Something I put together at lunchtime. Starting to look good. I don’t understand why X-Over pro is predicting a slight bump in 13khz region even with Zobel but I don’t think it will make a difference. The drivers natural roll off will compensate for it. If necessary, a notch (LCR) could be added later,,,,,, if it’s really audible.
From the graph it looks like Z-tags are also going to sound heavy from 1000hz down, but I don’t think this will be the case because of the eminent baffle step comp. In order to determine roughly and predict it, I would need the geometry and positioning of the drivers.
If it is indeed the case with a pair of Z-tags over powering the tweeter, then a simple L-pad will correct the problem. However, it has to be decided by you after the x-over is together and the drivers are installed.
BTW, what is the spacing between the center of the tweeter and the center of the lower woofer? It will actually help determine correct crossover frequency.
I don’t have enough knowledge and experience to design a 2.5 x-over and I am not a big fan anyhow. Attached picture is a predicted model based on available info and woofers, wired up in parallel. I’ve also changed from Butterworth to L-R type which gives slightly flatter frequency response.
Any other opinions?
I don’t know how to post a high res picture so there it is.
From the graph it looks like Z-tags are also going to sound heavy from 1000hz down, but I don’t think this will be the case because of the eminent baffle step comp. In order to determine roughly and predict it, I would need the geometry and positioning of the drivers.
If it is indeed the case with a pair of Z-tags over powering the tweeter, then a simple L-pad will correct the problem. However, it has to be decided by you after the x-over is together and the drivers are installed.
BTW, what is the spacing between the center of the tweeter and the center of the lower woofer? It will actually help determine correct crossover frequency.
I don’t have enough knowledge and experience to design a 2.5 x-over and I am not a big fan anyhow. Attached picture is a predicted model based on available info and woofers, wired up in parallel. I’ve also changed from Butterworth to L-R type which gives slightly flatter frequency response.
Any other opinions?
I don’t know how to post a high res picture so there it is.
Attachments
fantfool said:
Usually, the difference is only heard after you know there might be a difference.
Use reasonably good quality components without worrying about if they are good enough, or if the caps are in the right direction. Inductors should be low DCR.
I'm not familiar with the generic PE parts, as I use Solen caps (not the really expensive stuff). They are Canadian, so I don't get murdered with the shipping and the ridiculous cross-border brokerage fee that USP charges.
The improvements that high quality components give are smaller than the improvent from good crossover design/values. After building you may want to tweak by ear, and you may be disappointed if you want a different value than your expensve components. After you have settled you will be free to buy the expensive parts.
In other words, you could build a crossover using sand cast resistors, electrolytic capacitors and other things that people say are sub-standard, and still have a pleasing result.
I do feel that some improvements are had from using nicer capacitors and resistors.Some parts of the crossover, and some parts, wll produce less difference than others. Some will be totally worth it. I don't however, feel you should get carried away and blow your budget in this area.
In other words, you could build a crossover using sand cast resistors, electrolytic capacitors and other things that people say are sub-standard, and still have a pleasing result.
I do feel that some improvements are had from using nicer capacitors and resistors.Some parts of the crossover, and some parts, wll produce less difference than others. Some will be totally worth it. I don't however, feel you should get carried away and blow your budget in this area.
Sorry I didn’t make myself clear. I need a distance between the center of the tweeter to the near woofers center. I also need complete dimensions of the existing boxes (actually volume will be enough) but I’d need a size of the front baffle to figure out compensation.
Considering the quality of the components.
Resistors: some people claim audible differences.
I could never hear a difference between an inexpensive sandcast wirewound and more expensive non-inductive resistors. I prefer Mills in my crossovers just for the looks and higher W rating then other non-inductive resistors. Not a $$$$$ deal anyway.
The fact of non-inductance technically makes your x-over more accurate. How much more? Don’t know. I don’t believe it’s that important.
Capacitors: Again, some people claim huge differences.
I remember building a 2 way x-over with electrolytics and then replacing all electrolytics with metallized poly. I THOUGHT I heard a difference. Notice the word “thought”. Anyhow, it is considered by many professional x-over designers (to my knowledge) that electrolytic capacitors don’t behave well on higher frequencies. Electrolytic capacitors also have a tendency not to age as well as poly.
To make a long story short, get something like Solen or inexpensive Jensen and you will be fine. Neither will break bank.
Inductors.
I prefer to use 14 gage on the woofers and 18 gage on the tweeter. Bigger the gage, lower the DCR. If you want to save a couple of $ you could get a Iron core inductor. Some people say that Air core is the best, I don’t have an opinion because air core is only type that I’ve used in parallel with the woofers.
