the woofer crossed at.alan-1-b said:
It's a 5.6µF Bi-Polar Electro on the tweeters. I'd like to cross them at somewhere between 2500 & 3000HZ considering that the woofer doesn't do mids too well.
A 5.6µF should be crossing it at around 3500 +/-10% so I was thinking of placing a 1µF film across it to bring it down to around 3000HZ. I was even thinking adding a small value bypass cap to hopefully improve the performance of the high-pass.
I'll put together a Zobel for the woofers tonight also as I have the right resistors for it on hand. Hopefully I can wring what performance is available from what I've got that way.
I'll also measure the DC resistance of the tweets and post them when I get home. Unfortunately that may be a bit late in England as you're 5 hours ahead of me.
RDV
the original crossover
If your 0053 inductor is a .53mH , then with the 10uF cap. it resonates at 2.184kHz and into an Impedance of 7.28ohm . This will not be the cross-over point you are getting because that Goldwood driver does not have an Impedance of 7.28ohm at 2k184 , but probably has an impedance of somewhere between 12 and 14 ohms at that frequency - and part of it is Reactive there - Inductive , owing to the voice coil winding , thus your crossover is now misterminated . I do not know the freq. it will be effectively operating at - it would have to be measured . Also the Q of that filter is 1 , which is neither of the commonly used Qs for L.C networks {usually Q is made to be .5 for Linkwitz-Riley filter , or , .7071 for Butterworth filter} , so perhaps that Q was chosen to specifically match the frequency characteristics of the original driver in that cabinet , or perhaps it was simply clumsy misunderstanding design . I'll post more later - have to go now .
If your 0053 inductor is a .53mH , then with the 10uF cap. it resonates at 2.184kHz and into an Impedance of 7.28ohm . This will not be the cross-over point you are getting because that Goldwood driver does not have an Impedance of 7.28ohm at 2k184 , but probably has an impedance of somewhere between 12 and 14 ohms at that frequency - and part of it is Reactive there - Inductive , owing to the voice coil winding , thus your crossover is now misterminated . I do not know the freq. it will be effectively operating at - it would have to be measured . Also the Q of that filter is 1 , which is neither of the commonly used Qs for L.C networks {usually Q is made to be .5 for Linkwitz-Riley filter , or , .7071 for Butterworth filter} , so perhaps that Q was chosen to specifically match the frequency characteristics of the original driver in that cabinet , or perhaps it was simply clumsy misunderstanding design . I'll post more later - have to go now .
alternate bass drivers .
Audax Aerogel drivers had a good reputation for clear midrange , but now the Company has stopped manufacturing them {typical !} . That particular 6 1/2 " available is a left-over . Looking at its data sheet it seems to have been designed for use as an upper-bass to lower-midrange driver for a 4-way system . It could be got to work as a bass-mid in a 2-way , but there would be a lack of low bass , and a problem in the upper mids - look at its data sheet - see the similar high-midrange spike there to what is in your Goldwood . This will require a more sophisticated cross-over than I can describe here . _____ For a bass-mid to use with your Audax tweeter , and with a simple cross-over I can suggest to you to start with that will give listenable results , better than with your Goldwood , I recommend the SEAS CA18RLY , available mail-order from Madisound . It will work in your cabinet if its internal volume is between 12 and 18 litres and sealed - no vents or ports for this SEAS driver , {and even more-so not for your Goldwood } if you want quality bass .
Audax Aerogel drivers had a good reputation for clear midrange , but now the Company has stopped manufacturing them {typical !} . That particular 6 1/2 " available is a left-over . Looking at its data sheet it seems to have been designed for use as an upper-bass to lower-midrange driver for a 4-way system . It could be got to work as a bass-mid in a 2-way , but there would be a lack of low bass , and a problem in the upper mids - look at its data sheet - see the similar high-midrange spike there to what is in your Goldwood . This will require a more sophisticated cross-over than I can describe here . _____ For a bass-mid to use with your Audax tweeter , and with a simple cross-over I can suggest to you to start with that will give listenable results , better than with your Goldwood , I recommend the SEAS CA18RLY , available mail-order from Madisound . It will work in your cabinet if its internal volume is between 12 and 18 litres and sealed - no vents or ports for this SEAS driver , {and even more-so not for your Goldwood } if you want quality bass .
to better use the Goldwood driver .
