Seas ER18 design thread

[peter_m]

You might want to keep it at 16.5L as Zaph suggested it in a posting for people building their own enclosures... As for tuning he mentioned he would go slightly lower, in your case making the port slightly longer.

Peter

[ThomasW]

Is it really a big deal to get new baffles cut for the PE boxes I already have?

It's no bigger deal than ordering any other product from PE.

[LNeilB]

It's no bigger deal than ordering any other product from PE.

That was not the essence of my question, but thanks.

This was:

What are the reals sonic risks involved, considering how close the PE and MAD baffles are in size...

[---k---]

Your question really wasn't clear. You didn't even say what PE enclosure you had or include links. Maybe it is lost is the thread somewhere, but I don't remember it.

So I went and looked up what the SR71 kit includes:
MD14BP cabinet (.5cf, rear ported) - 9.1” W x 15.4” T x 11”D.

versus what I assume you have from PE:
Dayton TW-0.50CH - 14" H x 8.5" W x 12" D.

If you can make everything fit: 9.1" wide versus 8.5" wide, or a 7% difference. It shouldn't make that much of a difference in the sound. It the real world, 99% of the population probably couldn't hear the difference (with Zaph, Thomas, JonMarsh, CJD, etc. making up the 1%).

If you're still concerned, you could download the Baffle Diffraction Simulator excel spreadsheet from FRD and model the two baffles to see how much of a difference there is between them.

[cjd]

Honestly, you'll see more difference from the difference in edge treatment than anything. If you can do a felt diffraction control any difference should be negated beyond any sensible ability to hear.

C

[LNeilB]

Thanks for your replies folks, guess I was a bit vague to begin with, but I was referring to the PE .50 cabinet. Sounds like felt is a quick way to try it.

I'm just trying to compare before committing to cabinets. Sounds like it will be close enough to get a decent impression.

Thanks for the help!

[SQconstable]

sr-71

Ok, I got three kits of the sr-71. I've got a lot of space to build, especially height-wise and depth-wise, so what's the max I should build these without comprimising the pre-assembled crossover? I want maximum bass extension but without causing problems with the crossover. Is there a point where you shouldn't make the enclosure too big before changing the crossover?

[peter_m]

SQconstable,

Zaph mentioned he would do the enclosure at about 16L if he wasn't using the one from Madisound. He also mentioned he would have preferred a slightly lower port tuning frequency. You can make the front panel as tall as you want but never change the width! Let us know how i turns out... would love to hear it!

Peter

PS: I have a speeding ticket.... can you help me with it?

[SQconstable]

Oh I'm not a constable/cop.. I chose SQconstable for Sound Quality Constable. Yeah that's right, I'm the big bad cop of sound quality!

I'm curious how the phasing of the bass will change if my rear port is not traditionally placed perpendicularly. For example, my enclosures will have angled sides so the ports will not aim straight back.. instead they'll be at a 45deg angle or so. One good thing I guess is that the port won't aim straight at the wall, but I understand the orignal design used a plain 'ole rectangle cube for these crossovers.

[peter_m]

LOL, SQ constable also stands for Surete du Quebec Constable... or as we call it SQ for short... Funny coincidence!

I hope Zaph reads your question and answers you first hand but maybe he already did... On his site he had an experiment regarding front and rear port placement. His conclusions were that either was fine. I think you could extrapolate and assume your configuration will be adequate without fear of getting a SQ arrest warrant issued in your name ;-)

Peter

[SQconstable]

Hehe yeah that is a coincidence! I'll make these babies soon and place some pics up here and at diyaudio.

[SQconstable]

BTW, I'm making my enclosures around 16-16.5L and lengthening the port tube some. I'll use the ears on this one...

[peter_m]

Oh man, I can't wait. I had decided to build the SR-71/ER18 but for financial reasons had to put it off.... To get my feet wet, I decided to build a pair fo Zaph's B3S and see if I really was up for the woodwork. Now I am left pining for the ER-18 design knowing I can do it.... Patience Pete, patience.

Make sure you show and tell for the rest of us!

[SQconstable]

Will do.. btw.. why do they call this a 7" woofer? The ER18 isn't any larger than my Peerless 6.5"


I'm reading product descriptions saying both things:

"The ER18RNX is an 18 cm (7’’) cone driver"

"ER18RNX is an 18 cm (6,5’’) cone driver"

Make up your mind!!! haha

Ok that's it.. I'm measuring this thing with a digital caliper.

[Ray Collins]

Probably frame vs. cone.

