I have a pair of 604-8ks waiting for cabs and this is the first I've heard of this problem. Is there a solution to correcting the midband by replacing, or modifying the stock XO?Quote:
Originally Posted by Todd W. White
Cheers
Printable View
I have a pair of 604-8ks waiting for cabs and this is the first I've heard of this problem. Is there a solution to correcting the midband by replacing, or modifying the stock XO?Quote:
Originally Posted by Todd W. White
Cheers
Maybe the speakers just aren't for you, especially if you're fighting their basic nature so to speak.
In my expirience problem with the bass is in the lowpass part of the crossover. Due to the rising response of the 604's bass/midrange driver, low pass crossover should be done so that it compensates it otherwise 604 speaker would have "no bass" and "shouty" sound.
Seadweller: try it with the crossover I posted
SD-50 -
The only real solution I ever had work for me was to eq the things way more than I'd like - I'm a "Cut Only" type guy and don't liek to use much of that, even (I probably get that from Don Davis and the AcoustaVoicing training), so I've used 1/3-octave units on it, as well as parametric's on them (used to own the Altec relabeled Dynacord parametric - it was NICE!). The parametric's were really nice, because I could dial in the correction, but, that said, you can do it with a 1/3-octave equalizer, especially if the filter skirts are constant-q, rather than variable-q. I like variable-q for live-sound eq, but not for this application.
The other solution is a custom crossover with eq built into it.
We used to have all kinds of problems with the acoustical consultants not allowing 604's because "they don't have any bass" back in the "K" years, when, in reality, as we later found out, the bass WAS there, but it was being overwhelmed/covered up by the midband...kind of like adding two dozen trumpets to your concert band, but not increasing the number of low brass players (tuba, baritone, trombone) to compensate and balance them out - the group will sound "shouty" and "harsh", not because the bass isn't there, but because the other guys in the trumpet section are playing TOO LOUD.
Does that make sense?
Oh - and thanks to all the other fellows who echoed my advice to take them outside: I remember years ago talking with my friend the late Dr. Paul Veneklasen about his experiences in measuring the Altec speaker systems out at the RKO Ranch in the late '40's: he said they had to get the speakers way up off of the ground to minimize LF reflections, but they still had to EQ the curve to compensate for them.
That said - even though he might get some bass boost from the ground reflections, he's still got a way better chance to hear what they're really doing without the interference one gets from the reverberant field...
Thanks Todd.
Gustavo Fitzwilliam
I had a suspicion that the "shoutiness" was coming primarily from the woofer, and not the horn...I disconnected the horn and confirmed this, but it's hard to tell just how much the horn itself is contributing.Quote:
Originally Posted by vuki
Is it possible to install an L-Pad in a crossover that would lower only that range of frequencies? I have to imagine that a 16" driver would have a difficult time with midrange frequencies. Doesn't a driver of this size start to "beam" at higher frequencies? Is it possible to drop the level with some sort of notch filter?
Thanks...So, does the active crossover allow you not only to set the XO point and slopes, but also adjust specific frequencies, similar to an equalizer?Quote:
Originally Posted by Panomaniac
Seems to me that XO's are more than just slopes and crossover points, that there are notch filters, uplift circuits, and other circuits that manage the drivers output.
Yes indeed, all that can be part of a crossover. Or the crossover can be very simple.
The nice thing about a digital crossover like the DCX2496 is that it allow you to do so many things. Most active crossovers have fixed filter types and don't allow to adjust each section independently.
But of course that comes with a price. Tweaking can be endless, if you let it.
Yeah I agree...and if you are an analog purist...it's still the best way to quickly determine the response you want...and THEN emulate the curves in the analog world!Quote:
Originally Posted by Panomaniac
I have a pair of the GPA 604 H-3's in a six foot box. I also use a small band pass sub with the speakers. I love the sound and the sub helps. Very good bass.
Nick
Wait !!! You've gone through six different crossovers? Hmm ! The gentleman from Chicago should have suggested L-pads for his second build. This would have given you some control of the slope and the balance (similar to the model 19 crossover).Quote:
Originally Posted by Seadweller
Seadweller - have you tried removing the 3 ohm resistor on the bass circuit? Or replacing it with a different value one? I know the MLTL design for these drivers call for it, but if the bass is anaemic in your room, maybe this might help tune to your bright room?
