Tuesday, March 31, 2009

Good by Junior, Hello Buddy.

I have not been blogging much lately because I have been doing a lot of hardware testing of the FLEX-3000 along with writing a rig.xml file to use PowerSDR (the "software" part of a software defined radio). Last week was very busy and good week in the test department. As with any beta testing, you are going to find or shake out bugs or make improvements in the product and that is what I have been doing. It really is not that important what those specific issues were as long as they get identified and incorporated into the production units before they get shipped.

The modifications to the FLEX-3000 hardware have been done for the production units, but Junior, who was a pre-production (beta) unit, does not have those improvements. As a result of those changes, Junior was no longer able to transmit more than about 5 watts out, which curtailed my on air activities the middle of last week. Junior need some hardware changes to make him whole again.

So, as sad as it may be, Junior is going back home to Austin. I knew when I got him that, that our time together would be short, but I was not expecting it to be less than two weeks. I will pack him up tomorrow and ship him back but not before I receive "Buddy" (the replacement FLEX-3000) later this afternoon.

Buddy is on the truck and I will have him on the air later this afternoon, probably doing some digital modes with Fldigi to test out the rig control file I created last weekend.

I'll post a notice when I will be QRV

Tuesday, March 24, 2009

Testing, testing, testing...

I didn't get on the air last night. The "day job" was to hectic and I had to cut the lawn before this weeks rain arrives later tonight.

I have been spending time testing some of the alpha/beta code for improving the TX and RX calibration tests for the FLEX-3000 and FLEX-5000. Mr. Bob is doing some masterful work for inclusion into the released version of PowerSDR that will officially support the FLEX-3000 that will come out the day the FLEX-3000s start shipping (in a few weeks). The algorithms are being tweaked to provide deeper nulls for image and carrier rejection for TX. For RX image rejection, the new RX-AIR (adaptive image rejection) code will eliminate the need to run image rejection calibrations for RX.

Hopefully today will be an easier work day and I will be able to play radio with Junior tonight a little bit.

Sunday, March 22, 2009

What a weekend.

This has been an action packed weekend. Lots of "playing" radio and the FLEX-3000 was a great toy to "play" with. I am VERY impressed with it. The receiver is HOT. It approaches the FLEX-5000 and K3 in performance and for $1600, there are no radios in that price class that even come remotely close to matching performance and features of the FLEX-3000. The receiver is very quiet. Better than the SDR-1000 and like the FLEX-5000 it is noise and DDS spur free. I found the frequency stability to be very good for the digital modes I was working. The TX audio is fantastic. I participated in the FlexNet today on 20m and I swear that Dudley, who has the other FLEX-3000 was on his FLEX-5000. It is that good.

The FLEX-3000 is a value that can't be beat. It has the receiver performance numbers of an Orion II which costs over $4900 fully loaded and comparable to the FLEX-3000! It beats a Icom 756 Pro III hands down and that radio costs $1100 more than the FLEX-3000. In addition, the great thing about the FLEX-3000 is that with the frequent software upgrades it just keeps getting better where all of these other "firmware defined radios" have significant limitations in there capabilities and the DSP is not upgradable, like it is with a software defined radio.

FlexRadio broke the price/performance paradigm again that all other radio manufactures adhere to so strongly. In their price structure, the more money you pay, the higher performance class of radio you get. To get one with IMD DR3 numbers in the +90 dB range, you empty your wallet to live in that rarefied air. It looks like FlexRadio puts performance first and only charges the premium price for a radio that has the greatest flexibility in I/O capability and modularity, something that not all hams want or need. What a concept!

You do not need a calculator to work the numbers. It is not even close. This radio is a winner by a long shot.

I need to get ready for the upcoming workweek. I will be operating the FLEX-3000 during the evenings and I will keep on reporting on Junior’s adventures. Good night.

