Tuning 604 Project - High Q and High Output Impedance?

Seadweller

Member
2007-03-20 6:55 pm
Greetings,

Having been an audio nut for 30+ years, I finally decided to take the plunge, and build my own pair of speakers...They are based on the Great Plains Audio (Altec) 604 coaxial driver, along with a tested cabinet design. I'm using the stock Great Plains crossover, which was evidently modeled for a cabinet of a similar baffle width, but I'm still struggling with the "tuning" of the system. Let me say right off the bat that I'm a complete novice, but I've done quite a bit of research to try and educate myself. I now have more respect than ever for speaker designers! It’s certainly a far cry from putting a driver in a box!!

With that said, I still need to be fed information at the kindergartner level. :xeye: What I am faced with is a system that has what seems to be the typical baffle step issues. Output from the vocals up is extremely forward and overwhelms the rest of the frequency spectrum, to the point of being “shouty.” I have experimented with a baffle step correction circuit, as detailed by the very kind and generous Martin King, but I don’t believe it applies to a two-way driver, as I’m experiencing odd results (vocals become recessed, but the remainder of the spectrum remains elevated).

A couple of folks have mentioned that the 604 driver is “high Q,” and designed to be used with an amplifier with high output impedance. The cabinet design evidently was designed around this parameter as well. Sadly, I am in the dark with this.

My current solid state amplifier is an obvious mismatch for this system, so I am wondering if moving to a tube amplifier with high output impedance would change the character of the system? If so, how (in laymen’s terms if possible)?

Thanks so much!!! Despite my lack of knowledge, I'm having fun, and am determined to make the system right!!! There's magic hidden within this system that I want to break out...
:smash:
 
Right. Most SS amps have a high output impedance (or low damping factor if you prefer). With a low Q driver, which is essentially over-damped in the bass, you will have a relatively weak LF response, because there's only so much the cabinet can do to boost things, and it's not getting any help from the amplifier. Shifting to a lower output impedance / higher DF amp will bring things back into line. Meantime -you could try sticking a resistor in the hot lead to the driver, which will artificially lower the amp's DF a bit.

In the long term, you might want to change your amplifier -doesn't have to be to tubes -Nelson's First Watt F3, or a couple of his F4s with a high-gain pre-amp spring to mind. :bawling:

604s. Oh well, one day, when I'm rich.
 

Seadweller

Member
2007-03-20 6:55 pm
Thanks everyone.....

Is there a straightfoward way to determine the critical damping factor of the driver (or my system in total) in order to identify the proper amplifier match?

Once identified, I would assume that I could calculate the optimum output impedance, correct?

Or, am I looking too far into this? Do most tube amplifiers have sufficiently high output impedance to accommodate?

Thanks again!

:cool:
 
By using a BSC filter, you have already simulated the impact of the low damping factor that can be achieved with a tube amp. Above a few hundred Hz, you essentially have a DC resistance downstream from your SS amp in series with the driver. This is the effect you would be looking for in a tube amp. However, your DC resistance is adjustable since it is not amp design dependent.

The other property of a tube amp that might be of assistance is a rolled off high frequency response. This might help eleviate a painful top end. At one time I had an old Adcom GFA 555 amp which was a very dark sounding amp. At the time, my Focal inverted metal dome tweeters sounded very nice. When I switched SS amps to a newer vintage the tweeters sounded too bright.

In effect, you could tune your speaker further by selecting a tube amp. But which one? I would recommend trying before buying to make sure that the tube amp's high frequency properties compliment the speakers high frequency response.

Or, if you want to experiment some more add a low pass filter at some higher frequency to roll off the highs gradually. This would definitely be a trial and error effort and might take some time. I have no idea what starting values to recommend since I have never worked with this driver.

Hope that helps.
 

Seadweller

Member
2007-03-20 6:55 pm
Thanks again everyone.....

Regarding the BSC filter duplicating the effects of a lower damping factor, what value of resistance would provide an accurate representation of having an amplifier with a higher output impedance? Would 2ohm's of resistance be a fair representation, or would it be much higher, on the order of 10ohm's?

I still have the circuit in place, so it would be a simple matter of changing out the resistors.

I am also consider a different crossover, likely the Jeff Markwart design, where L-pads can be used to attenuate the higher frequencies. I'd prefer to get the new amplifier in place first to determine its impact on the system.

Thanks!
 
nullspace said:
It's not as though they don't have that reputation.

