So.......
At this stage I will stick in more familiar waters and rewalk the BR route.
So while I have nothing great to listen to at the moment, I will fast track another pair of floorstanders. I will try to get them firing in the next few weeks then let the years of tweaking begin! While this is all taking place, I plan to invest in the nesessary testing gear, and bringing myself up to date with algebra, and TL theory among other things.
Promise i'll get it right(er) this time!
Mick.
At this stage I will stick in more familiar waters and rewalk the BR route.
So while I have nothing great to listen to at the moment, I will fast track another pair of floorstanders. I will try to get them firing in the next few weeks then let the years of tweaking begin! While this is all taking place, I plan to invest in the nesessary testing gear, and bringing myself up to date with algebra, and TL theory among other things.
Promise i'll get it right(er) this time!
Mick.
TL
TLs are very forgiving about design, much more than bass reflex. I built a 6 ft. taped line with a Dayton RS270S-8 10. It is pretty simular to the Tritrix line but is 12" wide with a little longer line, done by increasing the length of the back baffle. The enclosure volume is about 3 cubic feet. It is 3 dB down at 40 Hz. It sounds better than an Eton 10" in a 100 pound bass reflex. There are dozens of TL designs on the web and most are pretty forgiving of the driver used.
http://www.parts-express.com/projectshowcase/indexn.cfm?project=Tritrix
http://www.flickr.com/photos/7567821@N05/?saved=1
TLs are very forgiving about design, much more than bass reflex. I built a 6 ft. taped line with a Dayton RS270S-8 10. It is pretty simular to the Tritrix line but is 12" wide with a little longer line, done by increasing the length of the back baffle. The enclosure volume is about 3 cubic feet. It is 3 dB down at 40 Hz. It sounds better than an Eton 10" in a 100 pound bass reflex. There are dozens of TL designs on the web and most are pretty forgiving of the driver used.
http://www.parts-express.com/projectshowcase/indexn.cfm?project=Tritrix
http://www.flickr.com/photos/7567821@N05/?saved=1
Finished TL withe Zebra wood veneer
Finished TL mounted wth spikes to floor. Sideways mounting loads driver with 2 pi steradians floor loading for better efficiency. One of the spikes is closely coupled to the woofer mounting frame with short stiff bracing to minimize motion of the woofer motor. This can be achived with high enclosure mass or coupling to the concrete floor with spikes. My back doesn't like moving 100 pound plus speakers...
Finished TL mounted wth spikes to floor. Sideways mounting loads driver with 2 pi steradians floor loading for better efficiency. One of the spikes is closely coupled to the woofer mounting frame with short stiff bracing to minimize motion of the woofer motor. This can be achived with high enclosure mass or coupling to the concrete floor with spikes. My back doesn't like moving 100 pound plus speakers...
Attachments
Thanx CosmoA,
Thats a very nice looking speaker, is it fullrange?
I have enough material (I think) to put together a test TL, I think this will give me a better understanding of the concept in general. However I'm not at all sure on which type of TL would give me a good foot hold. Based on the driver on post #1 (I have 6 of these and nothing else!), could you suggest a good starting point?
Thanx Mick.
Thats a very nice looking speaker, is it fullrange?
I have enough material (I think) to put together a test TL, I think this will give me a better understanding of the concept in general. However I'm not at all sure on which type of TL would give me a good foot hold. Based on the driver on post #1 (I have 6 of these and nothing else!), could you suggest a good starting point?
Thanx Mick.
G'day,
Homework seems to be paying off (I think/hope) WRT TL & Qts.
I got it oh so wrong earlier in this thread, I'm thinking its a simple case of Qms, Qes and their 'relative' ratios, rather than Qts itself being the defining factor.
ie Qms = 12, Qes = .276, = Qts .27
Qms = 2.2, Qes = .31, = Qts .27
Qts for both drivers are the same, but the way they act in a given TL will be very different. Am I on the right track this time?
Thanx, Mick.
Homework seems to be paying off (I think/hope) WRT TL & Qts.
I got it oh so wrong earlier in this thread, I'm thinking its a simple case of Qms, Qes and their 'relative' ratios, rather than Qts itself being the defining factor.
ie Qms = 12, Qes = .276, = Qts .27
Qms = 2.2, Qes = .31, = Qts .27
Qts for both drivers are the same, but the way they act in a given TL will be very different. Am I on the right track this time?
Thanx, Mick.
Greets!
You're welcome!
So get unconfused, study up on basic pipe/horn theory here first if MJK's site is too tough a read, then with that background all his info should be a 'walk in the park' (skip over the math if you want to).