Ray

[benchtester]

application of ER18

Before the introduction of the ER18RNX. Zaph published his Waveguide TMM (LR2) design with CA18RNX drivers; would there be any benefits to using the ER18RNX? How challenging would it be to shape it to a LR2 roll-off? :roll:

[jimangie1973]

Before the introduction of the ER18RNX. Zaph published his Waveguide TMM (LR2) design with CA18RNX drivers; would there be any benefits to using the ER18RNX? How challenging would it be to shape it to a LR2 roll-off? :roll:

I think you would need to have the TDFC in the waveguide to do an LR2. An LR2 withouth the waveguide would be difficult.

The ER18RNX has lower distortion in the midrange than the CA18RNX. How audible this is I don't know. I just did a design with the CA18RNX+TDFC with slopes close to LR4. The crossover came out very similar to Zaphs so I can confirm his SR71 kit will sound excellent.

[peter_m]

Has anyone built Zaph's TMM waveguide? I searched but couldn't find any builder's feedback.

[benchtester]

I think you would need to have the TDFC in the waveguide to do an LR2. An LR2 without the waveguide would be difficult.

I have that covered. As a tip to all, I got a bargain pair of TDFCs on ebay because the mounting plates were buggered up. No problem for the waveguide configuration.

Thanks for the comments, they are helping me to refine the questions.

While searching for comments on the waveguide TMM, I found a Zaph response emphasizing that the TDFC waveguide is LR2 at 2000 Hz only.

So, what are the issues at with achieving LR2 at 2000 Hz with the E18? (I am still on topic, although it isn't always apparent .)

[peter_m]

So, what are the issues at with achieving LR2 at 2000 Hz with the E18? (I am still on topic, although it isn't always apparent .)

Overall the ER-18 has lower distortion, but it misbehaves a little a more at 4500hz... might require a notch or a steeper slope to avoid the 4500hz area... maybe?

[jeff_free69]

Just finished my SR-71 kit!

...Took a little longer than planned (doesn't it always?) since I built the cabinet from scratch. It was coming out pretty well, so i decided to go for a glossy finish (which is a great way for OC types to spend their nights . Since I was only using spray cans I had a good excuse to finally say "perfect enough". I did purchase the front baffle, but shaped a lot more roundover on the edges and contoured it into the cabinet.

So on to the listening; this is all very preliminary since I am listening in a studio which is just being redone with more extensive treatment (ie bass traps , etc).

As advertised, they are a very clean and smooth with nothing hyped or scooped. i imagine they would measure fairily flat, as the design page shows.

I may also be hearing a little of the extra "warmness", also suggested on the design page. The speakers are away from the walls and on stands, but adjacent to a mixing desk, so its not exactly as intended for the baffle-step compensation. And the front 1/3 of the room is pretty dead. So I will probably start experimenting with some of the subtle tweaks suggested to lift the tweeter . Reducing R0 from 4 ohms to 3 or 2 ohms should be easy enough by temporarily running a wire thru the port and adding R's in parallel. (maybe I'll end up with a pemanently mounted switch).

The Bass is pretty solid and satisfying, and I may no longer need my subwoofer as I did with some JBLs of similar basic desgin that I've been using lately. Prior to the JBLs+SW I had a 12 year old DIY 3way (12" - 5" - 1"). if you just go by the charts, the SR-71s look like they might be a little thin, but i can't complain when listening to well recorded CDs. At moderate levels at least; I cranked it once to impress myself, but I get nervous when I see those 7' spekers start jumping around. So the jury is still out on the Sub - won't know for sure till i get started recording again and I can see how well the mixes translate. Not that I need it loud, but if i mix in too much bass it may be boomy put in the "real world".

Of course when listening to older rock CDs originally recorded in the 70's they tend to not have any real deep extension, as they were originally intended for LP and had to be rolled off or compressed way down there. the SR-71s may not go all the way down there but, they least let you know whether its there or not. (And thats the beauty of an active subwoofer, where you can switch it off and adjust level and XO point. I keep the amp in an easily accessible rack, not built into to SW cab. I like a lot of bass

I'll post some photos as soon as the studio makeover is complete.

Anybody else try these yet?

[peter_m]

Jeff,
congrats on your new build. Can you elaborate on the mid-range and the treble? How do the vocals sound (male & female) ? There is a whole world out there beyond the 40hz range... I swear! :lol:

Peter

[SQconstable]

Definitely use that sub for mixing, Jeff

I'm a hiphop producer so I heavily rely on a sub for checking things. In fact, I've made it a common procedure to turn off the mains and only listen to the subwoofer when checking subbass. I'll bring the sub close to me where I rest my fingers on the rubber surround and take note of "beats" which is when overlapping close, low-freq tones occur simultaneously. I do a visual check on the cone movement too. If there's a lot of movement on overlapping freq's (like the kick and bassline) but not a decent audible loudness to match the movement, I'll retune the kick so it doesn't coincide so much (or I'll "frequency carve" the kick away from the bass line, or vice versa).