The 604 series don't and one reason why they were the recording studio standard for many decades, it's the old 300 - 800 Hz sectoral and multi-cell CD horns that do when driven with a low output impedance and no CD horn EQ, so it's how you implement them that determines whether they 'shout' or reproduce a signal more accurately than all but a very few of their contemporaries.

GM
 
Taming the 604 ... Way too long!!

Ok..... were to start??.....

This topic is a mine field.... and no mater what one says... there are going to be major disagreements.

So..... In the interest of disclosure.... I have worked with the 604-8K as produced by Altec... Mid 80's variant. And the 604E as well as the 605A and own several of these wonderfull machines.

Have not worked with or heard the GPA 604. Hoped to hear it in Tulsa and Dallas at the Midwest Audio Fest and Lone Star Audio Fest.. but GPA and their fine products were not displayed there the last couple of years.

Moving on..... The amp and damping factor issues..IMHO are WAY over played. Are they factors in determining the sonic signature??... yes..no doubt about it. But cross over and correct box implementation swamp the amp issues. The load is right or not. Period!

What I have found is.... with much experimentation ... is cross- over is nearly EVERYTHING with All of the Altec 600 series of drivers.

No way around it!!!.... So.... Like all drivers .... ultimate results are influenced by the combination of five major components...
1) The box (and alignment.... reflex...sealed or open air)...

2) The driver itself (of coarse and remember they do vary) Gotta measure the T/S

3) The crossover and

4) room &...

5) Personal - musical taste components.

Suspect at this point several inmates are rolling there eyes... but I related all of this in deference to your original inquiry.

Questions.... what box are you using? General room charactaristics... your personal listening - music preferences?

As for Crossover...... with the low QTS (Low QE) the woofer wants to roll off early down low and has a rising reponse up high...above 1,600 HZ and tends to peak around 3k. And as Jeff Marquart has demonstrated.... there are time alignment issues to be considered because of the horizontal separation of the LF and HF acoustic centers.

Again.... sighting Jeff..... a spread frequency 2nd order Butterworth seems to ..... if properly implemented cure the horizontal distance issues. Proper pading of the HF is also part of the solution. The HF.... like the woofer can also be hot in the upper 2,000's in to the mid 3000's and a voltage divider or trap is needed to tame this hot area. And there is a premature roll off at the very high end that also requires attension. All of this is crossover stuff.

Now on the bass end you can use some major baffle correction.... and low frequency roll off compensation to get this thing to exhibit tonal balance. We are talking some large value inductors with by pass resistor to do this correctly. Any way you look at it.... you have to trade away 3-4 db of woofer effieciency to get it to play with authority into the low 30's. You can use the traditional 9-10 cu.ft. box with a Extended Bass Shelf alignment... but you pay a price... and that is a 2-4 db or more sag in the bass response from the mid 50's to the mid 200's. Some of that is gained back with a wide 35-40 inch box width.... but it never quite compensates. Again... I know there are those out there shaking their heads... Go measure and you will see this IS the case.

OK... so.. if you want it right.. you combine Jeff's crossover techniques with added bass correction in around 8 Cu. Ft. box tuned to mid 30's depending on the room ...and the box needs to be about 30" across. You will have to fiddle with the H.F. pad and or traps. The good news is after you figure it out... it is one of the best.... most musical drivers available.... It is a thoroughbred and requires much care and feeding.

My $0./02 cents worth...!!!

Good luck and keep the forum informed as to what you do and how you end up.
 
Wow John, that certainly was $.04 worth!! 8)

Much appreciate the input....I've been chastised by some for undertaking this project with the little knowledge I have, but so many kind and generous individuals (Jay Fisher, Greg Monfort, Martin King, Jeff Markwart, and many, many others) have been so unselfish with their time and knowledge, that I’ve enjoyed every step of the way! When all is said and done, I know I’ll end up with speakers for a lifetime, which is my ultimate goal.

With that said, to answer a few of your questions, my listening room is around 14x20, and fairly bright acoustically. My listening preferences run the full gammit, from classical to rock, but my primary interest is in contemporary and acoustic jazz. I listen mostly at moderate volumes, but when home alone I do like to rotate the dial to the right at times!

The box I am using was designed/modeled by Jay Fisher and Greg Monfort. It’s a bottom vented enclosure with an MLTL alignment, and around 10.5 cu/ft I believe. My dimensions vary ever so slightly from the original design, but I kept the same cross sectional area, which I was told is critical for the MLTL alignment. The crossover I am using is from GPA (their standard version, with no L-Pads).

