Sleeve Valve Plus

Happy New Year.

Sleeve Valve - The picture below shows a new sleeve valve I am testing. The sleeve valve is much smaller than the ball valve I have been using. I have high hopes this will give me the ability to evacuate the chamber, close the sleeve valve and not need to re-evacuate for a long period of time. I also think the sleeve valve is very well built and looks appropriate for the job.

Other work - I have done a lot of testing on the new cable feed through described in my December 31st post. It has a very low leak rate indeed. I also have been testing an absolute vacuum sensor that is installed inside the vacuum chamber. This sensor eliminates the changes in barometric pressure that contaminated the readings I get from the differential vacuum sensor in my vacuum test set-up. Between the sleeve valve and internal vacuum sensor, I can eliminate all the Viton tubing external to the enclosure so I expect an even lower leak rate. Finally I am getting a solution I am happy with.

One more thing - I just started testing a version of my acrylic vacuum enclosure with improved optically clear solvent welded joints and thicker end plates. The new end plates are 9 mm thick instead of 6 mm thick.

More on all this later.

Sleeve valve near centre of image.

New Cable Feed Through

Near the center of the image below is the new cable feed through. It has solid metal pins instead of a flex cable with insulation. The insulation on the flex cable is one of the expected sources of vacuum leak (see December 7th post). The metals pins are surrounded by white Agilent Torr Seal epoxy. This epoxy is specifically formulated for vacuum applications with very low off-gassing.

The epoxy should finish curing by tomorrow morning and I will test it. I have high hopes.

New cable feed through.

A Little More Information

For a while I have wanted to put one of my acrylic vacuum chambers under water to determine the location of any air leaks. Good news is the solvent welded joint between the acrylic cylinder and the end flanges is leak free. I am still studying techniques to build an optically clear joint at this point but this would be a cosmetic improvement only because the joints already appear to be air-tight. More on this later.

But, I found a small leak at the viton seals. The end plates are attached to the flanges on the acrylic cylinders using four fasteners as seen in the picture below. The end plates are 6 mm thick acrylic which flex just enough that there isn’t sufficient pressure on the viton seal at the mid-point of the end plates. I have ordered 9 mm thick end plates to test if this solves the problem. I’ll provide updates later.

With thicker end plates and a better feed through as described in my December 7th post perhaps there is life in the acrylic version yet. Parts for the improved feed through are being couriered to me as write this blog.

Other progress over that last two weeks includes testing a breadboard version of the Microchip Technology AT42QT2120-SU capacitive touch sensor integrated circuit. I plan to use this for a remote control. This will be an alternative to using the iOS remote described elsewhere on this website. Again more on this later.

Acrylic vacuum chamber under water.

Progress on Two Fronts

In my November 26th blog post, I described a slow vacuum leak that I thought was due either to reusing the Viton seals (black ring in photo below is rear seal) or a leak in my vacuum testing apparatus. It turned out to be something altogether. After testing all the components individually, I convinced myself the leak is in the cable feed through.

Near the centre of the picture below, there is a flex cable passing through an opening in the aluminum plate. The opening is sealed with epoxy. I speculate that flex cable insulator is just porous enough to allow a small amount of air to enter the vacuum chamber.

A commercial vacuum chamber cable feed through is very expensive but one feature they seem to have in common is metal pins encased in epoxy. The pins don’t have any electrical insulation. I think the electrical insulation on the flex cable is the weak link. My next step is to redesign the cable feed through to use uninsulated metal pins.

As a side note, I use the same cable feed through on the acrylic version of the enclosure so after I redesign the cable feed through, I will test it on my acrylic enclosure design - there may be life in the acrylic version yet.

Rear aluminum plate of vacuum chamber showing flex cable surrounded by epoxy.

My other progress is with the frequency response testing of DÆ Phono Preamp v2.0. In my November 30th blog post, I described the fine tuning of the RIAA filter capacitors using a B&K 880 Precision LCR meter. Following fine tuning the filter capacitors, the filter matches, the ideal RIAA response to within 0.11 dB from 20 Hz to 20 kHz and within 0.03 dB from 50 Hz to 20 kHz. The low frequency roll-off is caused by the DC blocking capacitor and could be further improved with a larger capacitor. The measurement is 0.03 dB may be a limitation of my measurement process. Next, I will look to improve my measurement process.

DÆ Phono Preamp v2.0 Fine Tuning

This week I received a new B&K Precision 880 LCR meter. This meter is capable of measuring capacitors, inductors and resistors with an accuracy of 0.1%+2 digits.

A phono preamp RIAA filter requires two resistors and two capacitors. The accuracy of the RIAA response is sensitive to the accuracy of the components. 1% tolerance resistors are readily available and some 1% tolerance capacitors are available. To get a more accurate RIAA response than what is possible with 1% tolerance parts in common series values (E96 resistors for example), multiple resistors and capacitors can be used in series and parallel combinations. As an example if two equal value components are combined to make the desired resistor or capacitor, the effective tolerance is improved by 1/√2; so two 1% parts make a 0.7071% tolerance combination.

My resistor tester project provides the measured tolerance statistics for 1% metal film resistors to aid in understanding the likely results of combining resistors.

As an alternative to combining several parts, the DÆ phono preamp V2.0 uses two capacitor multiplier circuits. The capacitor value of the capacitor multiplier circuit can be trimmed using a potentiometer. Using the B&K Precision LCR meter the capacitors can be trimmed to 0.1%. It would take one hundred 1% components connected in parallel (or series) to achieve a 0.1% tolerance component.

Next I will test the frequency response of the phono preamp to confirm the capacitor fine tuning provides the desired very accurate RIAA response.

Fine tuning phono preamp with B&K Precision 880 LCR meter.

Ladder Attenuator Installed - Its Very Quiet But..

I added a sample relay switch ladder attenuator into the vacuum chamber. It is very quiet but there appears to be a small leak. The leak may be caused by re-using the Viton seals a second time or possibly a leak in my vacuum test set-up. The leak is very tiny but I still need to track it down.

One step forward, two steps back…

Relay Switched Ladder Attenuator installed in glass vacuum chamber.

Success with Vacuum Test

Success!

After many trials, the attenuator pictured below successfully holds a vacuum for 24 hours and likely much longer. The attenuator has a glass cylinder and aluminum end plates. The end plates include the flex cables openings to allow electrical connections between the vacuum chamber and the control and connector printed circuit boards.

Next step - build a new set of relay switched ladder attenuator sections.

Successful vacuum test on this attenuator including glass and aluminum vacuum chamber.

Initial Post

First post! I hope to add more posts as I got along.

Currently working on aluminum/glass version of vacuum enclosure for the stepped attenuator.

The acrylic version had too high a leak rate to allow the vacuum to last for months/years. Aluminum and glass are much better vacuum materials. Glass is 3 mm thick and aluminum end caps are 6 mm thick. Laser cut Viton seals are used.

Prototype aluminum/glass version of stepped attenuator enclosure

Prototype acrylic version for comparison