TMWW... First DIY project

[WernerM]

I think so too, and I may upgrade later on, I know it means remodeling, but if I am not completely satsfied, I will be up to the task. I have considered modeling it with the Vifa XT25TG30 dual ring radiator tweeter, and I may still do so if costs and benefits outweigh what I currently have. That tweeter looks to have a very good off axis reponse also. May take you up on your offer too, depending on if it will still be available... Plenty of time still to figure it out.

[arlis_1957@yahoo.com]

damn marc. you have just zoomed by so many in a very short time. cant wait to see these finished.

[rpb]

I think so too, and I may upgrade later on, I know it means remodeling, but if I am not completely satsfied, I will be up to the task. I have considered modeling it with the Vifa XT25TG30 dual ring radiator tweeter, and I may still do so if costs and benefits outweigh what I currently have. That tweeter looks to have a very good off axis reponse also. May take you up on your offer too, depending on if it will still be available... Plenty of time still to figure it out.

I really like this tweeter. I'm currently using it crossed around 2.5k if my memory is correct. Since it's a horn , the acoustic center is back a little. I like it a lot.
http://www.parts-express.com/pe/show...umber=264-1028

[Pete Schumacher ®]

I think so too, and I may upgrade later on, I know it means remodeling, but if I am not completely satsfied, I will be up to the task. I have considered modeling it with the Vifa XT25TG30 dual ring radiator tweeter, and I may still do so if costs and benefits outweigh what I currently have. That tweeter looks to have a very good off axis reponse also. May take you up on your offer too, depending on if it will still be available... Plenty of time still to figure it out.

With that crossover setup, you could get the XT19 to work well, and have even better off axis than the XT25.

[rpb]

Thanks for the vote of confidence, does this mean you think I will still get good imaging with the drivers chosen? GRS-8PR8, Vifa TC9FD, Goldwood GT-510. I hope so as the goal from the beginning was to beat retail speakers in the same $400-$500/pair range. Right now if I were to buy the crossover components today ($135), I will have a total of $275 invested and would only be lacking damping material, solder, bead board for crossovers, and screws to mount drivers, and build cabs. So figure about another $30 in materials, and I will have speakers ready to listen to, but unfinished. So just over $300 invested, leaves me around $100-$200 for finishing materials. Not too bad!

To add to what Don said, I think imaging is best achieved when the speaker setup in the room is symetrical, and when the frequency response of the two speakers are very closely matched. I think it is also improved when the speakers are located a few extra feet from the walls if the room allows for such a placement.

[WernerM]

With that crossover setup, you could get the XT19 to work well, and have even better off axis than the XT25.

You have intrigued me Pete, may have to look into that tonight after work.

[WernerM]

To add to what Don said, I think imaging is best achieved when the speaker setup in the room is symetrical, and when the frequency response of the two speakers are very closely matched. I think it is also improved when the speakers are located a few extra feet from the walls if the room allows for such a placement.

Right, which is why I am going to order the extra resistors and such as well. My setup is going to be a little quircky, but I can make it work. If not, then I will just have to move my living room!!! LOL...

[rpb]

Here's something else you should sim. After seeing that your summed response in the midrange is louder than the mid alone, I speculate that you can put the woofers in parallel.
So if you leave the mid and tweeter as they are, but with less padding maybe, you could make a couple minor changes to the woofer x-over and
have a more sensitive design. It may not be ideal for your amp, but if someone else decides to build the design, they might prefer the higher sensitivity.

The other thing I would sim, is a higher tweeter x-over point. I think that the mid you selected is very good even up high. The Goldwood tweeter is an unknown, and
a 2k to 2.5k x-over is pushing it even for many tweeters with a good reputation.

[WernerM]

Here's something else you should sim. After seeing that your summed response in the midrange is louder than the mid alone, I speculate that you can put the woofers in parallel.
So if you leave the mid and tweeter as they are, but with less padding maybe, you could make a couple minor changes to the woofer x-over and
have a more sensitive design. It may not be ideal for your amp, but if someone else decides to build the design, they might prefer the higher sensitivity.

The other thing I would sim, is a higher tweeter x-over point. I think that the mid you selected is very good even up high. The Goldwood tweeter is an unknown, and
a 2k to 2.5k x-over is pushing it even for many tweeters with a good reputation.

