• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Box tuning for Altec 604-D

Well, good news and bad news. Good news first.

I have them up and running (though not completely finished), and with the Markwart crossovers with the midrange attenuation and they sound pretty great. The bass is actually quite amazing compared to what I'm used to, which is only bookshelf speakers. The mid attenuation has proven key, as they got a little bit shouty in the upper mids. Dialing that down with the pot (in this case L-Pad) is very useful and makes them much more pleasurable to listen to.

Now the bad news - I had a lot of trouble getting consistent and understandable readings when I tried to measure the impedance curves. I ended up with completely backwards results. When I was reading the impedance, the "spikes" that I found were far from obvious and were sometimes not repeatable...so I'm sort of at a loss. I would suspect the frequency generator, but when I switched my DVM to Hz that was the one thing it read stably and consistently. I hooked two DVMs at the same time across the woofer and they both got drastically different Ohm readings as well. I was confused. I had them set to resistance in Ohms...was that right? This was a different setup from when I measured for the Fs - I measured AC voltage across a 10 ohm resister, instead of reading ohms across the woofer with a 1k resistor in series with the sine wave generator. Should I try the voltage method again?

EDIT: Just to be clear, I did the measurements with just the woofer, not through the crossover (I installed those after attempting the measurements).
 
It's AC voltage you should be reading, not Ohms.

In the one case, you're reading voltage drop across the resistor, and in the other, it's across the driver.

At the impedance spike, the drop across the resistor will be lowest, and at the driver, highest, so the two measurement schemes are read inversely.

Plot your readings against frequency to see what's going on.

I believe you'll find the results to be similar either with or without the crossover in place, as the crossover isn't doing anything in the frequency range of interest here.... :yes:
 
Ah! Ok...well I guess I'm an idiot. I probably should have figured that out. I guess since impedance is given in Ohms I assumed I should be reading Ohms across the driver. Oops! Back to the lab for more tests! It's going to be hard to convince myself to disconnect these mothers and stop listening to records on them! Maybe I'll cut the 2nd piece of pipe I have into 5" pieces and measure with 4" and with 5" and record the differences if any. If the 4" is suitable I can hack an inch off the 5" pieces, and if 5" is needed, I think I can stretch my budget the extra $3 for another 2' piece of PVC pipe...I need to make another trip to the store anyway.

Speaking of which, what is a good way to fasten those giant air core inductors in the crossovers to the board? I built them on roughly 10"x10" plywood but have no clue how to strap them down. Hot glue? The 14ga 4.7mH inductor is like 4" high and weighs like 2 pounds. What will hold it on? Am I wrong in assuming I should avoid metal fasteners?
 
They have wide heavy-duty cable ties at Home Depot, hanging in the electrical department, as I recall. Drill slots in your board.

currituckco said:
The mid attenuation has proven key, as they got a little bit shouty in the upper mids. Dialing that down with the pot (in this case L-Pad) is very useful and makes them much more pleasurable to listen to.
Click to expand...
You have discovered the importance of employing HF compensation with compression drivers. It's accomplished with midband attenuation in those crossovers.... :yes:
 
More data! Less wank!

Here we go!

So I had slightly strange measurements on the first woofer/cab combination. Here's what I got. All measurements taken using the 4" ducts. I haven't cut the 5" yet.

0 ports open (Fc) = 71.6
1 port: Fl=21.92, Fm=42.7, Fh=74.72, calculated Fb=30.6
2 ports: Fl=23.93, Fm=47.1, Fh=76.28, calculated Fb=35.6
3 ports:Fl=29.75, Fm=48.4, Fh=85.8, calculated Fb=55.9
4 ports: Fl=30.67, Fm=52.2, Fh=87.6, calculated Fb=59.1

the 3 and 4 port figures seemed higher than I expected.

The other woofer/cab yielded more expected results:

0 ports open (Fc) = 66.3
1 port: Fl=25.54, Fm=37.48, Fh=68.0, calculated Fb=29.7
2 ports: Fl=25.8, Fm=39.63, Fh=71.5, calculated Fb=37.2
3 ports:Fl=26.14, Fm=41.4, Fh=73.9, calculated Fb=41.8
4 ports: Fl=26.15, Fm=48.6, Fh=75.38, calculated Fb=44.4

When I first got one speaker up I listened briefly with different plug combinations and thought 1 plug sounded possibly best, but have been listening with no plugs. It looks like 1 plug might well be closest to the Fs of the woofers.

So what's my next move? Would it be useful to try 5" ducts? Should I swap woofers and cabinets and see if the weird readings follow the driver or the cabinet, or neither?
 
WHOA, beautiful work!! :thmbsp:

If the second cabinet is correct, you're right where you want to be. Our target was 40 Hz, and you have that bracketed nicely around 1 port closed. You just listen and decide which tuning gives you the bass response you like best.

Yes, I would pursue the "problem" with the first cabinet further. Woofer swap sounds like a good starting point. Also look for something like one cab is missing a sealing gasket, etc.

Something's up. Fc should be nearly the same on both cabs with the same driver. If not, look to the cabs. If yes, then look to the driver. Did you get both drivers sealed properly?

Another check you can do on the cabs: Buy a cheap $15 plastic stethoscope at the big drugstore. Take the bell off, so you're just listening to the open-end tube. Close all four ports, and play 30 Hz tone on the speaker (doesn't have to be loud, just so the cone is obviously moving,) then go around all seals and seams listening for "Motoboat." Where you hear it, the cab is leaking.

