Altec 604E Rear Chamber Modifications

**tonymjb** · February 16th, 2004, 06:15 PM

Hi,

This is my first post here.

I have Altec 604E drivers in custom designed 9 cu ft enclosures. I have made some mods to my drivers earlier (bug screen removed, my own crossover and the horn "separators" removed - the drivers came without the separators).

I have performed the "last" 2 remaining 604E rear chamber modification (both are reversible):

1. Removed the internal wires (with fork terminals) and replaced them with solid core
magnet wire. The wires were soldered to the same place where diaphragm wires are soldered.

2. Removed the Bakelite cap on the rear of diaphragm. 3 layers of 1/16" - 1/8" wool felt were placed into the rear cover cavity. The internal round Bakelite cavity (previously holding the Bakelite cap) received a thin strip of felt as well on its rounded internal wall.

The high end decreased SIGNIFICANTLY so a boost of quite a few dB will be needed in the crossover. Is this decrease to be expected due to decrease of compression on the rear of the diaphragm? (Nothing touches diaphragm. There is a small chance that after assembly the felt could possibly touch the diaphragm wires but it should not happen due to Bakelite cavity.)

Any ideas or experiences?

Regards,

Tonymjb

**Jim Easley** · February 16th, 2004, 09:05 PM

Tony,
The rear cover shouldn't require more than a single layer of 1/8" felt.
From what you are describing, it sounds like you have a loading cap on the diaphragm. This part should not be covered with any material due to it's close proximity to the diaphragm. In fact many people remove the loading cap completly and install shorter screws when they have a rear cover.
Generaly a loading cap is used in three instances:
(1) When you are crossing over below 800 hZ. to limit diaphragm excursion.
(2) OEM in higher output applications. (again, limits excursion)
(3) OEM in concealed driver applications such as the Model 14 or the A-8. This was done to cut manufacturing costs.

Good luck,
Jim

GM · February 16th, 2004, 09:56 PM

Welcome!

Not surprising. The driver's front/rear volume is sized to get the desired BW from a specific throat/mouth area and pathlength, so the combined front chamber/throat pressure is counterbalanced by the air 'spring' of the rear chamber. Increase the rear chamber's volume and the diaphragm's excursion becomes increasingly non-linear with increasing frequency and unable to 'squeeze' the HF out with enough acoustic pressure to overcome all this acoustic resistance. Not only do you lose the HF, but also increase distortion somewhat across its entire useable BW if sufficient EQ is used to flatten it back out.

FYI, depending on where the roll off begins to require EQ and the Q of the gain slope, it's possible to burn up a VC with EQ even in the HF if these are used at ~'live' levels.

Anyway, while I'm not experienced at tweaking the Duplex series, I can't think of a good performance reason to omit the loading cap on them like on some other drivers that were designed without one and it was added to increase its power handling at the expense of a peaky response.

If there's some peaking with it installed, then without removing any more felt than required to refit the loading cap, you can shim it up a tiny amount to create a two stage loading chamber (experimentation/patience required!), reshaping the loading curve while not killing so much of the top end if you would rather not use so much electrical EQ contouring.

GM

**billfort** · February 17th, 2004, 01:23 PM

Hi Tony. I have performed these same mods on my 604G but used a little less felt ? I lined only the back cover with a single layer, maybe 3/16 thick, but I didn?t put anything near the diaphragm or leads. It is my understanding that the loading cap is there for the 3 reasons mentioned by Jim and since none of these apply to me, I removed it. I?m using the stock Altec crossover, a tube pre-amp and 300B SET amps.

The one thing that bothered me about my 604?s was the tendency at times to sound ?shouty? in the upper mids. From my point of view, this mod essentially cured that problem with only a barely perceptible decrease in treble. Maybe it?s just my limited hearing ability beyond 14K but I found this tradeoff well worth it. The nice thing about these changes is that they are easily reversible and I have gone back and added the loading cap to make sure I?m not ?throwing out the baby with bathwater?. My personal preference, in my system (which is a huge variable), is no loading cap and light felt lining in the cover. This is obviously a major case YMMV.

Billfort

**tonymjb** · February 17th, 2004, 02:24 PM

Jim, GM and Bill,

1. I do not have the rear Bakelite loading cap on the diaphragm now (that is what I removed). There was no felt on the loading cap earlier. I have 2 layers of thin felt on the rear cover and a felt ring screwed to the Bakelite diaphragm ring (spaced about 3/16" (5mm) from the Bakelite ring.) My crossover is around 1500Hz. The loading cap was removed to see if it eliminates potential standing waves between the diaphragm and the loading cap.