OK, you still haven't posted the cab's internal dims/woofer cutout location and I assume the vent is attached to the bottom plate (can't get the thumbnails to open). Regardless, I see nothing to indicate a ~80 Hz dip other than possibly an air leak that will cause a suck-out (had this happen the other day) and with it simming a max flat alignment, such a small cab needs some form of BSC filter to tonally balance it down to ~40 Hz if not a near-field app. This will cover most recorded music if the dip can be be rectified.
GM
Why? With all the info available today there's no reason why you can't have a good performing TL system without learning anything beyond how to use simming software. I mean if you want DIY speaker design as a long term hobby, then learning the theory, etc. is beneficial, but hardly a prerequisite these days.
Who knows? Converting to a TL may remove the mysterious dip.
GM
G'day GM,
Amused, bemused AND still somewhat confused!!!! Ever tried typing whilst laughing full heartedly? It looks like this......sfu9p ibn no izs ui iv yyoiu!!!!!!!
A blind man with eyes peeled, I feel like!! with a twist of yoda!
Anyhooo, The cabs int. dimensions are 200x320x420 and the port is attached to the bottom. The port is made from 12 mm mdf and is actually 195mm long internally +25mm cab wall thickness (sorry, I think I quoted 190 total in a previous thread). The bass driver was centre and centred 210mm from external bottom (185 internal). Both channels seemed to have the same problem and talcam powder was used to find any leaks, although none were evident. Is this a sound method for finding leaks?
A bsc was used, however it didn't seem to do much (if anything to the dip) except tone down and clean up the midrange, ....seemed to intergrate the whole sound alot better.
Not at all sure on 'near-field app' to try that approach.
Sorry, not sure whats going on with the thumbnails.
Thanx again, Mick!
Amused, bemused AND still somewhat confused!!!! Ever tried typing whilst laughing full heartedly? It looks like this......sfu9p ibn no izs ui iv yyoiu!!!!!!!
A blind man with eyes peeled, I feel like!! with a twist of yoda!
Anyhooo, The cabs int. dimensions are 200x320x420 and the port is attached to the bottom. The port is made from 12 mm mdf and is actually 195mm long internally +25mm cab wall thickness (sorry, I think I quoted 190 total in a previous thread). The bass driver was centre and centred 210mm from external bottom (185 internal). Both channels seemed to have the same problem and talcam powder was used to find any leaks, although none were evident. Is this a sound method for finding leaks?
A bsc was used, however it didn't seem to do much (if anything to the dip) except tone down and clean up the midrange, ....seemed to intergrate the whole sound alot better.
Not at all sure on 'near-field app' to try that approach.
Sorry, not sure whats going on with the thumbnails.
Thanx again, Mick!
So get unconfused, study up on basic pipe/horn theory first if MJK's site is too tough a read, then with that background all his info should be a 'walk in the park' (skip over the math if you want to).
Your just showing off now!!!!...........HE,HE,HE!!!!!
Last time I walked in the park, I stood in doggy doo doo, got mugged & then arrested and sent to timeout for skipping math!!!!!!! and you know it!!!!!!
And here's me rolling in the park!
Just light hearted fun. But in seriousness I might just build a BR (cause I luv cuttin' up mdf anyway) and a 'practice' TL to tinker with. Just so I don't lose interest. BRs and trying figure xovers has kept me interested and amused this long, I'm sure they'll teach me a lot yet!!!
Thanx again, Mick.
Your just showing off now!!!!
...........HE,HE,HE!!!!!Last time I walked in the park, I stood in doggy doo doo, got mugged & then arrested and sent to timeout for skipping math!!!!!!!
and you know it!!!!!!And here's me rolling in the park!
Just light hearted fun. But in seriousness I might just build a BR (cause I luv cuttin' up mdf anyway) and a 'practice' TL to tinker with. Just so I don't lose interest. BRs and trying figure xovers has kept me interested and amused this long, I'm sure they'll teach me a lot yet!!!
Thanx again, Mick.
Hi Mick, couple of comments. I don't think a side firing woofer was probably a great idea if it was playing up to 600Hz especially if it is a 1st order crossover! way to much into your midrange to not be being fired forward with everything else.
2nd would you be able to post the internal dimentions of your cabinets (the tall ones) and the port spec. It would be interesting to model it in unibox. I did the 23L and it seems to be not bad at all. Only thing I noticed was that one of the impedance peaks was around 75Hz Not sure if that has any bearing on things, it has been four or more years since I read the theory on BR and I can't remember, but I did think that there was something that would cause dips...