More data, indeed. If what you have found is valid, you might have been months wondering what the heck was going on with your speaks.... :yes:

Could we have some pics of your speakers and measurement setup, please, so readers may have a better idea of what is going on here?
 
I'll get some pics up tomorrow.

Thinking about it, I did not replace the gasket on the one speaker that had existing gasket material (which is probably 50 years old now), and I believe that is the one that had the weird readings. At least there's a 50% chance it's that one ;) I'll investigate further and let you know what I find. Maybe before I swap drivers and cabs I'll give the weird one some attention and see if I can't chase down a problem. If nothing is obvious I'll do the swap. The stethoscope is an interesting idea, too!
 
Okay - here's a lesson in properly sealing cabinets.

I removed the driver from the cabinet with the weird readings. In fact it was the one I had put new weatherstrip in. So I figured it was not the woofer's seal to the cabinet. I did notice, however, that I had not done as good of a job sealing the baffle to the cabinet on this one when I dismantled it. So without swapping woofers and cabs I just reinstalled the same one in the same cab but took pains to seal it better. I also sealed the back better. Here were my findings:

0 ports open (Fc) = 69.0
1 port: Fl=too low to measure - below 20Hz, which is as far down as my signal generator goes, Fm=?, Fh=70.7, calculated Fb=?
2 ports: Fl=21.32, Fm=41.7, Fh=72.6, calculated Fb=31.1
3 ports:Fl=23.95, Fm=49.9, Fh=74.38, calculated Fb=36.7
4 ports: Fl=26.26, Fm=50.5, Fh=75.11, calculated Fb=39.6

Wow, what a difference! This is pretty much spot on to Zilch's estimates before - 40hz with 4 open, and down 5hz with each port closed!

I bet if I work to better seal the other cabinet I will get similar numbers there as well. I think the key was putting weatherstrip along the back bottom where the wire exits. I think that, along with the baffle, were the seal weaknesses.

These numbers look damn good, though!

And now, I will go to bed, a little bit satisfied (though never completely)!
 
Until we became conscious of how important it is, homebrew (and many professional) box builders paid little attention to QL, the quality of the construction, a measurable parameter. A 1/2" hole remaining open after removal of a level adjustment control in an otherwise perfect cabinet may shift the tuning by several Hz; it's an "Auxillary" port. Leaks do the same thing.

I remember building a pair of Heathkit AR2As, now nearly 50 years ago. The final step in the instructions was to stick the supplied rubber tube in your ear and listen around all of the seals for leaks.

Looking back to what your cabinets were when you first acquired them, it's easy to see that they were out of control, and there was little likelihood your 604s were ever going to get a fair hearing.

[Get the stethoscope.... :music: ]
 
I cut the 5" sections and repeated the tests on the same cabinet/driver and found even closer numbers to Zilch's simulation:

0 ports open (Fc) = 68.4
1 port: Fl=below 20Hz, Fm=27.0, Fh=70.4, calculated Fb=?
2 ports: Fl=20.19, Fm=29.3, Fh=72.3, calculated Fb=30.9
3 ports:Fl=22.32, Fm=43.2, Fh=73.5, calculated Fb=34.9
4 ports: Fl=24.62, Fm=45.1, Fh=75.5, calculated Fb=40.3

So I got another PVC pipe and cut 5" pieces out of it today. Here are some testing/work in progress pictures:

IMG_0829_2.jpg


IMG_0830_2.jpg


IMG_0832_2.jpg


IMG_0831_2.jpg


The last one is one of the completed Markwart crossovers.

I still have some assembly to do, and need to fix the inductors to the boards in the crossovers and install them into the cabinets permanently, but all the elements are there now.
 
Heh, heh.

Loudspeaker design engineering at CurrituckoLab. :thmbsp:

[We're certainly having fun NOW! :banana: ]
 
Looking good curritucko!

It'll be interesting to hear what you think of the sound of these after a good listen.
 
currituckco said:
The stethoscope is an interesting idea, too!
Click to expand...
Yes and it works well as I've had my DIY speakers at the top secret ZilchLab for testing. Great project and I'm glad you decided to make new baffles instead of using the original ones with all the holes. I remember reading about you buying these speakers but just found this thread. Great job on your speakers and enjoy listening to them. :thmbsp:
 
I'm glad I made the new baffles as well, I learned a ton!

Thanks for all the words of encouragement. I'll post more pics when they are actually up and running. Later on I plan to refinish them, and at the very least stain the baffles. I am currently putting silicone caulk around the ports.
 
currituckco said:
I'm glad I made the new baffles as well, I learned a ton!

Thanks for all the words of encouragement. I'll post more pics when they are actually up and running. Later on I plan to refinish them, and at the very least stain the baffles. I am currently putting silicone caulk around the ports.
Click to expand...
There's a lot to learn from Zilch and Bowtie! I just painted my baffles black with some spray paint and they look great.

attachment.php
 
Nice! I think I may stain mine a dark wood color - make them look more furniture-ish in case I want to run them with the grills off. The grills look none-too-shabby themselves, though - except for the fact that I stapled the grill cloth inside-out, but I kinda like it that way, at least that's what I tell myself. Whatever, it's a 10 minute fix if I get sick of it.
 
Update

Installed and playing the hell out of a Skeeter Davis LP! These speakers really sound incredible to me, with plenty of low end, amazing and smooth response throughout the spectrum, and a really wonderful character.

Thanks to all who helped me out with this project, especially Zilch, whose footprint is seen in many a happy listening room thanks to this excellent forum.

IMG_0893.jpg
 
You must log in or register to reply here.
Back
Top Bottom