Why Doesn't PE Offer A Crossover Design Service?

[Flacjunky]

I cant be the only one out here that is refraining from buying a bunch of Dayton RS Drivers because I dont know how to properly integrate them.

It appears the few pre-assembled crossovers PE does sell are universally panned on the forum they host, and as I understand it, for good reasons.

I see that Madisound offers such a service, but only for the expensive drivers they sell. It just looks like PE would sell a lot? more drivers if they offered such service. It would pay for itself, not just with the design / assembly fee, but also increased driver sales.

Either that or offer a much broader array of Kits w/o cabinets. Where's the 3-way kits with 12"+ drivers?

This board is stocked with knowledgeable Hobbyists who have spent decades refining their PCD skills, but there has to be a large pool of potential DIYers out there who just want to build 1 nice set of speakers and move on to other areas of interests. I would think an in-house or 3rd Party design service catering to those of us with just the carpenter skills could make a hefty profit marking up assembled boards.

Just my 2 cents.

[Face]

Re: Why Doesn't PE Offer A Crossover Design Service?

I see that Madisound offers such a service, but only for the expensive drivers they sell. It just looks like PE would sell a lot? more drivers if they offered such service. It would pay for itself, not just with the design / assembly fee, but also increased driver sales.

https://www.madisound.com/store/product_info.php?products_id=8462

No, they'll design a crossover for any driver they stock.

[Pete Schumacher]

Re: Why Doesn't PE Offer A Crossover Design Service?

There are so many projects completed out there to choose from.

PE's crossover design service is called TechTalk. ;)

[robwest]

Re: Why Doesn't PE Offer A Crossover Design Service?

Thought I read somewhere here on the PE site that their techs would be happy to help you design an x-o...

[mzisserson]

Re: Why Doesn't PE Offer A Crossover Design Service?

There are so many projects completed out there to choose from.

PE's crossover design service is called TechTalk. ;)

Costs a lot less too !

[Flacjunky]

Re: Why Doesn't PE Offer A Crossover Design Service?

PE's crossover design service is called TechTalk. ;)

OK Pete, since your volunteering. ;)

Here you go...... :D

I have a 7.5 Cu. FT. Box with a baffle thats 27" wide x 21" Tall....



I want to use 2 Dayton RS150-4s for the Mids, and an RS28F-4 for a tweeter, and a Dayton ST385-8 15" Series II Woofer, and be as close to 8 Ohms as possible.

Do you take Paypal?

[turn2]

Re: Why Doesn't PE Offer A Crossover Design Service?

OK Pete, since your volunteering. ;)

Here you go...... :D

I have a 7.5 Cu. FT. Box with a baffle thats 27" wide x 21" Tall....



I want to use 2 Dayton RS150-4s for the Mids, and an RS28F-4 for a tweeter, and a Dayton ST385-8 15" Series II Woofer, and be as close to 8 Ohms as possible.

Do you take Paypal?

Your idea is kinda "out there". Take a look at this design and see if it leads you to some, uh, better ideas. CJD's RS150S-8 MTM

MTM on a bass bin perhaps? Good luck!

[Taterworks]

Re: Why Doesn't PE Offer A Crossover Design Service?

Madisound's LEAP crossover design service will get you only so close to a working crossover, but it won't be optimized in other ways. They'll find the spot where frequency and phase are closest together, but they may not necessarily consider baffle step compensation, proximity to room boundaries, nonlinear distortion, enclosure loading, or time domain performance. These are all things that need to be considered, and some of them can't be determined without having the drivers in the box and making measurements of the completed speaker system.

Here's how the design process needs to go, as best I understand it:

1) Choose the drivers for your project based on your goals for the project's intended use. The frequency response of both drivers should be fairly smooth in the intended crossover region, without much ripple due to diaphragm breakup. Also make use of sites like Zaph's that give harmonic distortion and cumulative spectral decay information. For distortion, pick drivers that behave as similarly as possible at the crossover point. Then for CSD, pick drivers that have the cleanest plot, over the intended passband, particularly through the upper midrange.

2) Simulate absolutely everything you can. Jeff Bagby's spreadsheets work great. You need Windows and Office to run them, because some functions rely on Visual Basic code, but you should really own Office anyway, if you own a computer, and even the cheapest flavor of Office 2010 comes with Excel. Start from the manufacturer's measurements, and follow Paul Carmody's step-by-step process for designing without measurements.

