XG-8 Mk3s, gas problem

Read more on the Dayton Wright XG-8 Mk3s in our home audio section

Mike Burns2004-02-23 22:23

I have just purchased a set of Dayton Wright XG-8 Mark 3s. One of them seems to be in fine shape with good tension on the Mylar(?) seals that keep the fluorine gas in. The other has 2 small holes that have let the gas out. I am intending on repairing the holes and refilling the gas. Each speaker has an air valve with compression fitting (like a car tire) that seems to be used to fill them with gas.

My questions are these :

1. How much (in volume or pressure ) fluorine goes in one of these? Do I need to void the air that is in there now before putting in the fluorine?

2. Can I use a butyl rubber patch kit to seal the holes?

3. The original owner told me that he had a ”Threshold” mod done to the step up transformer. He advised me to use an amp that will handle a 1 ohm load. Is there any information out there on this mod?

4. It is my understanding that the difficulty of driving these older panels was derived from the variable impedance they presented not by the low ohm rating. Am I off base for these speakers?

Any help or even direction where to get help would be greatly appreciated.


Dave Pyatt2004-02-23 22:23

I own a pair of XG-8 Mark 3s and I remember speaking to Dayton Wright about the fact that one of my speakers seemed to have too much gas in it. The force required to depress the bag was much greater for one speaker. Dayton Wright\'s advice was to turn the speaker upside down and remove or depress the valve stem and release the gas until it was at atmospheric pressure. The speaker has to be turned upside down because sulfurhexafluoride gas is denser than air. So I imagine that when he fills the bag, the gas will fall to the bottom and simply force out the air.

By the way, I drive my speakers directly with an electrostatic amplifier that was written up in the magazine The Audio Amateur,issues 3 & 4, 1977. The design is by David Hermeyer. Prior to that I had tried a Crown DC300A (150 w/ch) and a Dunlap Clarke (also 150 w/ch) with the matching transformer. I have never regretted building this amplifier although it uses 8068 output tubes which are getting expensive and hard to find. I recall an article written by Mike Wright in the magazine Electron - long since extinct - in which he said that something like 75 % of the distortion in the speaker system was due to the transformer. I believe him because the improvement in transparency was immediately noticeable. Obviously with this amplifier I am not using the piezoelectric super tweeters which I never did like.

Paul Young2004-02-23 22:23

Here are some answers:

1. The volume /pressure of SHF gas is very simple. The speakers internal pressure should be exactly equal to the ambient air pressure, which can be seen to be true when the outer skins are neither convex or concave. The volume (in cc.) will be about 10% less that the calculated volume of the metal speaker frame. In actual day-to-day use, the ”perfect” volume is rarely achieved due to the fact that the speaker acts like a huge barometer and actually swells and shrinks as the ambient atmospheric pressure changes. This is normal and is not a problem until the deviation of the center of the skin from flat exceeds approx. 2cm. Then you can invert the speaker (very important!) and press the valve stem for ~5 seconds trick to get them flat again. This causes some eventual dilution of the SHF gas - but not much, even after years of adjustments.

2. Re getting rid of the air before adding SHF, this is only a problem if the speakers contain mostly air. To refill a speaker that is mostly air filled, it is necessary to SLOWLY push SHF into the box at its lowest point, while allowing the existing air to rise to the top of the enclosure and escape via a second hole. This hole can be of any type - as long as it will let the air out as fast as the SHF goes in. You can create an air exit hole burned into the outer skin with a soldering iron tip - then tape it over with some 5 cm wide clear plastic tape of the type used in shipping depts. to pack commercial packages. This will hold for years - just inspect it from time to time. The butyl rubber patch will likely not stick to the outer skin very well. Even QUAD uses this type of tape and it has proven to stick for 25+ years.

3. Re the Threshold Mod. - I have no data on this.

4. Re the possible difficulty in driving these speakers - their impedance is both significantly non-resistive and at certain frequencies reaches a very low value. If your amplifier has trouble with reactive or load impedance loads, it won\'t like the IM-10/XG-10 combination. The lowest impedance value I measured was 2.4 ohms at 14kHz. Many modern solid state amps will drive such a load with stability. Some amps will current limit their output due to the low impedance at some point as you increase the volume, but all amps eventually limit their output under some condition, so this is really a matter of ”will it play loud enough before it distorts”.

Rob Bruce2004-02-23 22:23

I have a pair of XG8 MK 1 speakers. I refilled them a year or two back. Yes, the SF6 gas is heavier than air. What I did was cut two small holes in the top of the mylar (The Mark 1s did not have a valve), and stuck a hose from the tank of SF6 in one side, and let the air vent out the other. I weighed in rougly 1.5 pounds of SF6 per speaker. The XG10 manual recommends 1.1 pounds of gas I do believe, but I had a 3 pound tank, and decided that I might as well use all of it. This of course means that all of the excess gas escaped through the vent hole I had cut, but it did ensure that the speaker was full of gas at standard operating pressures. Fill until the mylar starts bulging, stop the flow of gas, and let the excess pressure flow out. Repeat until the required amount of gas is in the speaker. Don\'t forget to be in a well ventilated area, SF6 will also displace air in your lungs if inhaled, (think of carbon monoxide poisoning).

Post a reply

Your name will appear on the website next to your contribution. Your email address will only be used to contact you if something is wrong with your contribution. It will not be shared with others.