David
This is not the best solution but you can try this for an immediate temporary solution.....Quote:
Originally Posted by Seadweller
Take any one box, and continue.
1) Arrange for some AC capacitors of 2-4-6-8-10 micro-farad values.
2) Arrange for some lamp cords. Connect it directly but separately to the 604's woofer & horn terminals, and bring them out thru the vent opening. Then strech them to a distance of around 10feet, so that you can sit & listen from a comfortable distance.
3) Bring out a similar connection from your amp-output to the same spot.
4) Take the GPA crossover, and connect the amp-cable to its amplifier marked terminals.
5) Now connect the woofer to the X-Over's woofer-output.
6) Do not connect the horn now.
7) Play some known music
8) Now start by shorting the woofer (across both the connection points) with a high value capacitor say 10mfd, and see how much of the shouty-ness has reduced. If too much reduced then use the next lower value.....and go on like that till you reach a position that you like.
9) Now while the woofer stays connected with the chosen capacitor......Connect the horn to the GPA X-Over's horn terminals....and listen.....to the whole thing...and try to asses if any further amount of that shouty-ness is now added due to the horn.
If you feel so, then disconnect only the +ve wires of the horn, and insert a cap (highest value) in series....and listen......if still shouty....then try with a lower value cap.....and continue till you get what you want.
Once you reach a suitable spot, copy the connections to the other box and listen in totality.
Hope this will be of some help and make you carry on with your project :).
Aditya
Find some 904-8A or 604-8K crossovers. They are identical and match the manta horn.
Or possibly some 604-8H units, I believe they are the same.
As for the comment from Todd on the -8K magnets. I doubt this is true for the 904-8a as I have never felt mine shout from the midrange. Granted, the 904 has a diff woof cone, and I modified the HF to mimic the -8K and -8H.
I love the sound, mine don't go real low, but subs fix that in spades. The mids blend so well, these things disappear and all you hear is music suspended in space.
I've never heard the GPA 604's.
But I wonder.......
Are they really that good?
Are the crossovers really that good?
Altec got the dual shelved crossovers for these reasonably right, and they flat work.
Your test of removing the horn tells a lot in my opinion......the schematic for the stock crossovers for the woof is a simple 12dB/oct setup.
Maybe the woof of the GPA drivers just sucks?
Ever tried one of the vintage versions?
Ron
I think that Vuki's passive is worth a try. It sims well. Nice spread frequency with a dip in the middle and a rise for the top end of the horn. Looks like a ~2.5K notch.
I might replace the 4R7 and 27R with an L-PAD just for fine tuning. It's doing about ewhat my active crossover was.
Depending on one's listening preferences, room acoustics and system, varying (3R3 - 6R8) the value of 4R7 resistor would give wide enough range for tuning high freq. energy. There is no need to change 27R.Quote:
Originally Posted by Panomaniac
One cannot tune low/midrange with L-pad. It is done with series inductor value. Varying capacitor parallel to woofer would change only the upper part of woofers frequency response.
@ron,Quote:
Originally Posted by RonSSS
before I decided to purchase the GPA 604H series II I had the opportunity to audit 604E in 602 EMI enclosures and 604H in 612B enclosres side by side. finding a serious pair of vintage 604H (last with alnico and first with tangerine phase plug) is a jackpot. only reason why I would shift from GPAs to vintage Hs when ever I get the chance to buy a pair of those vintage Hs is pure PSYCHOLOGY! and I'm quite sure that if I swap them I already miss the GPAs because I fell in love with this in-between vintage and modern sonic that might distinguish the series II from the original Hs I've audited ...
@seadweller,
I use the 604H series II in 'gutted' 612B enclosures. they are lifted from the ground via spikes, bent and sloped. also my room is quite huge (>40sqm) the distance while auditing is 2.5 to 3 meter. the bass is lovely delicate and subtle I've never heard before with my former audiophile systems that contained all tube gear and speakers for more than 15k. however in my experience cables and amplification are quite important next to the crossover. in my case triode in single ended is an outperforming match ...
good luck
I used a simular procedure when putting together a 3-way system last year. I also used DPDT switches to reverse the polarity on the tweeter and mid range speakers. It turned out that the mid range +/- were stamped incorrectly. I had to reverse the polarity on the mid range to connect it in phase.Quote:
Originally Posted by aditya
Again, don't know much about 604s. But could the L.freq section of the 604 be accidentally connected out of phase with the H.freq section?