The really odd thing about your latest box is that if it really is 30L then provided your port is right that should be pretty much optimal for a single one of these drivers... your comment about the shape may have something to it, you might be getting pipe resonances.... but not sure how much of a problem that would likely be.
Tony.
2nd would you be able to post the internal dimentions of your cabinets (the tall ones) and the port spec. It would be interesting to model it in unibox. I did the 23L and it seems to be not bad at all. Only thing I noticed was that one of the impedance peaks was around 75Hz Not sure if that has any bearing on things, it has been four or more years since I read the theory on BR and I can't remember, but I did think that there was something that would cause dips...
The really odd thing about your latest box is that if it really is 30L then provided your port is right that should be pretty much optimal for a single one of these drivers... your comment about the shape may have something to it, you might be getting pipe resonances.... but not sure how much of a problem that would likely be.
Tony.
Attachments
Hi Tony,
Geez, I gotta say, that impedence peak is sitting right there isn't it!
I agree that there may be somthing in that, almost certainly, its awfully sharp, and bang where I was hearing the problem!
The 30 ltr box internally measures h990xw150xd220 and the port measures 80x50x145, I agree that the side firing idea was not a good 1, huge bsc troubles i think is the biggest
trouble here, although the high xover point is also a big problem. I built these on a whim,
and was doubtful of their success from square 1, which is why they are not long for this earth.
But again that impedence line is scarily close to where I was hearing the problem!
Far too close to be a mere coincidence I feel.
I'm interested to see if the 30 ltr box shows the same problem?
I think if there is the bad Baffle step situation is hiding it well.
If so could you help me with an optimum box/port ratio for this driver
There is a problem with my pc (windows) at the moment and I can't open zip files (ie. unibox).
I'm guessing, (cause i only work by ear) that a rising impedence could well be the cause (as Dean also suspected),
because hooking up my zobel does seem to lift and clear the mids with this driver, so it seems logical to me that it could well be the cause! GREAT WORK!
Well done and thanx aplenty, Mick.
Geez, I gotta say, that impedence peak is sitting right there isn't it!
I agree that there may be somthing in that, almost certainly, its awfully sharp, and bang where I was hearing the problem!
The 30 ltr box internally measures h990xw150xd220 and the port measures 80x50x145, I agree that the side firing idea was not a good 1, huge bsc troubles i think is the biggest
trouble here, although the high xover point is also a big problem. I built these on a whim,
and was doubtful of their success from square 1, which is why they are not long for this earth.
But again that impedence line is scarily close to where I was hearing the problem!
Far too close to be a mere coincidence I feel.
I'm interested to see if the 30 ltr box shows the same problem?
I think if there is the bad Baffle step situation is hiding it well.
If so could you help me with an optimum box/port ratio for this driver
There is a problem with my pc (windows) at the moment and I can't open zip files
(ie. unibox).I'm guessing, (cause i only work by ear) that a rising impedence could well be the cause (as Dean also suspected),
because hooking up my zobel does seem to lift and clear the mids with this driver, so it seems logical to me that it could well be the cause! GREAT WORK!
Well done and thanx aplenty, Mick.
yep it does plot the theoretical SPL but no effect shown for the impedance peaks... the current for the amp is interesting though, and may be more telling, ie due to the higher impedance the current at that freq is quite low, as power = Volt's X Amps I guess it stands to reason that you will have a dip at the impedance peak. I'll have to read up on BR again (as this could well be complete rubbish), This may be one of the tradeoffs (much poorer transient response was the main one I could remember).
The zobel shouldn't affect this particular impedance peak. From memory it is due to your box tuning. Bass reflex has two impedance peaks not one.
One thing I did think of this morning though was the positioning of your port in the 23L boxes. Since it is on the floor of the box rear facing and directly behind the woofer you may be getting some sort of diffraction or other effect I noted some quite marked effects in woofer response (sealed box) in nearfield measurements just from having a 1" sqaure brace directly behind the woofer, this was at around 700Hz but removing the brace removed the anomaly completely. I wouldn't rule out the port being the culprit. Do a search on port placement, I think I vaugly remember something about NOT putting it directly behind the woofer.
Tony.
Attachments
Greets!
You're welcome!
OK, using MJK's software which is the most accurate I have, cab tuning is right at 50 Hz and has strong vent pipe harmonics out in the 700-900 Hz BW, though MJK has noted that the accuracy falls off above 500 Hz, so could be insignificant. Bottom line, I don't see anything in the cab design that could create a ~80 Hz dip.