The Markwart crossover in on my list of “to-do’s,” but I have not taken this step because I don’t feel comfortable practicing my soldering skills (which are poor at best) with $700 worth of components. I also am very wet behind the ears when it comes to reading schematics. Actually, I just sent Jeff an e-mail yesterday to see if he might know someone that I could commission to build the crossover for me.

The articles I have read about damping indicate that a systems ideal operation occurs when the driver is “critically damped.” When it’s over-damped, based on the response curves I looked at, the low end rolled off quickly and fairly sharply. This leads me to believe that a higher output impedance amplifier would definitely work better with this driver, but I’m sure that's only one piece to the puzzle, and maybe even one of the smaller pieces….I don’t know for sure.

To move forward, would there be some consensus that the place to start would be to install the Markwart crossovers, followed by a move to an amplifier with higher output impedance?

Does the Markwart crossover work to tame the drivers’ anomalies (proper padding of the HF, dealing with the woofers hotness in the upper 2,000's in to the mid 3000's, premature roll off at the very high end, etc.)? If not, how do I address them economically?

Lastly, will I still require a BSC filter, or can this be accommodated with the crossover somehow?

Serious Stereo evidently received rave reviews for their GPA 604 system, which is essentially the same system that I constructed. Now if only knew what crossover they were using, I might be set!!! I guess that would take some of the fun out of it though, now wouldn’t it?;)

Thanks so much everyone.....I can't tell you how much I appreciate the assistance!!!!!!!!!!!!!:D
 
GM said:


The 604 series don't and one reason why they were the recording studio standard for many decades, it's the old 300 - 800 Hz sectoral and multi-cell CD horns that do when driven with a low output impedance and no CD horn EQ, so it's how you implement them that determines whether they 'shout' or reproduce a signal more accurately than all but a very few of their contemporaries.

GM

Hey Greg --

Certainly you have more experience with the 604 than I, so I'll defer to you on this. I'm going off comments made by Staephen at 6moons here and Billfort here. Also, if I'm not mistaken, there is/was discussion of GPA doing a 604 with some variation on the UREI -- I assumed it was to help with issues of forwardness/ aggressiveness/ 'shout'. I myself have a pair of 802s on 50hz conical horns and can state categorically that they can be a bit on the hot side at times (though I suspect that is a flaw in implementation moreso than the driver/horn's fault).

Regards,
John
 
Re: Wow John, that certainly was $.04 worth!! 8)

nullspace said:

.........there is/was discussion of GPA doing a 604 with some variation on the UREI -- I assumed it was to help with issues of forwardness/ aggressiveness/ 'shout'.

Greets!

Urei added woofer(s) to fill in the bottom end just as Altec did as wider BW monitoring speakers became necessary and did the necessary XO and HF horn updates required for this plus the transition to all solid state electronics because by then Altec was 'on the ropes' financially thanks to corporate raiding 'bleeding it dry'. Indeed, their increasing lack of build quality and willingness to do the upgrades combined with high pricing eventually forced them to quit using Altec altogether.

Anyway, if you drive 604s with the electronics they were designed/voiced with, then with the original stock XO their only real shortcoming is a poor timing match through the XO BW and the 'soft' lows and highs that was typical for virtually all multi-ways at least through the '60s.

With all SS though, it takes an $elaborate$ passive XO to mimic all the complex interactions of tube systems with highly variable reactive loads. Bottom line, get it right and it will easily outperform the original and over a much wider BW to boot. That said, the horn is too small/XO point too high (the original at 1 kHz was about right), so until this is dealt with they are going to fall short of being all they can be.

GM
 
GM said:


Anyway, what amplification/XO, any contour, etc., filter(s) are you using?

GM

Signal chain is Laptop (Foobar 0.9.4) -> TDA1541A DAC -> 27 linestage -> 56 driver & 45 output amplifier. The setup is 802-8Ds on Bill Woods' 1000hz conical horns with the network he sent me, which is 2nd order ~1400hz (electrical) + a notch filter for ~2khz + CD compensation. For the woofers, it's a third-order BW (electrical) also at 1400hz.

Only so much Bill can do from up there in Canada, so I'm cobbling together some measurement gear and burning the midnight oil with D'Appolito's "Testing Loudspeakers". The biggest hurdle so far has been that I didn't have an amp suitable for testing -- I ended building a mono LM3886 amp to use. The rest of the stuff should show up next week, so you'll probably start seeing cries for help on the board in about 10-12 days...

If you want to talk about it more, absolutely feel free to email me. Or you could post here.

Regards,
John