Not meaning to say "you" connected the 604 incorrectly. :)Just wondering if internally the 604 is connected out of phase. Have you tried reversing the polarity between the H.freq section and the L.freq section? Won't hurt.
That's cool.Quote:
Originally Posted by vuki
Vuki did you know that your HP filter (before the contour) has the topology of a 3rd order, but the electrical function of a 1st? You might double check me, but it looks like the filter functions is very near 1st order at 4KHz. A single 5uF cap would do the same. The inductor and resistor do pull it down a little faster. Did you do that by ear?
No, I didn't do it by ear. I did it through measuring and ear. You probably analyzed it with some simulation software which didn't tell the whole story. In reality there is significant difference between simple 1st order and this crossover. With 1st order there is no way to get crossover at 1600hz and no midrange hump (at aprox. 2-2,5khz). Anyway, as I said it measures and sounds pretty good. ;)Quote:
Originally Posted by Panomaniac
Hi Todd:Quote:
Originally Posted by Todd W. White
In 2005 you said on this forum, "You see, during the years EV owned Altec Lansing (after the move to OKC), they forced Altec to standardize on a few magents, rather than the large number Altec was using. The early, Anaheim K's were actually quite good - almost as good as the "H"....but the change to a different magent hurt it a lot."
How do you tell an early vs late (Anaheim vs OKC) K? Serial number, but I recall no one kept a serial number list in all the reorganisation. I have K's and am quite satisfied with their overall balance, using K-xovers and Stonehenge encl's. S/N's are in the low 1000's.
-Jon Archibald
I believe the serial number stickers tell the difference: Serial numbers beginning with a printed A are for OKC (believe it or not), and those beginning with a printed K are Anaheim.
Gary can correct me if my memory is wrong on this.
There have been L-Pads in all the XO's he's built, but it's a struggle to control the horn energy without killing the highs...Quote:
Originally Posted by sba2
Perhaps some experimentation with the range that the L-Pad covers would help?
Yep, I've tried it without, and have tried varying values...I actually even tried a baffle step circuit just to satisfy my own curiosity...Quote:
Originally Posted by gortnipper
I've been wondering how to tell these apart too, but my 604-8k's don't have an A or a K on the serial number sticker, only 02191. Are there any other tell tale signs, like magnet sizes?Quote:
Originally Posted by Todd W. White
Dan
The date code will also tell you - it will be something like 8234 - whcih means 1982, week 34.
If I remember right, Altec moved to OKC around 1984, so anything before that would be the original design, but beware - they had lots of QC problems then...
The other is the size of the magnet - it's huge.
OK, both mine are labeled 239792. Is it possible that 8k's were made as early as 79?Quote:
Originally Posted by Todd W. White
Thanks,
Dan
There should be a by-pass cap to let the extreme highs through unattenuated, but it sounds like either it's not there and/or not a high enough value and/or not wired to be inversely proportional to the L-pad's setting. Ditto for the mids L-pad.Quote:
Originally Posted by Seadweller
Bottom line, I don't know if Altec built enough adjustment into these later designs and GPA has shown a penchant for maintaining the status quo, so not sure how any of these sound since my 'hand's on' experience is limited to the 604E, which at the time only required adding a bypass cap by ear around the single attenuator once it had been adjusted to balance the horn's lower BW to make it sound pretty @#$% smooth, tonally balanced all the way up as high as vinyl, RTR tape recordings went.
The more I've read about the later XOs though, the more I wonder if the extra adjustment causes more trouble than they're worth as just like adding drivers causes a ~exponential increase in design complexity, so it seems with two separate L-pads for only a two way speaker system, at least WRT the 604s.
GM
Thanks for the wonderful input everyone...So it looks like I have several options here:
1. Go with an active digital XO
2. Try the XO designed by Vuki
3. Experiment with caps and such
Any concensus which way to go?
By the way, how much of a difference does the port configuration make when it comes to bass response? Is diameter more important, or is length the key? Could a 6" port be too small for this cabinet?
