Homebuilt khorns get some tweaking

[homemade]

I built some khorns from speakerlab plans about 15 yrs ago and while visiting this forum it got me thinking about modifying them. I used an Eminence 15inch speaker from McGee Radio out of K.C. I used a 6.5 inch Audix HRM17 midrange driver that has 99.6 db efficiency and the EV T35 tweeter. I was pretty satisified with the sound if I used an eq. to reduce the midbass and increase the lower bass. For the crossover I wasn't using anything on the woofer. I ran it in this mode for the past 14 yrs or so. I read some posts about adding some dampning material to the back chamber and doing away with the coil. Well, I added some fiberglass sheets to the back chamber and didn't notice much difference. I than proceeded to add the coil back to the woofer. I located some 6.2mh coils and added them. I noticed less midbass and more balanced sound. Did some quick measuring and found the coil cut the output to woofer aprox. 1.5 db and lowered the output significantly between 150-400hz. I'm probably running too low high of mh coil but it does sound more balanced. The next step was to email Eminence and find out what I was actually running for woofer compared to the K33. Here's a quote from the email:

"The 5415's that you have were custom manufactured for McGee. We have a

couple of drivers that are close. Our Delta 15 is a good parameter match

but it has a much shorter coil which will give you less excursion. Our

Delta 15LF is a good match as far as soft parts go but the parameters don't

match as well. Below are the T/S params for your McGee's and the Klipsch

K-33. The McGee's should be good for 300-350 watts and the K-33's 200-250

watts.

01-SPEC 15467 04-DATE 06/06/94

02-PART # EM5415HI8

03-MODEL # 1525F5637

05-RE OHMS 5.37 13-FS HZ 35.38

06-LE MH 1.42 14-MMS GMS 80.19

07-QM 7.91 15-CMS mm/N .2523

08-QE .480 16-RMS NS/M 2.2545

09-QT .460 17-VAS LTRS 259.89

10-XMAX MM 7.30 18-SD SCM 856.34

11-BL TM 14.08 19-EBP 73.3

12-EFF % 2.30 20-SPL dB 95.6

01-SPEC 15162A 04-DATE 11/30/92

02-PART # K-33

03-MODEL # 1520F5637

05-RE OHMS 3.39 13-FS HZ 34.46

06-LE MH .96 14-MMS GMS 78.59

07-QM 7.39 15-CMS mm/N .2714

08-QE .410 16-RMS NS/M 2.3037

09-QT .390 17-VAS LTRS 301.66

10-XMAX MM 8.20 18-SD SCM 889.59

11-BL TM 11.88 19-EBP 84.4

12-EFF % 2.91 20-SPL dB 96.6

"

I noticed he said that my woofer was rated for 300-350 watts and the Klipsch speaker 200-250. Interesting. I only lost 1 db efficiency (I'm running the 8ohm speaker too) and 1mm less xmax. Does anyone know what the effect the differenct QT and VAS would have on my speaker from the K33? I than proceeded to measure the effeciency of my speaker. I ran white noise from my amp and measure 2.73volts at the speaker and my Radio Shack meter read 103db after I backed off the tweeter adj. because it made the meter peak past 109 db. I'm guessing my 6.5 inch midrange is over 100db efficient and possibly close to 103 which is what the meter read with the woofer and midrange going. I only thought my speakers were in the 100db effecient range. Maybe my voltmeter is off! Maybe I'm getting to the point of removing the 15 band eq. in my system! Now, I'm thinking of buying an Adire Tempest 15inch subwoofer to add some needed lower octave power?

[ricktate]

Hi im from KC area to maybe we could get together and talk speakers?I got pair of Ks im veneering and fixing up.Rick

[WMcD]

I had read about people buying from McGee in SpeakerBuilder. That was probably a bit more than 15 years ago, though.

It looks to me that PWK was pleased to encouage home builders early on and factory approved woofers were available. Then, maybe, he became annoyed at commercial copiers and shut down on supporting others to build his units.

But now Klipsch will sell you a K-33 at a very reasonable price.

Gil

[D-MAN]

The Speakerlab K is a virtual knockoff of the Klipschorn right down to the small slot size of 3x13" used for the K33E. It's clear that Speakerlab was copying and not really understanding why things were the way they were. If you REALLY want to stay with 8 Ohm woofers, then cut the slot wider to 6x13", and you'll find that your problems go away.

Actually, you'll have to cut the motor board slot AND the throat cavity slot wider for each cabinet (I have the SL plans, too).

The K33E at $130 each is a splendid match without any modifications, and after all, that's what the stock SP-K and the Khorn are specifically intended to use, anyway, at least with the small slot size.

DM

[homemade]

D-Man, what can I expect if I cut my slot to 6x13inches? I'm thinking about doing the mod and even doing away with the motor board. Will it change only the upper limits of the woofer or the low end also? I tried to do some research and didn't come to any conclusions about the low end but some did claim to have better low end by doing away with the motor board. I can't imagine this little change would make that great of difference.

