Monday, December 6, 2010
Currently only 2 render clients are online as the front view is rendered by flightsim server. In the final configuration, I'll have the server displaying the instruments panels, control panels and the GPS on two separate monitors and the front view will be rendered by a dedicated machine. The whole display setup in its final version will be composed of 3 monitors for the 135 degree panoramic cockpit view (top row) and 2 additional monitors for the instrument panels (bottom row).
For the radio stack, autopilot control panel and the switch panel I am using separate hardware panels (by Saitek). The flight controls are Pro Flight Yoke system and throttle quadrant and Pro Flight Rudder pedals by Saitek.
The major item left to be completed is the display setup. I have to figure out how to raise the panoramic view monitors so I can have the instrument panels on the bottom row. I'll post pictures once the whole setup is completed.
I've done some crude testing (using 3DMark06 ) to compare Quadro FX 4600 vs. GeForce 8800 Ultra. Both cards are based on the same G80 nVidia chip and are very similar - the Quadro is the "pro" version with a few extra hardware features enabled and using special drivers aiming at high accuracy rendering. The Geforce is the "Gamers" version where high FPS is the primary goal - the G80 nVidia chip is a bit overclocked in the "Ultra" version. The test was done in 1920 x 1080 dual monitor setup /w 2x2 AA enabled and bilinear filtering. As it turned out the Geforce 8800 Ultra is the faster card so I am using it in the main flight sim box.
Tuesday, September 28, 2010
Both transverters share exactly the same design and PCB with only a few differences in the BOM - the frequency of the Local Oscillator (180 MHz for the 2m version and 400 MHz for 70cm) and the the components of the Band Pass Filter (different set of components for each filter, changing their frequency range).
I started with the 2m transverter kit from Ivan, VE3IVM. It was an easy and pleasant build - Ivan provided a partial kit - PCB, LO, BPF coils, balun transformer for the bridge, both mixers and a power splitter- the rest of the BOM is from Digikey.
The top side of the PCB. The 4 coils of the BPF are shielded, each in its own compartment. It is important that they are centered and lifted from the ground plane in order to increase the Q and minimize losses. The other components are the LO, both mixers, RA SMA connectors and the voltage regulators - +6V and +5V. The most time consuming task here is to make and install the shield for the BPF coils - a certain level of precision is required while fabricating the RF shield as the tolerances are not very large. The top cover is soldered during the finalization stage. I used tin-plated brass sheet to make all RF shields - this material is very easy to form and solder.
Most of the components are installed on the bottom side. Again, a lot of attention is required when building the BPF and the bottom shield for it. After adjusting the BPF, the bottom shield is installed and the filter is checked and re-aligned again as the shield might de-tune it a little. The attenuators are pretty small and should be soldered very carefully. Same thing goes for the 100 Ohm 0604 bridge resistors. The two amplifiers should be soldered carefully to avoid over-heating. Ivan provides specific instructions for installing the variable caps - no flux (or very minimal) should be used and the board must be washed with alcohol or flux remover to avoid mechanical damage to the trimmer capacitors. This picture shows an incomplete and bypassed Low-Pass filter on the low-side output. The jumper is removed and the rest of the components are installed during the finalization stage and after final alignment of the BPF.
This is a plot of the BPF frequency response from the initial alignment of the filter. I was aiming at 143 MHz - 153 MHz range. The BPF filter design allows for almost 10 MHz bandwidth in the 2m band and 20 MHz at 70cm. It can be moved a few MHz up or down from this general range if needed just by adjusting the variable BPF capacitors. During the alignment, it is important to achieve as flat as possible band-pass response and proper shape of the skirts. MyVNA really simplifies the use of transverters and takes care of all calculations - the user must enter the LO frequency and frequency range for the high side and myVNA translates this to the VNA's HF working range, while making a plot in the transverter's frequency range.transverter. The most visible difference between the two transverters is the configuration of the band-pass filter. For instance, on the 2m transverter there are some BPF capacitors mounted on the top side just under each coil - they should be installed before the coils are soldered. These are omitted in the 70cm version.
The BPF frequency response of my 70cm transverter. The insertion loss at 70cm is higher then the one of the 2m transverter at about -13 db. It can be lowered to up to -6 dB if some of the bandwidth is sacrificed but I wanted to cover the whole 430 - 450 MHz range.
The finalized PCB. Shown here is the completed Low-Pass filter and the RF shielding of the BPF. As Ivan, VE3IVM noted in his build instructions, after installing the BPF shields they will slightly de-tune the filter. As expected, I had to re-align it again. A fence-type shield was installed to divide the up-convert and down-convert paths in order to reduce stray coupling and cross-talk.
