Just in case you thought the engineering staff at Alpha was off honing their contesting skills or participating in a great DXpedition somewhere - we'd like to update you - yes REALLY update you on our Alpha 4040 tuner project. Our tuner is getting close to shipping - sure you've heard that before, but take a look at the details and the pictures below and form you own opinion.  We'll show you some of the guts that are going into making this the best tuner ever offered to the amateur market.

It hasn't all been a walk in the park getting to this point - we have stumbled a few times, and had to go back to engineering first principles and test, test and test again.  Right up front however, we want to admit to our biggest and worst mistake in the whole design process - and it was our first mistake too.

It wasn't the switches that you will see needed to be redesigned, or the inductor that has had some changes or the boards that needed to be turned again, or even the software..... 

Our biggest mistake in the Alpha 4040 tuner process so far was to underestimate the time and energy it would take to build our DreamTuner. Our 4,000 watt spec proved much more difficult to deliver than we originally thought. 

But, even though we've taken a bit of a beating on the forums, we stand by the Alpha way.  We won't ship this project until we feel it's Alpha quality and you'll be delighted. Once the DreamTuner leaves the factory, we don't want to see it again!  I'm sure everyone knows, but we didn't take a dime in deposits - all we have is a list that has grown to hundreds of future owners of the DreamTuner.   

We are in the midst of an intensive testing phase. High power testing.  Here are some pictures of parts of our test setup.

Lab Bench Test Setup

Component Test Setup 

All of the switching components were tested previously using a mixture of techniques at DC or 60Hz (hi-potting and over-current with DC) in an attempt to make sure there was plenty of margin. Of course, as the phrase on cars goes "your mileage may vary". It turns out (after building some fairly elaborate test setups) that there were some shortcomings when the components were tested at RF. These manifested themselves either as outright failures or as problems that would have resulted in reduced lifetime. 

At the beginning of the process, when we found the enabling element - the Vacuum Variables, we built a tester that would cycle the capacitor to the minimum capacitance, HiPot test it at 5,000 volts, record the leakage, run the capacitor to the maximum capacitance, and do the same test.  We tested multiple capacitors to failure.  When we saw that we could cycle these capacitors a minimum of 50,000 times without cracking the vacuum bellows, we approved the capacitors. They're rugged and we've been very happy with them for a long time.

Our edge wound silver plated copper inductor did well until we really cranked the power.  We found two problems with it - 1. The opposing wipers were causing half a loop in the circuit, and we needed to find a way to take much of the inductor out of the circuit when it wasn't needed.  Two problems found late in the game, but fixed.

Switches were a whole different animal.  Here is an example of a voltage breakdown failure in one of the beautiful switches we designed from scratch. Although there appeared to be a lot of margin in the design, something happened at 28 MHz when we applied the highest voltage that could be expected. RF at 28Mhz and very high power looking into a big mismatch.

 

lot of relays! Here is one that looked promising for a while, but eventually had "corona" occur on the open contacts, that lead to heating and ultimate failure. This relay is used in other company's tuner products, and so we thought it would be worth testing.

We tested many more. One that is touted as a "40kW" antenna switch emitted a strong odor of fish when the current through it was pushed close to what we expected at 4kW, into our specified SWR. Parts of it were getting so hot they were causing the dielectric parts to outgas. 40kW seemed fishy indeed...

Testing high up on 10 m

This "software tuner" has allowed us to look at a lot of things quicker than we could in the lab. Here is an example of the component stresses when the tuner is facing a 10:1 load at 28.5MHz and asked to pass 1.5kW of power. Without going into too much detail, the x-axis at the bottom goes from 0-360 degrees, representing one trip around the Smith Chart, so all the possible impedances that lie on the 10:1 swr circle. The important voltage stresses are shown on the left, for the two capacitors and more interesting, the output connector- which in this case sees close to 1kV rms- worth thinking about! The currents are shown on the left- in this case the inductor current goes up to 17 amps! Many of these values can be adjusted by changing the algorithm used to derive the component settings. This process is going on, and the difficult areas are being subjected to further testing in the lab, as mentioned above.

The final piece is the software. I think it is safe to say that no tuner ever made, commercial or amateur, has had such an ambitious computer focus. The idea has always been to create a tuner that is in line with all the modern transceivers whereby the functionality can be upgraded over the internet. The 4040 is the first Alpha product to tackle this head on in quite this way. Needless to say, it has provided us with another "educational moment" in product development. Again, there is probably too much detail to go into here. Suffice it to say, we have a team of people working on various parts of this, with a plan to bring it all together soon.

In case all the above sounds too negative, we have found solutions to all the problems that we have encountered - there is no impassable roadblock that is staring us in the face. Just the myriad of details required to pull the tuner together into a product worthy of the ALPHA name.

Don't just take our word for it - take a look at some pictures of the tuner in the lab being tested and the videos of it in operation. Here's a youtube video showing the elements in action.  

Inside the 4040

When Alpha releases this product we are certain that it will be the most thoroughly tested matching network ever released to the amateur market.  We feel this is essential based on the performance specifications we will claim and the price of the product.  This will be a tuner like no other seen before.

We're very happy with the quality we've built in, and very unhappy about the time it has taken to get this far.  As anyone who's in the engineering business, everything takes a long time.  Especially when you're building elements from scratch.

Well, that's it for now.  Back to work.