While some "mains" may seem like enough range to listen to the music, you'll always need to really put a "scope" (sub) on that low end to diagnose what many end-users will experience.

I'm still designing my front main pair of SR-71's, but I'm still planning on the complete setup with center and surround's for this particular mix setup. Like you, I am anticipating some warmness, not because of a mixing desk/console surface, but because of these being placed on a 90" wide, 23" deep tv stand alongside a 65" display. I am thinking I will design the mains with the xover accessible from the underside, and with the enclosures on either spikes or feet to distance it from the surface somehow. I'll update..

[jeff_free69]

Jeff,
congrats on your new build. Can you elaborate on the mid-range and the treble? How do the vocals sound (male & female) ? There is a whole world out there beyond the 40hz range... I swear! :lol:

Peter

I will try to get more specific shortly. all i can say for now is that after the first listening and noticing the "warmness" I went back to the article and noticed that in the freq chart theres a very slight hill just under 200hz, . This is combined with a slight dip from 1-3k and pretty tame behavior from there on up. so it makes sense.

I'm going thru lots of CDs in the collection and will also be setting up an A/B switch with the old monitors. (jumping thru CDs from across the decades really makes you realize what the loudness wars have done to dynamics in rock - but thats another thread). I'm still trying to narrow it down to a few representative recordings of good and bad. I'm fairly confident the SR-71 will let you tell the good ones from the bad ones and not just pretty everything up by adding hi end sizzle (that would be the option where you remove C4 & R5 altogether

[SQconstable]

Cool thanks for the details so far Jeff! Do you foresee unwinding the inductors for some baffle compensation eventually? You say you have some room treatment (wall absorbing material) behind the speakers? I wonder if that would help a little with the bump.

[peter_m]

...all i can say for now is that after the first listening and noticing the "warmness" I went back to the article and noticed that in the freq chart theres a very slight hill just under 200hz...

Zaph included this graph regarding the effect of a higher DCR value for the largest inductor: http://zaphaudio.com/SR71-options-L7dcr.gif

What type of inductor did you use?

Peter

[jeff_free69]

I used the Madisound prebuilt crossover, which looks to have their 15awg steel laminate.

First step will be just to drop R0 to 2-3 ohms for a gentle rise starting at 2k. i'm not looking to start unwinding the inductor yet a) because I'm not sure there's an "undo" button for that and b) that tweak looks like it will start lifting things at 600 - 2K, which may make it go from "warm" to "boxy" .
(Though that might help prevent mixing errors in this critical area , ala the famous yamaha NS10s)

As far as the Sub - I purposely have been listening with it off most of the time , just to get to know the SR-71s. right now its on but I'm keeping at fairly low settings. Its also Madisound (12" in a 3.5 cuft sealed box)

BTW - Other listening components are :
TEAC 300mk2 reference CD player,
Mackie Big Knob (preamp)
NAD 100W/ch power amp.

[jkrutke]

Hi Jeff,

If you are listening somewhat near field, I almost guarantee that you will need to unwind the big inductor as described. Possibly even more than that, but unwinding to 2.5 mH would be a good start. Correct though, no undo button for that. Changing only R0 will probably result in more brightness but nearfield boom will probably remain until the inductor is changed.

Tonal balance that works in a normal far field situation basically changes as the ratio of listening distance to baffle width gets smaller.

Room treatment behind the speakers will smooth out the midrange and improve imaging though it's somewhat less important for near field.

[benchtester]

Hi Jeff,

To prototype the change you could either parallel the 3 mH with a 15. mH (a value which you may not have). Alternately, you could parallel the 3 mH with your other 3 mH and put a 1 mH in series. (Check my math, I am rushing at work. :W )

[jeff_free69]

A couple of updates on the tweaking:

-PARTY : GIANTS WIN!
-sober up

- raised speaker stands to get proper height. My usual working/listening position will be sitting up and close. For the past couple of weeks I've been listening rolled back 6-8 ft and reclined.