I wonder when McGee Radio went out of business? I live near Wichita and not K.C. so I've never been in the store, only had their catalogs from years ago.

[D-MAN]

The general rule of thumb on the "slot" cavity opening is

"the wider it is, the higher the crossover point can be".

Of course, a 1:1 ratio with the throat area cross-section will also result in maximum efficiency.

The maximum allowable slot size in both the Khorn and the Speakerlab K is 6" x 13".

The reason that it will effect your drivers is that the 8 Ohm driver is raised higher than 8 Ohmns when put in the horn. The K33E (4 Ohm) goes higher, also, but being 4 Ohms, it ends up at around 6.3 Ohms when in the horn.

The Khorn 3x13" slot size was put in production by Klipsch in 1962 when they switched to using the K33 from CTS and its subsequent off-shoot Eminence, who currently makes the K33E. Up until that time, the slot was 6x13" and notably, the horn throat has remained unaltered in size (6x13 = 78 sq. in.) even though the slot size was changed to account for a peak in the K33 response somewhere around 400Hz or so.

The "sqeezing" effect of the smaller slot size reduced the response peak at the upper-mid-bass.

You will find with the wider slot size, the driver that you currently have will go a bit lower and be more efficient at the same time. The sound will "open up" noticably.

You will not have to change the crossover, it should remain at around 400Hz or so. I would also retain the motor board but make a new one that has a wider slot to match the horn cavity opening and then you can switch back and forth at will.

I can speak from experience in that my horns use the wider slot size, and I use the smaller slot in the motor board to allow for the use of the K33E.

DM

[homemade]

I thought the smaller slot was suppose to raise the response around 400 hz or so? I think I will go ahead and cut out the panel to 6x13 and mount the woofer directly to the housing and forget the mounting board unless I decide to go back to it I can just install it and be done. I understand the later khorns don't even use a motor board? How much greater efficiency and low end? I don't want to add much because I'm midrange limited by my 6.5inch Audax speaker. If I could achieve less response in the 1-200 range and more under 50 that would be great but I don't want to be adding any in the upper bass.

[D-MAN]

Klipschorns may have NEVER used a separate motor board as far as I know. I always assumed that the separate motorboard is a SpeakerLab "addition", which I chose to use in my DIY horns. However, it won't cause any problem if you leave them out. Some people here report better bass with a direct mount, but I remain sceptical myself as far as that goes.

The answer on the slot is that it was used to reduce the upper freq output of the woofer. PWK was quoted in an magazine article as saying that it was to correct a "dip" in the K33 response, but either he was mis-quoted or was just backwards in that instance. Narrow slots than the horn throat cross-section reduce upper frequency bandpass. Take it from me, I know from first hand experience, also it is in "How to Build Speaker Enclosures", Badmaeff and Davis, W.Sams Publishing, Az, 13th Reprint, 1978, if you don't believe me.

PWK also was quoted that the woofer went to 16 Ohms impedance when loaded into the Khorn. Certainly is NOT the case, either. That one gave me fits, but I decided to take the results of modern electronic testing as posted by John Warren over the PWK article in question.

I cannot estimate how much the efficiency will go up for you, that is dependent on the woofer. You will most likely like it, though. I doubt that it would be night-and-day kind of a difference, but you will notice it. You'll probably like it.

DM

[homemade]

Well, I couldn't wait and ripped the speaker apart and made the mods. I posted a freq. response first to verify any differences. This is for an uncorrected Radio Shack meter reading in the middle of the room which is far from ideal. (11.5x13x9ft with a door in the back and the left wall has an opening 5ft wide. 25hz is off the scale, 31--10,40--6,50--6,63-0,80-+4,100+8,125-+3,160-+8,200-+4,1000-0. I removed the motor board, opened up the opening from 3x13-6x13, took out the fiberglass fill, re-ran the tests and the lows actually went down very slightly and the 100hz setting wasn't at +8 anymore but more like +6. Put the fiberglass back in and re-ran tests and basically the same thing. I know some of the bass rolloff below 60 is because of the middle of the room and towards the back where I sit it picks up in the lower octave. Was this mod worth it? If it reduces the 125hz peak a little-probably so. I'll probably modify the other speaker to have a matched pair. I still like the sound better if I crank in 12db of gain in the 20-40hz range though. What I probably need is the Adire Tempest with 250watt amp in a sealed 3.5 cu. ft box but that'll cost me around $250 and somewhere to put it.(I already have a box).

Would anymore speculate whether the Tempest can put out decent enough volume at 20-25hz to keep up with the khorn in at least movie sound?

[D-MAN]

I would agree that less insulation would be in order. Put some directly BEHIND the driver to reduce reflections through the cone as shown in the Speakerlab plans. The VAS of your driver is slightly less than what is needed for the K33E. You might consider using the motor boards as a displacement to account for the difference in back chamber volume, rather than leaving them out entirely.