Junior gets voice lessons thanks to Julius

I knew what I had to do to improve the TX audio and that was to dump the HC-4 mic element. Now do not get me wrong, the HC-4 does a very good job for what it was intended, but it just is not any good for the FLEX family of radios. If you want to pinch down your audio, the 10-band EQ will do it for you and you do not need a specialized mic element to accomplish it. I can make my PR-40 sound like a HC-4 with aggressive EQ cutting.

As I mentioned before, the headset has an Icom iC element that is a more "full range" mic element, but needs the 5 VDC for it to operate. I normally run my TX audio through a W2IHY EQ+ (and VERY good audio processor for ham radio from my friend, Julius Jones) to enhance the TX audio for my Icoms. The EQ+ provides the 5 VDC needed for the iC mic element and has several signal processing stages to work with too. So if I hook the FLEX-3000 to the EQ+ I can use the full range iC mic element in the ProSet Plus headset. It looks like I have a plan. Now it is time to execute.

So it was off to the junk box to find an old shielded 100BaseT Ethernet cable and a 5-pin DIN connector to make an EQ+ interface cable. The pin out for the cable just happens to be in the FlexRadio Systems’ Knowledge Center. http://kc.flex-radio.com/KnowledgebaseArticle50404.aspx.

I found the necessary parts, including my reading glasses so I can see how to solder and I made an audio interface cable. If you are going to use an old 100BaseT Ethernet cable to make a cable that conencts to the FLEX-3000 mic connector, then this Knowledge Center article will be helpful in identifying the pins based on wire color. http://kc.flex-radio.com/KnowledgebaseArticle50405.aspx.

Here is a picture of the FLEX-3000 connected to the W2IHY EQ+

As the youngsters say theses days OMG, did that make a MAJOR difference! The EQ+ is Junior's BFF now. I bypassed the 2-band EQ on the EQ+ in favor of the 10-band EQ in PowerSDR. I added just a smidgen of compression so I could use the downward expander to eliminate some of the ambient fan noise and last but not least, a dash of reverb for a slight mellowing. Ahhhhhh.

The iC mic element is much better suited for the FLEX family of radios since it has a 100 to 8000 Hz frequency response and no large “emphasis” at the 2 KHz point. The EQ+ can put out a lot of signal (gain) so I cut back the Mic input gain in the PowerSDR Mixer allowing the EQ+ to provide the AF gain at the earliest stage in the audio chain for the lowest THD (distortion).

Does it sound like a PR-40? No, of course not, but it also does not sound like $#*! like the HC-4 mic element did. Junior has found his full range voice and I could not be a prouder Papa.

Digital down, Phone to go.

Saturday was "Phone Day" for Junior.

I have a Heil ProSet Plus iC set of headphones that I use with my Icom radios that has the iC and the HC-4 elements in it, so I figured this would be the easiest way to get the FLEX-3000 on the air using phone. FlexRadio sent me one of the RJ-45 to 8-pin Foster (male) "pigtails" to connect the Yaesu headset connector I had for use with the FLEX-5000. Connecting the two together and plugging it into the FLEX-3000 was just too easy.

Again, I connected a dummy load and started some testing while listening to it with a second receiver. I was surprised. No TX audio output. Darn. Gotta be the cable, right? Well, no. It was a PowerSDR database setting. In USB mode, I had VAC enabled - DUH! One click of the mouse and Junior was speaking and it was not the digital warbling babble it was on Friday.

I cannot use the iC element in the Heil ProSet Plus without wiring up the voltage pin in the connector (the iC element requires a 5 VDC to work), so I decided to use the HC-4 element.

For those who are not familiar with the Heil HC-4 element, it is the "DX" element that has a very narrow frequency range which rolls off sharply @ 500 Hz and has a 10 dB "boost" at 2K Hz for extra punch. This element is poorly suited for the FLEX-3000 and FLEX-5000 because it was designed for older analog radios that do not have a flat TX frequency response like the FLEX family of radios do. So I tried to “EQ out” the undesirable characteristics of the HC-4 element that makes the audio sound hollow and tinny sounding. There is only so much an EQ can do and I was hoping that the 10-band EQ could get the job done. Did it? Keep on reading...