If you want to continue with the Goldwood driver , then simplest is to copy Debertin's cross-over , and use your Pioneer tweeter - post its DC resistance and I'll post suitable resistor values to use with it instead of the L-pad and 4ohm resistor that Debertin uses . I could suggest a similar cross-over alternate , but it may not work any better than his . ____ Keep your Audax for use with a better bass driver , such as the SEAS CA18RLY . ____ Another idea for the Goldwood is to use it with a 3" or 4" cone driver that can be crossed over where the Goldwood starts to drop - at approximately 1.3kHz . Of the numerous available , there is only sufficient data for a few {that I have seen so far} . One was sold by MCM Electronics - part number 54-606 - but it is not listed in their current on-line catalog , so you would have to find a pair - try e-bay , and ask in the "full-range forum" if anyone wants to sell a pair of MCM 54-606 , but don't pay a lot for them . ___ The Fostex FE107 , available mail-order from Madisound , will work in this application . With the MCM and the Fostex you will not need a tweeter , and I can suggest a simple cross-over for them to the Goldwood . But , both are open-back drivers , so you would need a separate sealed enclosure for either . You could build one to sit neatly on top of your cabinets . The MCM driver will need a 2 litre internal volume cabinet , and the Fostex will need a 6 litre internal volume , if either is to be used with a simple cross-over . ____ You could use the Goldwood + Pioneer + a midrange driver , but fot a simple cross-over the mid driver will need to have a low Qt of less than 0.5 and a low Fs - preferably below 100Hz , and a frequency response plot without any major peaks or dips up to at least 8kHz . Look at data sheets of 4" mini bass-mid drivers to find a suitable one . Again , you would need a separate sealed enclosure . None of the manufactured sealed midrange drivers that I know of can be used with a simple cross-over with your Goldwood driver .
If you want to continue with the Goldwood driver , then simplest is to copy Debertin's cross-over , and use your Pioneer tweeter - post its DC resistance and I'll post suitable resistor values to use with it instead of the L-pad and 4ohm resistor that Debertin uses . I could suggest a similar cross-over alternate , but it may not work any better than his . ____ Keep your Audax for use with a better bass driver , such as the SEAS CA18RLY . ____ Another idea for the Goldwood is to use it with a 3" or 4" cone driver that can be crossed over where the Goldwood starts to drop - at approximately 1.3kHz . Of the numerous available , there is only sufficient data for a few {that I have seen so far} . One was sold by MCM Electronics - part number 54-606 - but it is not listed in their current on-line catalog , so you would have to find a pair - try e-bay , and ask in the "full-range forum" if anyone wants to sell a pair of MCM 54-606 , but don't pay a lot for them . ___ The Fostex FE107 , available mail-order from Madisound , will work in this application . With the MCM and the Fostex you will not need a tweeter , and I can suggest a simple cross-over for them to the Goldwood . But , both are open-back drivers , so you would need a separate sealed enclosure for either . You could build one to sit neatly on top of your cabinets . The MCM driver will need a 2 litre internal volume cabinet , and the Fostex will need a 6 litre internal volume , if either is to be used with a simple cross-over . ____ You could use the Goldwood + Pioneer + a midrange driver , but fot a simple cross-over the mid driver will need to have a low Qt of less than 0.5 and a low Fs - preferably below 100Hz , and a frequency response plot without any major peaks or dips up to at least 8kHz . Look at data sheets of 4" mini bass-mid drivers to find a suitable one . Again , you would need a separate sealed enclosure . None of the manufactured sealed midrange drivers that I know of can be used with a simple cross-over with your Goldwood driver .
Other suggestions .
I strongly recommend that you buy a copy of the "Loudspeaker Design Cookbook" by Vance Dickerson . The new 7th edition is available from Madisound and Parts Express . You will find much in it to help you understand all the matters I have posted about , and more , so that you can design and modify your own loudspeakers , including on 3-way cross-overs . ____ Also , have someone measure your 0053 inductor to find out exactly what it is - this can be measured directly , or mesured with reference to a known value series conncted resistor at a known audio frequency , and the the inductance calculated - it is not difficult to do for a knowledgeable person .
I strongly recommend that you buy a copy of the "Loudspeaker Design Cookbook" by Vance Dickerson . The new 7th edition is available from Madisound and Parts Express . You will find much in it to help you understand all the matters I have posted about , and more , so that you can design and modify your own loudspeakers , including on 3-way cross-overs . ____ Also , have someone measure your 0053 inductor to find out exactly what it is - this can be measured directly , or mesured with reference to a known value series conncted resistor at a known audio frequency , and the the inductance calculated - it is not difficult to do for a knowledgeable person .
Something I perhaps forgot to mention is that I'm also using a DIY 12" subwoofer with A DIY electronic crossover crossed at 150HZ. I've got bass. I was just looking for more clarity from the near fields. The Audax tweeters were a great help as was adding a 1µF cap across the existing 5.6µF in each 1st order high-pass crossover bringing the tweeters down to around 3khz.