Almost done. I installed a toroid choke on the power line. Ivan has designed the PCB to fit in a compact extruded aluminum box - Hammond 1455C801. The board fills up the space with very little room to spare - just enough for the RF choke. It was a bit tricky to drill the holes for the SMA connectors in both panels so they match perfectly the RA connectors - the board has a small horizontal play which helps it a little. A small piece of conductive RFI gasket material (neoprene covered with metalized fabric) is pinched between the base of the RA SMA conectors and the front aluminum faceplate in order to improve grounding of the aluminum enclosure.
This is the front panel. The labels on the top row are for Transmission mode measurements and the ones on the bottom - for Reflection. I made the small jumper (needed to configure the transverter for Reflection measurements) out of semi-rigid hand-formable RG-405 coax and used a spacer made from a strip FR4 board to fix the distance between the connectors.. The Hammond box is assembled in a way so the plastic trim for the front and rear panels is on the outside (instead of the "normal" way - between the enclosure and the panel plate). This improves the RF shielding as the end panels are in direct electrical contact with the enclosure - there is no gap.
The rear panel and power cable. The Hirose connector for the power cable is the same type that I am using with my RF-IV test head. The accessories connector on the VNA's back panel provides +9V DC and I can power the RF-IV or the transverter directly from there. Two small semi-rigid coax jumpers with Male-BNC to Male-SMA are used to attach the transverter to the VNA's ports. Each of the transverter's ports has a Male-SMA-to-Female-SMA port saver installed (not shown).
Friday, September 17, 2010
I picked up a nice calibration kit on eBay for a very reasonable price (these kits are normally VERY, VERY expensive, but I guess I was lucky this time). The kit is equipped with 3.5 mm connector. This type connector was developed by Hewlett-Packard and it is used in lab instruments for microwave measurements. The 3.5mm connector is using air dielectric and it is machined to extremely tight tolerances. It is rugged (compared to the SMA), allowing thousands of connect-disconnect cycles with repeatable results. The beauty is that it mates with standard SMA connectors. On the flip side - it is a precision lab grade connector and one should be extra careful when handling and using - it could be easily damaged when mated with low-quality SMAs or by poor mating technique.
The geometry and construction of the 3.5 mm connector provides mode-free operation up to 34 GHz. My kit is actually made of two different sets - the male cal standards are rated for up to 26 GHz and the female for up to 9 GHz. Such kit requires proper care and storage. The male pin and female receptacle are supported only in the back of the connector (air dielectric) and can be damaged by rough handling/cleaning. Enemy #1 of these connectors are low-quality or already damaged SMAs. When mated with such, a damage is almost certain to occur.
Tuesday, September 14, 2010
An excellent technical article about calibration standard definitions by William Highton (Chemandy Electronics) can be found here.
The main reason for matching the calibration reference plane and the mating plane of the VNA test port connector is to allow industrial VNAs to measure whole assemblies and devices, including their connectors. In the N2PK VNA (at least with the original software) the reference plane is assumed to be where the physical OSL calibration takes place. Instead of adding negative offset to move the reference plane back to the mating surface of the VNA's test port connector, it is accepted that the plane is right at the rear surface of the calibration standard connector (the location of the actual Open-Short-Load). The error introduced by the electrical length of the pin inside the connector is not that great at lower frequencies - for instance, assuming teflon dielectric (VF ~70%) in a standard SMA connector used for a home-brewed calibration standard, an offset of 5 mm will yield about 0.5 degree phase error @60 MHz. Commercial calibration standards use air dielectric, but this puts greater mechanical requirements during the fabrication process. By calibrating the VNA so the reference plane is on the back of the cal connector and fully exposed, it is more convenient to measure a single component but it means that the DUT test fixture should be made of (or at least include) the same type connector as the one used for the home-brewed calibration standards in order to maximize the accuracy. The setup can be as simple as the component (DUT) is just soldered on the back of the "test fixture" connector, while measures are taken to minimize external error (minimal lead length). With SMDs, this approach works well as the device is practically located directly at the "reference plane" when the test fixture connector is from the same batch connectors used to make the OSL standards. Another method for calibration is to use the test fixture connector itself for calibration - for OPEN the DUT is simply omitted, SHORT and LOAD are created on the spot during each calibration step and finally the DUT is soldered (Still, stray C for the OPEN must be known)
Tuesday, July 27, 2010
The outer edge of the meter's plastic bezel is transparent and can facilitate easy side-lighting of the meters. As a "bonus", the plastic has "prismatic" finish around the edge and acts as a light diffuser, resulting in a more even illumination of the face. I attached two yellow-orange high-intensity LEDs (clear lens, from RS) with hot melt glue in the middle, right under each meter (~1/4" away from the bezel - this widens the beam and covers more area). It might take some experimenting with the position of the LEDs in order to achieve best results. This is actually the most difficult part as the space is very tight to get the glue gun in and to fix the LEDs in the right position. The LEDs can be any color - i just like the warm orange glow - it reminds me of the long gone tube radio era.