There was a Russian DX contest on, I fired up Junior and start to search and pounce using the Panadapter (aka “shooting fish in a barrel”) I made about 30 DX QSOs very rapidly. I did not have any issues cutting through the pileups with my barefoot signal. I was pleased with the initial results.

I though I had done a fair job setting the TX EQ for the HC-4 when I received an e-mail from Dudley (WA5QPZ) telling me that he had just received the other FLEX-3000 test unit from Gerald and want to set up a SKED. You bet I jumped at the chance to have the first “real” FLEX-3000 to FLEX-3000 QSO and he could make over-the-air recordings so I could hear what the radio sounded like.

Dudley soon had his FLEX-3000 fired up after an initial Firewire glitch due to a slight assembly problem (this is why we shake out the radios before shipping them to you) and he was soon on the air. Propagation between North Carolina and Texas was very good Saturday afternoon and we soon were able to find a spot on 20m where the Russians were not consuming more than their allotted bandwidth (they have wide signals and the Panadapter doesn’t lie).

Dudley made me a recording and the HC-4 sounded like [insert your own explicative here] once I could hear myself with a receiver that has a wider RX filter range than the Icom. So we started tweaking the EQ setting, turned off the Leveler and it started sounding better, but as I said to Dudley, “…the HC-4 ain’t no PR-40”. We announced that we were QRV on the FlexRadio Reflector and were soon joined by several hams, like Bob (Mr. CAT-man-do) K5KDN, Dale, WA8SRA, Daryl KA1VVT, Marvin KM4GQ, Dale KP2CZ and Chris KF6VVL (?).

After the SKED / roundtable was over, I was “bugged” about the TX audio quality of the HC-4. I was mostly annoyed that it did not do the FLEX-3000 justice since Dudley’s audio was EXCELLENT with the BM-5 headset he was using. I hade to make some changes…and soon.

Friday, March 20, 2009

Where are the baby pictures?

OK, OK, OK. I hear ya all. The battery died in the camera and it is recharging. I'll get some pictures of Junior loaded this afternoon.

UPDATE:
I added a few pictures to the blog entries below.

Junior's first digi mode QSOs

It did not take long. I threw out a few CQs and EA4EQD (Javier) replied to may call. I received a good signal report and was told my trace on his waterfall display was clean. What a proud papa I was of Junior's first foray out into the real world and a DX QSO at that!

The next QSO was my first (and possibly the very first ever) digital QSO between the FLEX-3000 and another FlexRadio Systems' software defined radio with Carl WC0V. Carl was running a SDR-1000 using a compromise antenna. Again the signal reports were good and we were chugging right along with our rag chew when the first problem I have had with the FLEX-3000 arose.

Actually, it was not a FLEX-3000 problem, but a Vista x64 problem with Com0Com, the open source virtual serial port pair driver used to enable CAT between MixW and PowerSDR. The com port pair "froze" and stopped communicating with MixW in the middle of my transmission. Therefore, there was no PTT or frequency control and the radio hung in TX. I had to stop MixW and PowerSDR, but upon the subsequent re-start of those applications, the Com0Com com port was dead. I had seen this problem before and knew that I must reboot Vista to get it working again. To heck with that. I reverted to trusty XP and kept on truckin’.

So the PC was rebooted and XP recognized the new hardware (FLEX-3000) and went through the driver installation. You do not see this with Vista, as it is done silently and the user does not know the driver is loading for the hardware if the driver software is already loaded.

I copied the PowerSDR database from Vista to XP and fired up the FLEX-3000. All was working correctly again. I really wanted to shake out Vista x64, but the issue with third-party drivers is going to be troublesome. I have a support request in to the Com0Com developer so we’ll see where this goes.