2) Optimize the design in the theoretical realm, using the simulation and the driver manufacturer's measurements. Get it as close to perfect as you can. If you're designing a transmission line enclosure, tweak the values in Martin King's Mathcad worksheets. If you're designing a horn (not recommended for beginners), tweak the Hornresp simulation.

3) Build the system (box, drivers, crossover). Make sure things are accessible for future tweaks - you'll want to mount your crossover in the box in a location where it is accessible and removable, or else use a bi-amp terminal cup in the design so you can build the crossover externally and connect each pair of binding posts right to the drivers. If you're building a three-way (bad idea for your first few projects), just let the wires dangle out of a port, though I don't like the way this looks, personally.

4) Measure to verify the simulation. Find an acoustically-dead room that someone will let you use, or hang up heavy blankets over the walls and floor of your room, and make close-miked measurements of the speakers (1m, on-axis). Buying the gear to do this is not as onerous as it may sound - just buy a Dayton EMM-6, a USB audio interface with balanced XLR inputs and phantom power like the very affordable M-Audio MobilePre USB, and a mic cable. Use a free measurement tool like HOLMImpulse. Measure each driver individually without the crossover (but use a LF protection capacitor on the tweeter), then individually with the crossover hooked up, then as a complete system with both drivers hooked to the crossover. Save the data from the program so you can refer to it again. Gather frequency response, impedance response, and phase data.

5) Tweak to the measurements. If you discover anomalies that your crossover was supposed to address, but they still made it into the system output, then go back and tweak your crossover in the simulation program (and if possible, feed the actual individual full-range measurements of each driver into the simulation program so you're working with real data that reflects your project), and then substitute the components into the crossover, and measure again.

6) Listen in the room. Once your design has been tweaked according to measurements, it's time to tweak by ear in your own listening room. You may decide you want the tweeter to be a bit brighter, or that the design might sound livelier if you cross over a bit higher. Take notes, play many different types of music (including types of music you might not personally like) and be very critical. Listen quietly, then turn up the volume - it will sound different. Make separate notes for both performance regimes. You should play at least some recorded material that includes sounds for which you have some real-world reference, which means you probably shouldn't only use hip-hop and techno to audition - play some things with real instruments.

7) Tweak further as necessary. Based on the notes you took during listening, incorporate those insights into your crossover design in simulation-land, tweaking a dB here, and a dB there. You'll mainly be tweaking resistor values (thankfully, since audio-grade resistors are still relatively cheap compared to boutique capacitors or inductors). Then listen again. Go to a high-end audio store and listen to some speakers you can't afford, then come back and listen to your own design. If you can't pick out any shortcomings in your design, then stop. Your design is finished. However, if you notice any additional shortcomings in your design, then make notes, take a wild guess at what might fix the problem, and tweak the simulation.

Following this sequence of overall tasks, which I call the "Design Craftsmanship Criteria" (and will be applying in my own next design that will come to Iowa) will ensure that your project provides as much value as possible to yourself and anyone else who decides to build your design. These steps represent a bare-minimum effort to develop a design that covers all the necessary bases to bring your design from theory to reality, and represents the very best of your present ability. Without making measurements of your completed speaker, you won't ever know how well the conceptual design was realized, or whether there's anything else you could to to optimize your design further beyond the realm of simulation alone.

[czag]

Re: Why Doesn't PE Offer A Crossover Design Service?

I believe that the RS28f is interchangeable with the RS28a, and there are many MTM designs using the RS woofers and RS28A tweeter. There are even some good three ways based on them. The search function is your best friend in this case.

Chris

[Flacjunky]

Re: Why Doesn't PE Offer A Crossover Design Service?

Madisound's LEAP crossover design service will get you only so close to a working crossover .....

For my purposes you could have stopped right there.

Thats probably good enough for 90% of the people that just want to build themselves a nice set of speakers. Not everyone wants to invest the massive amounts of time it takes to understand / implement the rest of your post.

I'm glad you did go through the whole process, its great information to have ...... but most people are not going to go through that kind of effort, nor take their finished speakers to Iowa.

..... and those people, who wont invest the time it takes, may not deserve to have sonically perfect speakers with a perfectly flat response ..... we'll cede that realm to those of you that live and breathe this hobby.

I just want to build one nice set of speaker, and then sit back and listen to them for years.

I've learned quite a bit about speakers in the last few months that I've been following this forum. Because I want my one set of 7.1 set of speakers to be the best I can do / afford.

But once I finish them ..... if I ever finish them ...... I'm going to move on to other things and forget about speakers for awhile.