Two current limiting resistors (470 ohms) and a filtering capacitor (100 uF/10V, to avoid flickering as the 5V DC in the LED circuit is rectified by a single diode) were inserted in heat-shrink tubing and then attached with hot melt glue to the bracket holding the meters in place.
The LEDs are powered by the same circuit, powering the front panel "ON" LED. There is a second winding in the power transformer, used to supply the regulator board with 5.5V (hence the 0 to 15.5V adjustment range). The AC is rectified by a couple of diodes - one for the voltage regulator circuit and one for the front panel LED. This picture shows where my illumination circuit is connected on the regulator board - the positive lead is taken from the "hot" side of the front panel LED current limiting resistor (just after the diode V5) and the negative lead is on the 3rd pin (from the right, looking at the front panel) of the connector (black wire).
This is how the meters look in a well lit room...
... and in the darkness of my ham shack.
Monday, April 26, 2010
A possible solution to the SMA's durability problem is the so-called "port saver". This is an "in-series" SMA adapter - male-to-female. It is connected to the port connector and provides a "disposable" extension - after the adapter's front-end is damaged or worn off, it can be quickly replaced (and at a low cost too). This way, the actual front panel port sees very little use and there is no need for a complicated and time consuming replacement procedure.
Unfortunately, the "port saver" is not very well known/used adapter and it might be difficult to obtain. Thanks to eBay I was able to get a bunch for a couple of dollars a pop! An even better choice would be the stainless steel version. When disconnecting devices, 5/16 wrench should be used to prevent the port-saver from unscrewing.
Saturday, April 17, 2010
The Open, Short and Load standard are made of Male SMAs for PCB. The back side of the center pin of all three connectors is trimmed down to ~0.5 mm - the height of a 0603 SMD component. In addition, the SMA test fixture has its center pin also trimmed at 0.5 mm for improved solderability to the DUTs.For the SHORT standard, I used a small teflon washer (cut from the center dielectric of a MIL grade coax) placed on the center pin. The thickness of the washer is ~0.3-0.4mm. This washer serves as an insulator in order to raise the actual "short" to the very tip of the 0.5 mm center pin stub, (the back plane is covered with silver bearing solder to short the pin). For the LOAD standard I used the same Vishay FC series 50.0 ohm (0.1%) high-frequency precision resistor as the one in my BNC standards set (Digikey p/n FC0603-50BWCT-ND). This type of resistor is the preferred component for the LOAD - it is designed with maximum precision and minimal reactance at high frequencies.
For mechanical protection, RF shielding and easy handling, I used sleeves made of brass tubing (Stock #135, K&S Engineering, 3/8 x .014") to cover the back side of each connector. Each sleeve is 1 cm long and it is soldered on the inside wall to the 4 ground pins of the SMA connector. The end cap is made of closed-cell foam and it is held in place by the heat-shrink tubing. The OPEN standard also has an additional rear RF shield made of a tin-plated brass disc, soldered to the inside wall of the sleeve. The OPEN is more sensitive to stray RF than the others and must be fully shielded. Such shield could be installed on the LOAD as well. The SHORT does not need one as it is self-shielded.
SMA calibration kit. Heat-shrink tubing is used for color-coding and it gives a nice and smooth finished surface, protecting the brass from tarnishing.
In addition, I made a set of female standards in order to cover DUTs with any port gender. I used PCB female SMA connector (round base) and the brass sleeves were made of Stock #134, K&S Engineering, 11/32 x .014"
Saturday, April 10, 2010
Sadly, Mr. Craddock became a Silent Key on March 22, 2008 (R.I.P.). FCC was made aware of this at the end of 2009 and canceled his callsign. There is a two years waiting period mandated by FCC before the callsign is released in the general pool. This waiting period ended on March 22, 2010 and once the callsign was available, I immediately applied for it.
My application was granted today and my new (and probably last) callsign is now AE1S.
My last contact as AE1Z was HK1NK on Apr 3, 2010, 21:20 UTC on 10 meters / QRP.
It is time to start a new logbook! The change is a bitter-sweet one! AE1Z signing off!