2. In the original set up, there was no felt anywhere. Should there be felt on the rear cover when the loading cap is installed? I am not sure how good is the seal (if any) provided by the loading cap.

And as you said, there should be no felt in the cavity between the diaphragm and the loading cap.

3. My crossover is fixed in levels (I do not have a pot to adjust the tweeter level. So adjusting the tweeter is a bit more involved.) The crossover is modelled after Hiraga Altec VOT crossover, scaled to 16 Ohms and with notch filters trying to smooth out the response a bit.

4. My drivers came with the horn separators cut out (similar to Urei horn). They go up to 16kHz. There are some "funnies" around 2.5 - 3kHz and 8-9kHz (dips, if I remember correctly).

I got the original 604E horn replacements. I am not impressed by the way how the horn separators start on the inside in the throat. I would have expected nice knife edges. Instead, they start a a blunt, 90 degree cut. I would expect them to create standing waves between them and a phase plug.

By the way, I was once told by a 604A/B owner that 604A/B had a multicellular horn while 604E is a sectoral horn.

Tonymjb

**selmerdave** · February 17th, 2004, 05:40 PM

Tony,

I did similar mods to my 604E's and did not notice dramatic losses in the highs. I wonder if it has something to do with putting felt on the inside of the bakelite ring, if I understood correctly that you did that. Also like Billfort I used a single piece of felt like he described.

Did you try listening with the OEM horn? The CD horns on Urei's and later Altecs is a completely different design and principle, I don't think it's as simple as removing the slats. Yes the 604B is a true multicell and many consider it better, although I recently found out this is not unanimous.

I don't have distortion problems and have not damaged my tweeters since removing the caps.

You and Billfort being neighbors ought to get together and compare notes. I'm from Toronto and next time I get up there maybe I'll try to get in touch with you guys.

Dave

**tonymjb** · February 17th, 2004, 06:48 PM

Hi Dave,

Yes, I put 2 half circles of felt on the inside of the Bakelite ring. I was told by reputable people that mounting a driver behind the baffle causes resonances. The Bakelite ring looks just like that, just scaled down.

I did not try the original horns with dividers. The reason is that one of the horns is glued solid into the tube and I was afraid that I could break the tube. I would for sure file the throat ends of the dividers. I did not feel that the modified horn was bad off axis. Some people claim that multicell horns shout but others love them...

I may remove some felt first and see what it does. Or, just put the loading cap back with some felt behind it. Ideally, I would do it in stages. But removing the drivers is a pain: I have to put the speakers flat (each some 200lb = 90kg), the drivers are front mounted, they have to be disconnected, etc. - it takes about 2 hours.

Tonymjb

**jmarkwart** · February 17th, 2004, 07:05 PM

Tonymjb,

In your last post you mention the reason for removing your 604E loading cap - "The loading cap was removed to see if it eliminates potential standing waves between the diaphragm and the loading cap." Your comment reminded me of a French web site I stumbled across a year ago where the owner of a pair of 604-8Gs was polishing the surface of his phase plugs to a mirror-like surface. His rationale for this intrigued me so much I had a friend translate the text to english and found out the 604-8G owner too was concerned about standing waves. My enthusiam waned though when he reported the distance between the plug and loading cap would create standing waves at 100kHz, and he felt "they might be creating undertones that would beat with frequencies in the audible range". Sound waves don't work that way, they create overtones. Also, at specular frequencies above 500Hz or so, curved surfaces don't produce direct reflections, i.e., standing waves. Still, he did have pretty polished phase plugs!
I have removed Altec loading caps in the past and both measured and listened to the results. The biggest effect I observed was the lowering of the low frequency resonant peak of the diapghagm and the Q, or rolloff, of the diaphragm. I didn't note any loss of high frequency energy. It's possible that the newly lowered Q of you HF diaphragms and the notch filters in your crossovers are conspiring to lower the upper midrange output around 2-5kHz and this is what you're missing, not ultimate top end. Have you measured the output differences?
Jeff

**tonymjb** · February 18th, 2004, 06:54 PM

I have measured on axis response of both modified and unmodified 604E drivers. There is some difference but not large. Is there a way to post images here? Or, any recommended image hosts (it looks I can link an image file address only).

I wanted to measure the HF drivers without crossover but could not figure out how to set the lowest frequency in LAUD (the lowest display frequency can be changed but the measurements still start at 20Hz).

Tonymjb