You will find that once you get 2 cabs with the same mods, the changes to the sound will be very apparent, where I always find that one mod may be somewhat indicative of a change, it's not until I get the other cab done the same way does it really become apparent to the ear.

I would guess from the specs that those drivers are actually quite nice.

DM

[johnnyp]

"....I located some 6.2mh coils and added them...." ???

I think the drivers you have chosen are fine. Remove your 6.2 mH inductors and install a pair of air coil low-DCR 2.4mH inductors and rerun your frequency/dB test.

[D-MAN]

You COULD even run the woofers straight, that is, with NO inductor at all. The natural rolloff of the high frequencies will occur due to the nature of the driver, the horn and its folds. At one point, PWK even recommended this in one of the Dope from Hope papers, but it was later recanted.

Your 6+ mH coil should be fine for an 8 Ohm woofer and a slightly higher crossover point, which probably won't make it through the horn anyway... however, it will "take" some frequencies "away" from the midrange, so it might be noticable in a network "balance" sense.

As far as getting more lows, they should technically be there if you have a good seal in the corners. Adding more insulation will not increase the available lows, and too much will increase the IM distortion resulting from unequal cone excursion; this will sound like "mudiness".

DM

[johnnyp]

homemade,

One other thought. Are you running your 6.5 inch Audix HRM17 without any crossover components?

If not, one problem you may have is frequency overlap in the ~150Hz - 400Hz region.

[homemade]

I went ahead and modded the other speaker and re-ran a freq. response test with the meter in another location and couldn't really dicern any difference from the before and after mods. I did run the woofer without an inductor and it made the sound muddy and too tubby from too much midbass output. I'm actually satisfied with the midrange crossover point and output. The curve starts out at 60hz at 0db and increases to a high of +8db at 160hz and starts going back down to 0db at 400hz or so and than dips 2db slightly and back up to 0db at 1,000hz. I use my eq. to decrease the upper midbass output aprox. 4db and increase the 40hz setting 6db. I'm running a 12db/octave high pass filter at the midrange and tweeter. Now sure of the values without removing my crossover.

[lynnm]

FWIW

I modded my 1982 Khorns by rewiring the Type AA crossover to revert it to a Type A and liked the result a lot. I then bypassed the inductor for the woofers and found that the sound while still very good became somewhat less defined in the low frequency range and put the inductor back into the circuit. I find the sound more integrated without that modification but that said the bass was a little stronger without the inductor but overall the transition between the bass and the midrange became in MHO rather obvious.

[D-MAN]

I experimented with taking the inductor out of the loop, and I too think that it sounds best in-circuit. PWK came to the same conclusion, too. I got a kick out of it when I ran across that DFH installment, followed by the next month saying "oops, put it back in", more or less.

Now as to the reported frequency response, this is with the wider slot, direct mounted drivers in both cabinets, and the horns are well-seated and sealed in the corners?

DM

[homemade]

My Speakerlab plans are a bit different than the earlier plans in that the top board doesn't go all the way into the corner and there is no tailboard as such. There is actually an 8inch gap between the back of the top board and the corner. I have experimented with a top board to seal the walls but I wasn't able to measure any difference in low end response. I believe that sealing the khorn to the corner only affects the upper bass range. I was playing an organ recording that all of a sudden has a -6db 21hz pedal tone (loudest low freq. signal any of my recordings have beside maybe some dvd movies I've rented) and if eqing the low end too much and turning up the bass knob resulted in a reading of 110db at my couch with that note but noticable distortion (probably from the woofer bottoming out). The low note will shake the window in the other room. The khorn will easily take the low note signal if volume is kept below 3\8 and low end eq. is kept below 12db for a db level of around 95 or so. The funny thing about this low note organ is that the bass is out of phase between left and right and one subwoofer won't began to produce any tone. I'd like to have someone post this note from an mp3 I can make and see what their systems can do with it. Anyone interested? I could email you the small mp3 file.

[D-MAN]

Actually the seal against the corner completes the last section of the bass horn, allowing it to go down to 32 Hz. So this explains your lack of lows below 60Hz.

Can you post pics? I'd like to see what you are talking about.

You technically should be getting bass in spades so much so that you shouldn't need a sub, that would be way too much.

DM

[homemade]

Here's a terrible picture of it. (Sony TR730 camcorder in very low light). You can see that it doesn't have the sides extending to the walls and top plate that would seal it to the corner and no tailboard. I really don't believe this effects the low end at all. This is from a later Speakerlab plans that actually houses the 6.5 inch midrange and tweeter in an enclosure that is built into the front of the bass bin instead of the top. You can see that I put the midrange and tweeter in a separate box on top so I could rotate them like I wanted. They claimed a freq. response of 32-20khz +-3db in their add! I really believe that the midrange Audax driver gives the speaker a very nice sound without any kind of horn resonances in the midrange. I haven't listed to a Klipchhorn so I don't know for sure but I really love the midrange of this combo.

[Marvel]

A friend of mine had some that he always had out from the corners about three or four feet. They still sounded good (in their own way).

Marvel