Thats all I'm saying ..... not every DIY speaker builder wants to invest as much time in this hobby as most of you have. These speakers are just next on my list of things I need to build, so I can move on to other pursuits.

And right now the only options for people like me are generic pre-assembled crossovers from the place I normally buy my drivers (PE), or a optimized but not perfect crossover from a place that sells more expensive drivers (Madisound).

It just seems like as respected a player as PE is the DIY marketplace, they would have realized years ago that such a service is in demand ...... or maybe the demand is not as great as I assume.

[Taterworks]

Re: Why Doesn't PE Offer A Crossover Design Service?

Just to point this out:

PC - You already have this.
HOLMImpulse - Free
Dayton EMM-6 - $50
M-Audio MobilePre USB - $100
25' Microphone Cable - $20
Total - $170

That's your measurement setup, right there, for less than two hundred dollars. If you're not willing to spend that money, then you are much better served by building a pre-existing design, even though it may not look exactly like your sketch. Doing your own design involves taking a certain plunge into the hobby, without which your projects will fall far short of their potential. I know because I have done what you're attempting to do with my first multi-way speaker project, and was embarrassed when I measured my design. Now, if you're dead-set enough to go through with a made-from-scratch design of your own, then go ahead, but make sure you ask nicely when you ask a member of this board who has invested their time and money to become proficient at loudspeaker design to support your design effort (since there are those who will), and don't take their help for granted. Also, keep in mind that the best they can do is ask astute questions and try to simulate your application based on your responses.

And, if you read my post above, you already read my explanation of why the Madisound LEAP crossovers are only one step above generic, and can hardly be considered 'optimized'. Flat frequency response is only a tiny part of a complete speaker design, and once you put a speaker in a room (since I assume that you plan to listen in a building with rooms, and not outdoors in a free field environment), the shortcomings of the LEAP crossover design will become more apparent.

The reason Parts Express probably does not offer a similar service is because of the volume of business they do. One technician would be overwhelmed by the demand. That and the fact that the cake is a lie when it comes to having a crossover design service do your crossover design work for you, because of all the variables of your particular application that they cannot anticipate (as explained above).

I would be very disappointed if you decided that DIY loudspeaker design is a crock because things didn't turn out the way you wanted. There is a reason that the high-end segment of the audio industry exists, despite the fact that one can build similar products for less money when only the materials are considered, and that reason is because it takes effort to do a competent design. If Pete is willing to do that for free, then I hope that will hold some special significance for you, since he knows his stuff and will do the best job he can, and will be giving you something of real value.

[Flacjunky]

Re: Why Doesn't PE Offer A Crossover Design Service?

That and the fact that the cake is a lie when it comes to having a crossover design service do your crossover design work for you, because of all the variables of your particular application that they cannot anticipate (as explained above).

Cake is a lie ...... :D GLaDOS is still alive you know.

Boy, I'm certainly not wanting to cross swords with you here Rory.

I understand exactly where you're coming from, I just dont have the time to invest in all that right now.

I built a room, I built the cabinetry including the speaker box, I'm building the sub as we speak, then I need put drivers in the cabinet, build a desk, install lighting, carpet, TV mount, and get all my AV equipment in the new room by the time football season starts.

A year from now, if I'm not happy with my SQ, I might get back into this hobby with enough enthusiasm to follow up on your good advice. But if I'm happy with it my flawed setup, I wont.

I'm not taking anyone's advice here for granted. I appreciate all the time and effort that went into your experience in this hobby, and for all those that have published their hard work for others to benefit from. But I'm not as interested in pursuing the hobby long-term as some of you are.

I just want to put a cam and some headers on this thing and burn some rubber, I dont plan to blueprint the engine and take it to Daytona.

[Taterworks]

Re: Why Doesn't PE Offer A Crossover Design Service?

Cake is a lie ...... :D GLaDOS is still alive you know.

Boy, I'm certainly not wanting to cross swords with you here Rory.

I understand exactly where you're coming from, I just dont have the time to invest in all that right now.

Meh, if you want to see real Nazis, there are some other forums I could link you to. I'll lay off, since this forum isn't those forums, and that's part of why I personally like it.

[Brian Steele]

Re: Why Doesn't PE Offer A Crossover Design Service?

FWIW, I used Madisound's x-over service twice. The first time produced very good results. The second time was Ok, but I suspect that's because the "two-way" they designed for me was put to use in my car.

[Face]

Re: Why Doesn't PE Offer A Crossover Design Service?

I used their service once also and it also came out ok. Eventually I tweaked it and came out with something much better.