Clifton Laboratories 7236 Clifton Road  Clifton VA 20124 tel: (703) 830 0368 fax: (703) 830 0711

E-mail: Jack.Smith@cliftonlaboratories.com
 

I've revised the non-linear transformer page to add three inexpensive Stancor transformers and the very high performance (and likewise very expensive) Jensen transformer. Click on the link in the above sentence to view the page.

I've been remiss in not updating this page recently, but I've been trying to get two new projects closer to completion.

By the end of the week, I should receive prototype/final printed circuit boards for the Norton amplifier and a high performance, expandable 4/8 output receiver multicoupler.

I've said prototype/final because I've been through several iterations on these boards already and I've gone ahead and ordered a small quantity of production quality boards, with solder mask and silk screening. Absent some blunder on my part, these boards should be ready to make into kits in the next few weeks. The major pacing item on the multi-coupler is the sheet metal, which has a several week lead time and I can't complete the layout drawings until I have the boards in hand and verify dimensions and accessibility.

I've also received four audio transformers to analyze and have completed performance measurements and started adding the new material to my non-linear transformer page. I hope to wrap it up in next couple days. One transformer I looked at in a Jensen DIN-2LI, and it's an amazingly good performer, with THD below the ability of my HP 8903B audio analyzer to measure over most of its frequency range. The individual  transformers packaged into the DIN-2LI run around $80-90 each so it's not suitable for every application, but it's the best  transformer I've measured. For a more budget-acceptable combination of price and distortion, I have a couple of recommendations on the non-linear transformer page.

I recently acquired a couple of Kepco BOP bipolar power supply / amplifiers and used them today to measure the forward drop of a 1N4007 silicon diode and a SR2010 Schottky diode at current levels up to 5 amperes using a 300 μs pulse. This technique obviates the junction heating Vf reduction found in my DC measurements. I've added the new data at the end of the 1N400x Diode Family Forward Voltage page.

I've added a SPICE simulation plot to the RC snubber discussion at Diode Turn-On Time.

I've expanded the relay snubber discussion at Diode Turn-On Time to include RC snubber networks, designed using the Zobel network technique. It works well, but requires a non-polarized capacitor.

The myth that a 1N4007 power diode isn't fast enough to work as a snubber and clamp the inductive kick from a relay has surfaced again. I made an extensive series of diode turn on time measurements in December 2007 showing that the difference in turn-on time between a 1N4148 signal diode and a 1N4007 power diode is small at best. (Note that I said turn-on time, not turn-off time.)

To address this myth in even more detail, I've revised the diode turn on measurement page to include measurements of the inductive "kick" of a typical relay when controlled by a 2N7000 MOSFET transistor with three conditions:

• No diode snubber
• 1N4148 snubber
• 1N4007 snubber

These tests confirm what the turn-on time measurements showed; the 1N4007 works just as well as the 1N4148 as a relay snubber. And, the 1N4007 will handle much more current than the 1N4148, which may be a consideration in some instances. The page also looks at how much the relay's release time is lengthened by adding a snubbing diode.

For all the details, see http://www.cliftonlaboratories.com/diode_turn-on_time.htm.

I've revised the DC Transformer Saturation page to show the effect of DC in the winding upon intermodulation products for a Coilcraft PWB1010-L wideband RF transformer.

At the request of G3TJP for a shorter Z10000 manual, I've prepared an alternative version. It concentrates on using the Z10000 as an adjunct to Elecraft's K3  transceiver IF output port. You may view the alternative manual by clicking here, or via links on the Documents page or the Z10000 page.

I've also added a new option for Z10000 amplifiers—a die cast enclosure. Details at the Z10000 page.

I have a limited number of Z100 kits remaining in inventory and ask that anyone planning on ordering one contact me before sending money.

I've added a corresponding note on the Z100 page.

As normal for the first of the month, I've moved the March 2009 updates to the archive, viewable by clicking here or via the link table at the top of this page.
 

I put together an active antenna kit with the "Revision B" printed circuit board for a test builder over the weekend and the kit is now in the hands of  the builder.

I have parts for one more Rev. B prototype active antenna and, depending on the results reported by the first prototype builder, I may ask for a second volunteer builder. The next step will be to make a few changes in the printed circuit board layout for Rev. C, which should be the production version. I also have a very small change or two to the power coupler as well. Assuming all goes well, the active antenna should be ready to go around the end of April.

I've also worked on  the Norton amplifier design this week. I plan to shrink the PCB to fit into the same weatherproof enclosure used with the Active Antenna as a standardization measure. The Norton amplifier needs one more test PCB as a safety measure before ordering production boards, and I plan to finish  the layout and order another prototype round before the end of next week.

As I've said before, the Norton amplifier is quite a remarkable device. I re-measured the one prototype I built late last year on the first PCB and found IP3 above +40 dBm and IP2 above +80 dBm.

The Norton amplifier will be available as a kit or assembled amplifier, with an optional enclosure, either a die-cast indoor enclosure or a die-cast weatherproof enclosure. The amplifier will be provided with an option for external DC power feed or duplexed DC feed over the coaxial cable. This is particularly useful for outdoor use.