I fired up again on PSK-31 on 20m and had QSOs with:
WA7HDZ, Larry
IW0GNC, Rino

I switched over to RTTY on 20m running high power (~90 watts) and had QSOs with:
EA5DM, Jamie
YV5JBI, Juan
P40YL, Sue
CT1BXT, Fernando
KL7J, Les

The band started to die, so I switched to 40m PSK-31 and had a nice long rag chew with WA1NGH, Jack. While in QSO with Jack, I see a familiar call, W9OL. Bill, my good buddy from the FlexRadio Reflector. We had a short QSO before the band died, but that was the first digital FLEX-5000 to FLEX-3000 digital QSO on record.

I received a few e-mails from others who were listening and all said the FLEX-3000 look great on their waterfalls and the copy was 100%. I am pleased with the initial outing. After that, I went to bed a happy and content Flexer.

Junior finds his voice

Since I am planning to do digital mode operation first, I fired up MixW and got the VAC channels properly configured and started seeing signal on the MixW waterfall display. Turning down the RF output to a few watts (still running into a dummy load) I got prepared, paused for a moment and then executed the macro in MixW that turns on the transmitter, sends a test string of text and then shuts off the transmitter. I was finally transmitting! And startled.

Let me explain. The FLEX-3000 is in a very small package for a 100 watt transceiver, measuring 12.25" wide by 1.75" high by 12.25" deep. Stuffed inside this short transceiver (note the height measurement) is a 100-watt PEP 160-6 meter RF amplifier. RF amplifiers, being inefficient as they are, create heat. Lots of it and the transceiver has to be designed to remove the heat quickly. The FLEX-5000 uses an ingenious design that draws air from the sides and forces it through a plenum (air channel) where the PA heat sink is located. The heated air egresses from the rear of the FLEX-3000. If you have a lot of heat, it takes a large volume of air to effectively remove heat generated. To get a large volume of air, you need high velocity fans and the FLEX-3000 has them.

During receive, the fans are rotating at a slow speed, but once you start transmitting, the fans increase their rotational speed to the maximum and they do make a bit of noise when they do this. I was expecting the FLEX-3000 to be a bit noisier that the FLEX-5000, but they were loud enough to take me by surprise when they first kicked in.

Nevertheless, back to the important task at hand. I was transmitting with the FLEX-3000. Junior has a voice! The RF output was what I was expecting based on the AF signal strength input. A -5 dB input resulted in a 48 watt output. A 0 dB input (max) resulted in a 100 watt output. I ran several test and all look good, so it was time to connect the 20m dipole and let Junior use his new found voice and talk to the world and talk he did

Thursday, March 19, 2009

Junior learns to listen

..I heard static, but saw nothing on the Panadapter. Wait, is that the top of a signal there at the very bottom poking up? I went into PowerSDR and changed the lower limit on the Panadapter to -180 dB and there it is - the noise floor I was looking for. I know the sensitivity of the FLEX-3000 was going to be greater with the high gain preamp, but it was down @ -165 dB with the radio connected to a dummy load. That seems a bit low and I'll run through some of the calibration tests later this week to make sure. Turning off the preamp made the noise floor rise to a more reasonable level. Boy that was a momentary scare that I really was not expecting.

OK, so Junior has ears. Nice. I switched to the 20m dipole and the receiver came to life with signals. The receive audio is VERY good. I can't tell much difference between it and the FLEX-5000. I listened to a bunch of phone and CW QSOs on 20m. There are no artifacts I can hear in the audio, but I really have not listened that closely. Junior and I are still getting acquainted.



I am still getting used to the preamp. It has a lot of gain is isn't necessary to use it all the time. It is just another characteristic of the FLEX-3000 I am going to have to get to know better.

The receiver checks out fine. Now it is time to re-connect the dummy load and give the ol' transmitter a try.

It Arrived!!!

With much fanfare, the FLEX-3000 arrived today at 1:55 PM. I knew it was hear because my Shetland Sheepdogs tried to break down the door and bark him to death. But, Mr. UPS was unfazed by the EWAS (early warning alert system) and brought Junior home anyway. Below is Easu (1/2 of the EWAS) checking out the new arrival. It has to pass the "sniff" test.