Thanks!
I think the 6" port is small. But this is supposed to be a MLTL, so maybe it's OK. GM will know, he's the MLTL guy.
If it were a bass reflex with that size port, you'd want it short for the most bass. Like 3/4". That should be easy enough to try, right?
As far as crossovers, I would say try Vuki's. It won't cost much to put together. (about $60 each from Parts Express or Madisound) To make it even simpler, you can leave out L2, R2 and substitute a single 4.7uF for C2, C3. It won't be exactly the same, but it will be close enough to hear the potential. At the moment you are most worried about tonal balance.
And as Vuki says, you can vary R3 to change the level of the horn.
The active crossover will allow a lot more tweaking, but takes time to learn. Cost about $275
Did you get a chance to haul them outside over the weekend?
Unfortunately not...I was away from the house nearly the entire weekend, and now I've got a week's worth of business travel...This is what has made this project such a challenge to finish...Quote:
Originally Posted by Panomaniac
I was also going to buy an SPL meter at Radio Shack, and take some measurements using my Stereophile Test CD, but haven't had time for that either...
I'm having the gentleman in Chicago make the XO's, because the only thing I can do with a soldering iron is melt little plastic army men... :) My neighbor is an electrical engineer, so maybe he could at least help me with the parts layout...I'm quite unfamiliar with some of the components on Vuki's XO, such as "4R7." Could someone help me decode the components listed on the XO schematic? It looks like it might be fairly easy for me to assemble if I know what the parts are.
The first thing on the agenda is to buy an amplifier...
Absolutely !! What will you be going with tube or SSQuote:
Originally Posted by Seadweller
Yes, get the amp first.
You don't even need to solder to put this crossover together. Just use screw terminal blocks.
And we can help with the layout and the shopping list.
I will also be gone the rest of the week on business. Back at it next week?
Quote:
Originally Posted by Altec Best
I'm going back to SS...I sold my Parasound A-21 power amp and Sonic Frontiers SFL-1 pre-amp to buy a Cary SLI-80, and regretted it from day one...The Cary was a magnificent piece of equipment, but I'm in Florida, and it put off a tremendous amount of heat...I seem to prefer a more "analytical" sound, so the tube sound wasn't my cup of tea...My Sonic Frontiers SFL-1 was a SS/Tube hybrid, and it added just enough warmth to the presentation...That combination was incredible...Sure wish I could have it back...
Sounds great!!Quote:
Originally Posted by Panomaniac
You're welcome!Quote:
Originally Posted by Seadweller
High SQ digital is the way to go in today's 'two bit' world and even if you want a passive XO/response shaping filter(s), finding its specs with digital really simplifies the process considerably.
Quite a bit. The diameter + length sets its efficiency and tuning frequency, i.e. whether or not it may 'wheeze'/'chuff' at high power. For some reason I thought you cloned Jay Fisher's MLTL, but the driver is in the wrong place, so to answer your Q I need to know the cab's net Vb and the port's length.
If it's around 10.3 ft^3 like JF's, then a 6" dia. port is plenty large enough to handle the Iconic 704's/GPA 604-8H II's 100 W rating.
GM
I like the sound of both, I like to switch occaisonally depending on program material I feel like listening to,I think the Altec 604-8G's sound fantastic with tubes though,right now I'm using SS with my 19's and can't wait to get the 604's back into the mix again.Quote:
Originally Posted by Seadweller
I assume you have a laptop for business, so one of the freeware programs such as ARTA, HolmImpulse or similar would be the way to go: Re-Jigging the jig: Speaker testing device for Arta, Speaker Workshop & HOLMImpulse - diyAudioQuote:
Originally Posted by Seadweller
4R7 = 4.7 ohms
GM
I too prefer 'analytical', but there's a certain synergy between SET and compression horns, so tubes for HF and at least 200 'clean' watts for the woofer to ensure it's likely never to clip on transients (preferably Class A, but A-B is OK if at least the first 10 W is CL-A and its DF is < 80 which indicates a minimal amount of -nfb in CL-B mode) seems to satisfy both the bass 'sharpness' and inner detail/presentation the two excel at to ~ 'have your cake and eat it too', though not exactly the cheapest way to go.Quote:
Originally Posted by Seadweller
GM