- Subwoofer overlap: i realized it wasn't just sub-ing, it was woof-ing too. Its a Madisound plate and 12" in 3.5 CF sealed box, but even with the active XO set at its lowest freq, it had significant output in 100-200Hz, and unlike my previous DIYs and JBLs, the Zaphs don't need any help til you get way-way down. Fortunately (this can be expensive!!), in my surplus inventory i found parts for a 100Hz / 12db Butterworth filter to put on the sub . Let me tell you that cleans things up quite a bit! So now i can leave it on all the time, just to get the true lows.

- to facilitate the tweaking I have just extracted the crossovers from the cab, taking advantage of the bi-wire terminal cups. I've ran some short wires and installed some switches so i can A-B things from where I sit. Normal working area is within a 4 ft , equilateral triangle (so even though they are toe-ed in, I am still off axis when inside.

Right now i have 2 switches
- one for the R0 parallel resistor

- one for the C4/R5 cut out
(i didn't think I'd like this one since it looks relatively extreme on the charts, but it was too easy not to try. For casual listening it really opens up the soundstage in my very dead front end. Not sure I'd want to mix like this, but it sure makes em sound pretty.

So all in all things are starting to to sound a little less "heavy-warm"
Now that the guts are hanging out, I'll start playing with the inductors (buy alternates if I can't dig up /patch the right ones - I've got a bunch laying around, but a lot are so old the labels fell off

[benchtester]

Now that the guts are hanging out, I'll start playing with the inductors (buy alternates if I can't dig up /patch the right ones - I've got a bunch laying around, but a lot are so old the labels fell off

As an aside, I was lucky enough to inherit a LCR meter. Now I can't imagine being without it. I would recommend any speaker DIY'er to get one. :T

Good progress!

[ttan98]

JK or anyone,

Is the ER18 and its tweeter combo good enough to be considered as a mini monitor or very close to it?
How does it compared to a small speaker using better quality driver, scanspeak midwoofer and asso. tweeter, eg ZD5?

I am interested in building a mini-monitor or close to it.

thanks

[jkrutke]

JK or anyone,

Is the ER18 and its tweeter combo good enough to be considered as a mini monitor or very close to it?
How does it compared to a small speaker using better quality driver, scanspeak midwoofer and asso. tweeter, eg ZD5?

I am interested in building a mini-monitor or close to it.

thanks

"Mini-Monitor" is just a word that means different things to different people.

The ZD5 is currently the best sounding system on my web site. The way it does vocals and other midrange content is pretty much the best I've come up with. Too bad it's a little pricey and complex.

jeff_free69: Thanks for the studio usage and setup report. While studio monitors were not the primary intended application, It's good to see you're getting them to work in that situation. For others who are considering these for studio usage, I'll be sending them to this thread to absorb the work you've done.

[ttan98]

The ZD5 is currently the best sounding system on my web site. The way it does vocals and other midrange content is pretty much the best I've come up with.

One to one comparison between ZD5 and SR71, does the SR71 come CLOSE to the sound quality in terms of the mid range, dynamics, imaging and soundstage, etc? OR they just have different sound qualities?

Do you think ZD5 is better pair of speakers because you use better quality driver (in this case more expensive as well) with lower lower distortion measurements? Are there any other factors which you think contribute to better sound in your opinion?

You have designed so many speakers I just to know what are the main determining factors in determining the overall sound quality of the speakers. I am interested in the nature/properties of the driver rather than the x-over because they are all designed by you. I am just curious what you think.

Thanks.

[jeff_free69]

Hi Jeff,

To prototype the change you could either parallel the 3 mH with a 15. mH (a value which you may not have). Alternately, you could parallel the 3 mH with your other 3 mH and put a 1 mH in series. (Check my math, I am rushing at work. :W )

I think your math is pretty close, though the stock L is actually 3.3 not 3.0.
I do happen to have another 15.0 available, so that in parallel with 3.3 would give 2.7049180327868852459016393442623 (approximately

I love being able to parallel stuff - makes it easy to A-B test without hacking the stock board too much. If that gives a decent enhancement, I'll commit to buying & installing the suggested 2.5

[ttan98]

answer to query

One to one comparison between ZD5 and SR71, does the SR71 come CLOSE to the sound quality in terms of the mid range, dynamics, imaging and soundstage, etc? OR they just have different sound qualities?

Do you think ZD5 is better pair of speakers because you use better quality driver (in this case more expensive as well) with lower lower distortion measurements? Are there any other factors which you think contribute to better sound in your opinion?

You have designed so many speakers I just to know what are the main determining factors in determining the overall sound quality of the speakers. I am interested in the nature/properties of the driver rather than the x-over because they are all designed by you. I am just curious what you think.

Thanks.

John K,

Can you find some time to answer my query?

Cheers..