I tried simming in parallel, and there is a dip in the low end when I do that. I also tried simming a higher crossover point, but after many hours, I can't get past the 3-4dB dip in the 2KHz region of the mid. Now I'm not saying it isn't possible, just that I still don't fully understand what filters do what to compensate for it. I have basically just been using trial and error, and throwing crap at it from time to time to see what happens. This morning, I got a pm from Wolf suggesting to try a lower mH coil onthe tweeter, he said that a 1mH coil on a tweeter is basically a boat anchor. So I have resimmed again with a much lower value coil. Let me know if this will work. There are still plenty of things that you guys know which I currently don't, that you can pick out pretty quick. I wish there were some sort of suggested guidlines along with what Wolf has already created for ideal limits in crossover values on different drivers. And maybe some explanations on how certain filters work, I do like the tabs in PCD that give you calculators for several things, those have come in handy, but a good explanation in laymans terms of when series RLC's, RC's, RL's, etc., before and after crossovers, are suggested would be of assistance. The problem with the speaker builders bible thread is that several of the links don't work. I have tried contacting the person in charge of the thread, but haven't heard anything back. Anyway, here are the results of my last sim...

[WernerM]

damn marc. you have just zoomed by so many in a very short time. cant wait to see these finished.

I appreciate that, but there is still alot about a crossover I don't know, if it weren't for PCD there is no way I could even have come up with this.

[rpb]

There's nothing wrong with using a 1 mH coil if it gets the job done. Your other sim looked better. I hope you are renaming, or saving the different sims. What if you went back to your diffraction program, and see if moving the mid will flatten out that dip.

Here's the real thing. No dip!



Also here's Zaphs measurements. Theres not much of a dip in it as well. Maybe double check your trace, or the baffle effects. If it's the baffle, try putting the mid above the tweeter.

[WernerM]

Okay, I'll give it a shot. Thanks for the suggestion RPB. If I lower the mid 2" and the tweeter 2", and offset them 1" from center, it does lower the BDS about 1.5dB. You are right that it doesn't dip, what I am seeing as a dip above 2KHz is actually not a dip, but a rise in the 1KHz below region due to baffle diffraction. Now the tweeter will be at 34" instead of 36", and the mid will be at 30" instead of 32". For me this is better anyway, as my ear height is at 31"-33" depending on how far I recline...LOL...

[Pete Schumacher ®]

I'll guarantee that dip is diffraction related.

[WernerM]

Actually if I move the mid higher on the bafle, the diffraction is worse. So I am not sure it will work for my situation. Thanks for the thought. I just tried to remodel my graphs using the baffle diffraction and room boundary simulator, and wow does it look like crap. I'm not even sure anymore if my frd files are any good. I think I'm going to step away from this until after the holiday. My problem is I don't know which measurement data and graphs to go with. Zaphs graph is 15dB higher in SPL than the manufacturers graph. I know the manufacturers is at 2.83V/m and maybe Zaph's isn't, but if Zaphs is at 1W/1m, then shouldn't the SPL be lower in dB? I keep thinking that I'm on the right track, but maybe I'm not. The last sets of sims I did, I lowered the dB on the SPLCopy for the mid to match the manufacturers, and I am pretty sure now that this is wrong. All has not been lost, because atleast I am learning, but I definitely don't feel comfortable ordering parts anymore. For the GRS graph I copied the graph that Don sent me after measuring, and I adjusted the level at 2KHz to what PE says the SPL is, as per his advice. The Goldwood tweeter, I did the same way. So I guess what I need to know is what is the next step? Do I leave the SPL set to what Zaph shows, and just redo the Vifa mid sims, or are they okay? Are the GRS and Goldwood sims okay based on what I have currently done? What should I do?