I opened the box and gently lifted Junior out, removed the packing restraints and removed him from his clear plastic jacket. What a hansom fella he is. He is serial number B1009-0003

So, what is the first thing I did other than take some arrival pictures? Well, I didn't grow up with a screwdriver in my hand for nothing. I took him apart and looked inside.



A very nice piece of work, Ken and Gerald. My hats off to you both.

Cover removal was simple; three screws in the back, 8 screws on the bottom and you have to take the retaining nuts off of the Key and Phones 1/4" TRS connectors on the front and the cover slides off easily. Reassembly is just as easy too.

Now off to put it into it's operating position....

Today is the day!

12:05 PM

Mr. UPS says Junior is "in route". SWEET.

Time has come to prepare the temporary living arrangements for Junior. For the initial testing, I think I am going to pull the FLEX-5000 (aka “the Big Guy”) out of its operating position and take the opportunity to send it to the Mothership in Austin and let Ed work his magic on it in order to get the Powered Speaker Out jack repaired. In a moment of stupidity, I stressed the connector several weeks back and just could not stand to be without a FlexRadio SDR for a week or two. Junior (F3K) will be a more than adequate substitute in the Big Guy’s absence.

By taking the FLEX-5000 out, dropping in the FLEX-3000 in the same operating position, that will make getting on the air much faster. I had to make a little bit of room on the shelf because it is wider than the FLEX-5000, but that was an easy mod.

I will need to re-think the shack configuration for Junior’s semi-permanent operating position since he will be the portable SDR, unhooking him from the Matrix so he can take a walk on the wild side will be the primary design criteria. I have plenty of time to think this through before making a finial decision.

Wednesday, March 18, 2009

No more speed, I'm almost there!

Today @ 3:33 PM

Mr. UPS has been doing a great job keeping me informed about Junior's travel progress. Last night he went through Alabama and had a short stop in Georgia for breakfast. Then he headed north on I-85 to Charlotte. After dinner tonight he'll continue on to Greensboro followed by a short hop to Raleigh and then be taken by chauffeur in the Big Brown Truck to my QTH in Holly Springs sometime tomorrow.

Mrs. UPS was kind enough to call today the tell me Junior will need to be handed off to an adult upon his arrival or he will be taken back until I can make new arrangements to fetch him. Looks like tomorrow will be a good day to work from home, if ya know what I mean.

I am really getting excited now that we are both in the same state. I have been looking forward to this day ever since the first technical design meeting in December 07 when the core specifications were scribbled on the whiteboard in Austin and Gerald went into "engineer mode". It was a sight to see. I plan to get a good night's sleep. I was up for 25 hours yesterday working on a telecom SCP migration and only got 3 hours of sleep this morning. That "stuff" is for youngsters and I am getting too old to do it with any regularity. I may be real tired, but I'm giddy with anticipation.

Tuesday, March 17, 2009

A short stop in Mississippi

Today @ 6:37 PM

Mr. UPS tells me Junior (that is the new name for the FLEX-3000) made it all the way across Texas (after stopping off in Dallas) and arrived in Jackson Mississippi at 2:17 PM today. Keep on truckin’ little fella. I also received conformation that the microphone adapter cable was shipped separately. Oh well, that means the first QSOs will be digital.

Here is a picture of what the FLEX-3000 looks like. I'll be posting "real life" pictures after Junior arrives home.

And the trip from Austin begins...

Monday, March 16, 2009 @ 4:54 PM

I received an e-mail confirmation today that a FLEX-3000 has been shipped to me as part of the extended hardware testing team. This really is not a beta test, per se, but more of a shake down of the first units coming off the production line. I cannot wait to get the delivery date confirmation. I am assuming this weekend will be SDR centric and the XYL will be an “SDR widow”.

8:15 PM
I just received confirmation that the FLEX-3000 will arrive via UPS Thursday afternoon. I feel a case of Flexitis coming on for Friday