Skip to main content

SEM Adventures Pt. 2: Cooling and PSU Overhaul

title

Refurbishing the cooling and power supply of a scanning electron microscope.

This is the second in a series of posts that follows the progress of refurbishing a thirty-odd year old Cambridge Instruments Stereoscan 250 Mk III scanning electron microscope (SEM), together with learning experiences along the way and, hopefully, some fun hacks also. For an introduction to the machine, basic principles of operation, issues encountered and initial results, see Part 1.

Tired and disintegrated

title

Upon dismantling vintage electronics you are all too often greeted by disintegrated foam and rubber. The SEM was no different and above we can see a mounting bracket for a centrifugal fan, with an anti-vibration mount where the rubber has almost entirely perished.

title

The disintegrated mounts left two large fans loose and rattling around inside the bottom of the machine, which was obviously far from ideal, given that vibration becomes particularly problematic at higher magnification levels. Of course, having reasonably powerful fans tethered only by their mains cabling and essentially mobile inside equipment, is generally not desirable either!

title

Replacement anti-vibration mounts were sourced, fitted to the brackets and the fans secured in place.

title

Unsurprisingly, the foam media of a large filter that sat just behind the fans had also perished, resulting in particles of disintegrated foam being blown around inside the cabinet. This foam was cleaned up along with the filter panel, new media sourced and fitted in place.

title

In addition to the centrifugal fans there are two axial fans which operate in a push-pull configuration, forcing air through a rectangular box section which contains an array of linear voltage regulators. One of these fans was working intermittently and it was decided to replace both.

title

Original fan in the background, with replacement ebm-papst 9906 in the foreground

The original ebm-papst part appeared to still be in production, but not that easy to source. So a decision was made to replace it with a more modern alternative which provided a similar air flow, the ebm-papst 9906. A more compact part, this also gave improved access to adjacent connectors.

Fresh electrolytics

title

Electrolytic capacitors have a limited lifetime, due to the electrolyte evaporating at a rate which is determined by factors such as the applied voltage, ripple current and ambient temperature. Therefore it's common practice with vintage electronics to simply replace all the electrolytic capacitors and particularly where performance is a consideration, e.g. in high end audio equipment.

As with vibration, ripple and noise are not desirable in an SEM and so the decision was made to replace all the electrolytics in the power supply. Above we can see the PSU capacitor PCB removed from the machine, with one of the original eighteen (!) Philips 10,000uF/63V capacitors removed and next to this a replacement Kemet part. These are an old DIN standard solder pin form factor, with 5 terminals, 2 of which for +/- connections and the others simply for mechanical support.

There were also a 470uF/250V 4-terminal DIN electrolytic capacitor, a 10,000uF/25V rated one replaced with a 10,000uF/63V snap-in, 2x 10uF/450V and a 470uF/40V axial capacitor replaced.

It was quite a relief to discover that new replacements could easily be sourced for the old capacitors, without having to resort to modifying the PCB in order to mount more modern footprint parts.

title

Unfortunately, when it came to removing one of the original 10uF/450V capacitors, one of its terminals took with it the PCB through-hole plating.

title

Leaving a bit of a mess and nothing to solder to on the board underside.

title

However, a simple fix was achieved by bending one leg of the somewhat smaller replacement capacitor, removing solder resist from the track on the PCB top side and then soldering to this.

title

Two of the 10-way Molex power connectors had clearly been subjected at some point to carrying more current than they were rated for, with browning of the plastic and some of this breaking off. Above we can see one such connector after it had been removed (and lost a few pins in the process).

title

These were replaced and along with them the equally scorched mating cable mounted connector housings and pins in the SEM cabinet.

title

With that the capacitor PCB was ready to be fitted back into the main cabinet.

title

Above we can see the EHT power supply for the electron column removed, with the space where the capacitor PCB fits located below.

title

Once secured into position the Molex power connectors were reconnected.

title

At the rear of the main cabinet there is a module which indicates the status of the many different power supply rails and, happily, upon powering up this illuminated as it should.

Solid foundations

The power supply and, where required, cooling are really the foundations for electronic equipment and when not performing as they should be, all other bets may be off; there is little point in trying to fault find and improve the performance of circuits, if the power supply and cooling are degraded.

Now that we've first confirmed basic operation, before refurbishing the power supply and cooling, we can turn our attention to other improvements.

In the next post we'll replace the electron gun tungsten filament, the faulty current control for this and the oil in the rotary vane vacuum pump. Following which further testing, before looking at available access to signals and control, considering how we might add digital image acquisition.

— Andrew Back

Open source (hardware and software!) advocate, Treasurer and Director of the Free and Open Source Silicon Foundation, organiser of Wuthering Bytes technology festival and founder of the Open Source Hardware User Group.
DesignSpark Electrical Logolinkedin