[rpb]

Actually if I move the mid higher on the bafle, the diffraction is worse. So I am not sure it will work for my situation. Thanks for the thought. I just tried to remodel my graphs using the baffle diffraction and room boundary simulator, and wow does it look like crap. I'm not even sure anymore if my frd files are any good. I think I'm going to step away from this until after the holiday. My problem is I don't know which measurement data and graphs to go with. Zaphs graph is 15dB higher in SPL than the manufacturers graph. I know the manufacturers is at 2.83V/m and maybe Zaph's isn't, but if Zaphs is at 1W/1m, then shouldn't the SPL be lower in dB? I keep thinking that I'm on the right track, but maybe I'm not. The last sets of sims I did, I lowered the dB on the SPLCopy for the mid to match the manufacturers, and I am pretty sure now that this is wrong. All has not been lost, because atleast I am learning, but I definitely don't feel comfortable ordering parts anymore. For the GRS graph I copied the graph that Don sent me after measuring, and I adjusted the level at 2KHz to what PE says the SPL is, as per his advice. The Goldwood tweeter, I did the same way. So I guess what I need to know is what is the next step? Do I leave the SPL set to what Zaph shows, and just redo the Vifa mid sims, or are they okay? Are the GRS and Goldwood sims okay based on what I have currently done? What should I do?

Absolute spl doesn't matter all that much. Think about it. You made a sim raising the mid and tweeter several db with simple resistor changes. That's all you would have to do if it sounds off when you build it. Some tuning by ear is needed anyway. Take a few days off, and enjoy the holiday.

[duanebro]

Actually if I move the mid higher on the bafle, the diffraction is worse. So I am not sure it will work for my situation. Thanks for the thought. I just tried to remodel my graphs using the baffle diffraction and room boundary simulator, and wow does it look like crap. I'm not even sure anymore if my frd files are any good. I think I'm going to step away from this until after the holiday. My problem is I don't know which measurement data and graphs to go with. Zaphs graph is 15dB higher in SPL than the manufacturers graph. I know the manufacturers is at 2.83V/m and maybe Zaph's isn't, but if Zaphs is at 1W/1m, then shouldn't the SPL be lower in dB? I keep thinking that I'm on the right track, but maybe I'm not. The last sets of sims I did, I lowered the dB on the SPLCopy for the mid to match the manufacturers, and I am pretty sure now that this is wrong. All has not been lost, because atleast I am learning, but I definitely don't feel comfortable ordering parts anymore. For the GRS graph I copied the graph that Don sent me after measuring, and I adjusted the level at 2KHz to what PE says the SPL is, as per his advice. The Goldwood tweeter, I did the same way. So I guess what I need to know is what is the next step? Do I leave the SPL set to what Zaph shows, and just redo the Vifa mid sims, or are they okay? Are the GRS and Goldwood sims okay based on what I have currently done? What should I do?

Wow, 17 pages, this has got to be some kind of record. LOL

If you are resetting the spl levels to their rated value then you need to do that with all the drivers.

I just started a new job and haven't had time to do anything speaker related.

[rpb]

Wow, 17 pages, this has got to be some kind of record. LOL

If you are resetting the spl levels to their rated value then you need to do that with all the drivers.

I just started a new job and haven't had time to do anything speaker related.

Over 12,000 views too!

[WernerM]

Ahh, but just changing resistance in an L-Pad, or before crossover, etc, changes the reverse null. I don't mind modeling again, just need to make sure I am using the correct information. I don't like to assume, the whole ***-u-me thing, and I don't have a lot of money to keep buying parts. If you were in my shoes, who's SPL would you trust? I think the GRS and Goldwood files are fine, but if not can someone let me know? Thanks for the encouragement. I'm not giving up on this, I know I CAN do it with the correct information. My dad always said "Can't never could do anything".

[WernerM]

Wow, 17 pages, this has got to be some kind of record. LOL

If you are resetting the spl levels to their rated value then you need to do that with all the drivers.

I just started a new job and haven't had time to do anything speaker related.

I did reset all of them to manufacturers rated SPL, the thing that throws me off is that the Vifa TC9 is measured in 2.83V/m instead of 1W/1m. From what I have read, the difference is about 1-2dB less at 1W/1m from 2.83V/m. This is why I am uncertain of Zaph's measurements, even though his info is usually really good.

[duanebro]

I did reset all of them to manufacturers rated SPL, the thing that throws me off is that the Vifa TC9 is measured in 2.83V/m instead of 1W/1m. From what I have read, the difference is about 1-2dB less at 1W/1m from 2.83V/m. This is why I am uncertain of Zaph's measurements, even though his info is usually really good.

At 8ohms they are the same. They will be close.