As JNB said, design itself is more important then exotic x-over parts.
Considering the quality of the components.
Resistors: some people claim audible differences.
I could never hear a difference between an inexpensive sandcast wirewound and more expensive non-inductive resistors. I prefer Mills in my crossovers just for the looks and higher W rating then other non-inductive resistors. Not a $$$$$ deal anyway.
The fact of non-inductance technically makes your x-over more accurate. How much more? Don’t know. I don’t believe it’s that important.
Capacitors: Again, some people claim huge differences.
I remember building a 2 way x-over with electrolytics and then replacing all electrolytics with metallized poly. I THOUGHT I heard a difference. Notice the word “thought”. Anyhow, it is considered by many professional x-over designers (to my knowledge) that electrolytic capacitors don’t behave well on higher frequencies. Electrolytic capacitors also have a tendency not to age as well as poly.
To make a long story short, get something like Solen or inexpensive Jensen and you will be fine. Neither will break bank.
Inductors.
I prefer to use 14 gage on the woofers and 18 gage on the tweeter. Bigger the gage, lower the DCR. If you want to save a couple of $ you could get a Iron core inductor. Some people say that Air core is the best, I don’t have an opinion because air core is only type that I’ve used in parallel with the woofers.
As JNB said, design itself is more important then exotic x-over parts.
From center of the tweeter to the center of the first woofer is 8". The box volume about 1.65 cu/ft. Front baffle is about 39" but the last 12" are part of a chamber for ballast, the speaker part of it is about 27". Let me know if you need anything else.
I have been playing around with WINISD now that I have all of the Zetag Specs, pretty interesting, wish I would have had this 2 years ago.
I have been playing around with WINISD now that I have all of the Zetag Specs, pretty interesting, wish I would have had this 2 years ago.
I can email you a hi-res shot but it’s still a very raw version of x-over. I modeled it on the unshielded version of the tweeter. BTW the bump at 13khz is caused by a Zobel. It evened out response at the x-over but added that small hump. More playing around is due.
BSC will start around 1700hz and my guess is that it will gradually lower frequency response by 3db. Therefore, I wouldn’t worry about efficiency of the tweeter.
BSC will start around 1700hz and my guess is that it will gradually lower frequency response by 3db. Therefore, I wouldn’t worry about efficiency of the tweeter.
The tweeter I have is the TDFC unshielded version.
Do you think that the 2.5 way will ultamately sound better then the 2 way?
The only question I have on the PE x-over design is the dip around 1K and the rolloff at 100 and 10K. Does this look normal for those drivers and should I not be concerned about performance at either end of the spectrum?
Do you think that the 2.5 way will ultamately sound better then the 2 way?
The only question I have on the PE x-over design is the dip around 1K and the rolloff at 100 and 10K. Does this look normal for those drivers and should I not be concerned about performance at either end of the spectrum?
These effects are not huge but I would have done it differently. A dip after 1k is not unusual but that amount and that shape are not typical IMO. It looks as though you would have to sacrifice some overall sensitivity to help the response at the extremes. As it stands it may be OK closer to a wall.
fantfool said:
Hi,
My personal preference is for 2.5 way if using 2 of the same midwoofers. For practical reasons this can also reduce the chance of midrange lobing - the drivers spacing is important but not critical.
The FR chart for the combined response of the drivers using that crossover is just a simulation, not meant to represent the actual response.
Dips (or peaks) of 3-6 db in the response is normal.
MJL21193 said:
Ok, so how would you go about calculation of impedance curve of the lower woofer and it’s effect on the crossover performance? Also, Since BSC starts at 1700hz, would it make sense to cross it over higher? Do you design the x-over as a 2-way with Zobel and then add secondary woofer with coil and hope for the best?
I understand the general concept; I just don’t know how to go about calculations. I think it’s an interesting opportunity here. Would it be possible to come up with a good x-over design without actually measuring response of the speaker?
I have no idea if 2.5 are better then MTM.
BTW, did anyone notice that PE guy used Chebichev and Butterworth x-overs together and different x-over frequencies? Lobbing and tilting of Of-axis response will probably be quite big and then some.
I don’t have a problem with crossing over 2 drivers at different points, done it a couple of times (am I going to get crucified now?)
I would worry about a deep at a 1000 but not so much about gradual roll-off of the tweeter. JNB has a point there about efficiency and ruler flat FR.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.