Never tried talcum powder, just the traditional soapy water with the vent plugged if applicable. Did you plug the vents?
No, BSC shouldn't have affected any dip, just increased tonal balance to a lower frequency, which it did.
'Near-field app' just refers to sitting close to the speakers, like at your computer. If the dip was in anyway room related it would disappear, but from your comments I guess it's not.
Me neither, but apparently the problem's been sorted out as I see them correctly now with a zoom feature.
GM
Your just showing off now!!!!
???? Because I tried to help you?
GM
Yeah, not a good plan. Normally you want to limit the distance 'around the bend' to 1 WL of the XO's HF -12 dB point.
Unfortunately, Unibox can't even model the 23 L box fairly accurately since it assumes a ~uniform particle density in the cab, i.e. an ideal Helmholtz resonator, so no hope of accurately modeling a tower design (aka MLTL) that has some 1/4 WL pipe resonances affecting vent tuning.
The impedance peaks can affect response if the amp can't deal with them, such as a high output impedance tube amp that tracks the impedance peaks/dips or if it has a weak power supply such as the entry level 'T' amp that dips at a high impedance, but since more than one amp has been used with the dip still present I don't see how it could be the problem.
GM
Attachments
Thanx again Tony,
I now have unibox up and running, (I found my office pro 2003 repair disc), although it may take a day or two to learn how to drive it.
I'm not sure this peak is the one the zobel is aimed at, as I understand it, the zobel is aimed at the rising impedence that occurs as frequency rises, (further to the right of the graph).
What your saying is making a lot of sense, and the 'mysterious dip' may not be all that mysterious after all! (At least I know my ears arn't crazy!).
I think I have also read that BRs are meant to have two peaks, but this is the first time I have actually heard it, if that is its cause.
Could it be a combination of port placement/shape + the high impedence causing the spl drop THAT much.
At least it is an avenue to investigate further.
If anybody could shed some more light on this, My ears are pricked!!
Thanx again, Mick!
I now have unibox up and running, (I found my office pro 2003 repair disc), although it may take a day or two to learn how to drive it.
I'm not sure this peak is the one the zobel is aimed at, as I understand it, the zobel is aimed at the rising impedence that occurs as frequency rises, (further to the right of the graph)
.What your saying is making a lot of sense, and the 'mysterious dip' may not be all that mysterious after all! (At least I know my ears arn't crazy!).
I think I have also read that BRs are meant to have two peaks, but this is the first time I have actually heard it, if that is its cause.
Could it be a combination of port placement/shape + the high impedence causing the spl drop THAT much.
At least it is an avenue to investigate further.
If anybody could shed some more light on this, My ears are pricked!!
Thanx again, Mick!
GM,
Nah, Not 'cause you tried to help, but just because i'm a cheeky ferrethead, givin' you a bit of a ribbing is all! As I said 'light hearted fun'. Its in my nature. As its in your nature to be incredibly direct and straight shooting.
I only plugged the ports with foam, although it was pushed in fairly tight, mightn't have been tight enough. Soapy water, sounds like a good method.
Do you think it is probable that both speakers would have a leak that would cause the problem?
Anyhow They have been retired, they've had their 12 months of working time,
If the next pair do the same thing, then I will start worrying. Never really liked the sound of a Two way with an 8 inch driver anyhow!
And besides I love building speakers!
I'll get 'em right next time, (Don't we all say that?). Honestly, has anyone, ever been happy (I mean completely) with their finished work? I strongly doubt it, that it's a game of compromise and trade offs, is exactly what makes it so much fun!
Thanx again GM, Mick.
Nah, Not 'cause you tried to help, but just because i'm a cheeky ferrethead, givin' you a bit of a ribbing is all! As I said 'light hearted fun'. Its in my nature. As its in your nature to be incredibly direct and straight shooting.
I only plugged the ports with foam, although it was pushed in fairly tight, mightn't have been tight enough. Soapy water, sounds like a good method.
Do you think it is probable that both speakers would have a leak that would cause the problem?
Anyhow They have been retired, they've had their 12 months of working time,
If the next pair do the same thing, then I will start worrying. Never really liked the sound of a Two way with an 8 inch driver anyhow!
And besides I love building speakers!
I'll get 'em right next time, (Don't we all say that?). Honestly, has anyone, ever been happy (I mean completely) with their finished work? I strongly doubt it, that it's a game of compromise and trade offs, is exactly what makes it so much fun!
Thanx again GM, Mick.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.