If I get a better mid-bass driver, I'll buy a decent commercial crossover.
Thanks for your help Alan.
RDV
If I get a better mid-bass driver, I'll buy a decent commercial crossover.
Thanks for your help Alan.
RDV
Alan:
I forgot to ask what your opinion is on Dayton's Standard 6.5" Woofer?
It looks as though I could use it in a vented enclosure(which I have).
Thanks in advance.
RDV
I forgot to ask what your opinion is on Dayton's Standard 6.5" Woofer?
It looks as though I could use it in a vented enclosure(which I have).
Thanks in advance.
RDV
Reading this article has convinced me I just made a mistake with both original drivers. They might sound alright together in a large sealed box with the right crossover, but for the box I put them in they were woefully inadequate.
I may build a pair of cabs similar to the ones in the above article and use those drivers in it, but I believe I'll be ordering a Dayton 6.5" standard woofer and the components to build a 12db linkwietz-riley 2nd order crossover for the cabinets that I have. I think the Audax tweeter will work fine in that configuration. I just need to know how much pad to use on the tweeter. I'll have a look at the BR-1 crossover for some ideas on that.
RDV
I may build a pair of cabs similar to the ones in the above article and use those drivers in it, but I believe I'll be ordering a Dayton 6.5" standard woofer and the components to build a 12db linkwietz-riley 2nd order crossover for the cabinets that I have. I think the Audax tweeter will work fine in that configuration. I just need to know how much pad to use on the tweeter. I'll have a look at the BR-1 crossover for some ideas on that.
RDV
Not the Dayton !
Well , you've realized that you made one mistake , now please consider that you are about to make another mistake . ____ Look at the frequency plot for that Dayton - first at the right-hand end - see that peak at about 3.25kHz - that will be an extended resonance - similar to the one in your Goldwood . You will hear it , and it will not be easy to lower it sufficiently with a cross-over . ___ Next see the sharp dip just below it - at about 2.8kHz - that will highlight the 3k25 resonant peak . ___ Next see the broader peak at about 2.5kHz - that is the highest frequency this driver can be usefully used to - thus you will need a steep crossover on the 3k25 peak , AND , you will need a steep crossover slope on your Audax tweeter to crossover it at 2k5 , because that is only 1 octave above its resonant frequency . This crossover cannot be bought from any catalog - it would have to be specifically made for this application . ___ The SEAS driver I recommended is much easier to cross to your Audax , and is a much better quality driver than the Dayton , and with the SEAS and the Audax tweeter you will have a system that will be equivalent in quality to a loudspeaker pair that would sell for around $500 ! , and you can fine tune it for better performance as you learn more about speakers . ___ I spelt that book Author's name incorrectly - it is Vance Dickason , I think ? , I'll have to check spelling . I'll post more next time .
Well , you've realized that you made one mistake , now please consider that you are about to make another mistake . ____ Look at the frequency plot for that Dayton - first at the right-hand end - see that peak at about 3.25kHz - that will be an extended resonance - similar to the one in your Goldwood . You will hear it , and it will not be easy to lower it sufficiently with a cross-over . ___ Next see the sharp dip just below it - at about 2.8kHz - that will highlight the 3k25 resonant peak . ___ Next see the broader peak at about 2.5kHz - that is the highest frequency this driver can be usefully used to - thus you will need a steep crossover on the 3k25 peak , AND , you will need a steep crossover slope on your Audax tweeter to crossover it at 2k5 , because that is only 1 octave above its resonant frequency . This crossover cannot be bought from any catalog - it would have to be specifically made for this application . ___ The SEAS driver I recommended is much easier to cross to your Audax , and is a much better quality driver than the Dayton , and with the SEAS and the Audax tweeter you will have a system that will be equivalent in quality to a loudspeaker pair that would sell for around $500 ! , and you can fine tune it for better performance as you learn more about speakers . ___ I spelt that book Author's name incorrectly - it is Vance Dickason , I think ? , I'll have to check spelling . I'll post more next time .
I've read really good things about this driver. They say it compares favorably to the Seas Aluminum speaker. The price certainly can't be beat. I know the response looks crazy from about 3500HZ on but I'll cross it around 3khz if I get it.
I'm really looking for a driver that will work in a vented box.
I'm not rushing into anything yet. I realize the Seas driver is a proven performer. I may go that route.
RDV
I'm really looking for a driver that will work in a vented box.
I'm not rushing into anything yet. I realize the Seas driver is a proven performer. I may go that route.
RDV
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.