[jeff_free69]

I think your math is pretty close, though the stock L is actually 3.3 not 3.0.
I do happen to have another 15.0 available, so that in parallel with 3.3 would give 2.7049180327868852459016393442623 (approximately

I love being able to parallel stuff - makes it easy to A-B test without hacking the stock board too much. If that gives a decent enhancement, I'll commit to buying & installing the suggested 2.5

Spent the weekend listening to these mods and I am quite encouraged with the results. The vocals now have just the right amount of presence. I auditioned dozens of discs from all different genres and pretty much all sound "right" ( whatever that is!).

One CD that was problematic before the change and tipped me off is Radiohead's "In Rainbows" - the singer tends to sing in muted tones and falsetto and doesn't have a lot of "stuff" in his voice. So theres a lot of tracks where there's a relative large amount of fundamental in the voice (lets say 200-400 hz) and less harmonic overtones in the area usually considered midrange (~1K-4K). The change releived that slight muffled-ness and was gentle enough to not make everything else sound overbearing. Like i said , everything just sounded "right" to my ears

My test was a bit of a hack since i didn't have the exact parts available. So I ended up testing with about 2.7 mH and 2.7 ohms instead of the recommended 2.5 mH / 3 ohm. But now i can justify getting the exact values and making the change permanent (so i can shove the guts back inside - I've got quite a frankenshteen wiring harness for the switches hooked up now)

[hpolkerman]

I'm building the speaker you mentioned:

ER18RNX with the 27TBFC:

An answer to your problem check this website...:

http://www.geocities.com/woove99/Spkrbldg/ER18RNX_2Way.htm

Have fun!

[JonP]

I think your math is pretty close, though the stock L is actually 3.3 not 3.0.
I do happen to have another 15.0 available, so that in parallel with 3.3 would give 2.7049180327868852459016393442623 (approximately

I love being able to parallel stuff - makes it easy to A-B test without hacking the stock board too much. If that gives a decent enhancement, I'll commit to buying & installing the suggested 2.5

Just remember when paralelling inductors, you want to keep them as far apart and cross oriented, as you would any other couple of inductors. Just because they are both dividing down to be the "same" single component, doesn't mean that their magnetic fields wouldn't interact and produce different results than you'd expect.

[peter_m]

In the x-over diagram:


C4 and R5 labels and symbols are reversed on the wiring diagram... Are the values in the right column also affected or are they correct?

Peter

[jkrutke]

Fixed. The values are correct, the labels did not match the symbols. It's a Soundeasy glitch. Note that it doesn't matter which order those components are installed.

[tizeye]

Fixed. The values are correct, the labels did not match the symbols. It's a Soundeasy glitch. Note that it doesn't matter which order those components are installed.

Perhaps I am missing something, but by the schematic I see 2 "real" resistors. R0 (4ohm) and R5 (3.5ohm). On pictures on the Madisound site, as well as clearer pictures on another board of a kit the individual received, I only see 1 resistor. With different angles, it doesn't look like one is tucked back and hidden by the larger components.

What gives? Am I reading the schematic wrong or did Madisound alter the crossover slightly omitting/changing R0 or R5?

[jeff_free69]

Perhaps I am missing something, but by the schematic I see 2 "real" resistors. R0 (4ohm) and R5 (3.5ohm). On pictures on the Madisound site, as well as clearer pictures on another board of a kit the individual received, I only see 1 resistor. With different angles, it doesn't look like one is tucked back and hidden by the larger components.

What gives? Am I reading the schematic wrong or did Madisound alter the crossover slightly omitting/changing R0 or R5?

R0 & R5 are real and are there; R5 is just jammed in between other larger components.

[orlofstg]

SR-71 Build Photos

First DIY speaker attempt. Opted for custom Walnut cabinets and baffles (effort to match the trim work). Special thanks to John Krutke for this awesome speaker design and also Madisound for over the top service for a newbie. I thought about spending $700+/speaker for a high end Seas design but I listened to the value comments on the driver review section of John's website. I am blown away with the quality of this build. If you have any reservation about this speaker forget about it and buy this immediately.

[WillyD]

First DIY speaker attempt. Opted for custom Walnut cabinets and baffles (effort to match the trim work). Special thanks to John Krutke for this awesome speaker design and also Madisound for over the top service for a newbie. I thought about spending $700+/speaker for a high end Seas design but I listened to the value comments on the driver review section of John's website. I am blown away with the quality of this build. If you have any reservation about this speaker forget about it and buy this immediately.

Nice build. I know I'd have zero reservation. Love the Seas tweet and I bet the MW is just as good or better. :T