0x1

So, after surfing the internet for awhile, I found someone was already doing what I had planned!

Bill N4QA is switching in resistance to the zener diode on Dave Benson K1SWL’s Rock Mite QRP CW transceiver.

Needless to say, I just ordered one up for 7040 kHz. I know I wanted to do less kits and more straight homebrew, but this is too easy to let slip by!

I’m not sure I understand the theory behind adding the variable resistor to the zener yet. The zener is giving a regulated voltage to the oscillator, so adding resistance will change that voltage, which would change the frequency, I assume. I need to look at the Rock Mite schematic some to see if I can make sense of it. It seems a lot more simple than switching in a varactor.

So I’ve been giving some thought to an alternative to building KD1JV’s MMR-40 SSB transceiver, mostly because this might be a little difficult for a first transceiver project. Since my CW skills are pretty much useless, I really need another solution. (Well, I could just do computer CW, but that has it’s challenges.)

I figure that if I have a VXO controlled QRP CW rig, it should be easy enough to switch in a varactor or other capacitance to create the 170 Hz shift needed to create a RTTY “space”, the unshifted carrier frequency being the “mark”. I haven’t quite figured out how to do this yet, but many of these simple CW rigs have a provision for RIT, so a similar design may work.

In addition, I should resurrect my Arduino RTTY modem project. I doubt I’d have demodulation working in time for HAR2009, but at least I could have an encoder, and do demodulation with fldigi.

UPDATE: I’m actually wondering if I can use Hans Summer’s 40m QRSS beacon and then drive Ashhar Farhan’s JBOT linear ampilifier (that I was discussing with Mark K6HX). Hans uses a backwards LED as a varactor, pretty cool.

This method would still need a separate receiver. Maybe that is a good thing?

Well, now that I’ve got all the trails from Clark Reservation State Park on The Map, it’s time to tackle another park—Green Lakes.

So far, just a start. I’ve added a couple trails and the two lakes. The roads need to be corrected and there are a ton of trails that need to be mapped.

View Larger Map

So, I decided to build me one of Onno PA2OHH’s simple LED frequency counters.

I can start building this as soon as today, thanks to Mouser!

Zach Smith over at NYC Resistor has tamed ATMega168’s testosterone-injecting big brother, the ATMega644P. Introducing the Sanguino! Compatible with the Arduino, but more flash, more RAM, more IO! w00t!

Apparently the most awsome energy beverage of all time, Club Mate, played a supporting role, look closely in Bre’s video:

Sanguino: Arduino’s Big Brother from Zach 'Iowa' Hoeken on Vimeo.

[Link]

We had a piece of client software which would open a persistent Oracle SQL*Net (Net8) connection to the server. However, due to the ways the users would work, often this connection would sit idle for very long times.

We began getting reports from our service desk that users at remote sites and users connecting over VPN would get a bizarre Oracle error and the application would crash. After much digging, we discovered the software would attempt to issue an Oracle command, and it seemed that the server would send a TCP RST in reply.

It turns out, however, that the server was not sending the RST–it was a PIX firewall sending this. The default behavior on Cisco PIX and ASA firewalls is to timeout an idle TCP connection after 1 hour. This essentially is a broken behavior, as the RFC 793 says nothing about the longevity of a TCP session or about keepalives. However, firewalls are a reality, so we need some way of getting around this.

Increasing the global parameter on the PIX or ASA firewall is an option, but on a very busy firewall, this could begin to use up resources, so I don’t recommend it. It might be an option to set this to eight hours on a fairly quiet firewall:

timeout conn 08:00:00

A more attractive option is to use TCP keepalives. Oracle by default does not send keepalives, and, in a Windows client environment, the OS only typically sends keepalives once every two hours. So, two changes are needed.

1. In tnsnames.ora add the atom (ENABLE=BROKEN) in the (DESCRIPTION) container. This will cause Oracle SQL*Net (or Net8) to ask the OS to send TCP keepalives.
2. In the Windows Registry, add a REG_DWORD named HKEY_LOCAL_MACHINE\System\CurrentControlSet \Services\Tcpip\Parameters\KeepAliveTime with a value of 1800000 decimal. This corresponds to 30 minutes.

I’m fairly certain a reboot is in order, as this is Windows.

When we applied this to all our mobile users, the problem went away completely.

So, I got bored during lunch and I discovered that my ghetto-ass Motorola C139 has a “My Tones” ringtone composer. The language is sorta bizarre, but I figured it out quite well. So, those of you with Motorola phones with this “feature”, here you go.

Note: I newlined between each measure… when keying this in, just run it all together.

*4g2*4g2*4g2*4e3.*4&b4
*4g2*4e3.*4&b4*4g1
*5d2*5d2*5d2*5&e3.*4&b4
*4#f2*4e3.*4&b4*4g1

Everything about Disco is pushing the boundaries of interface, usability, and utter functional simplicity. Well, once you realize that Disco is emitting real time 3D interactive smoke as you burn, we start redefining the boundaries. Want to push it out of the way? Blow into your microphone and the smoke will react accordingly. Or, go ahead and flick at it with your mouse. Remember, you'll need a supported computer to run this. (emphasis added)

UPDATE: This no longer works. No more cheese for you.

http://moon.google.com/

Zoom all the way in.

This morning I'm awoken by my NexTel ringing its voice mail waiting reminder. I check it and call back my boss. Apparently he heard there was a power issue at our Syracuse office and he wants me to make sure everything is okay with the network. So, I VPN in and check a couple hosts there. Nothing.

When I get to the building (after throwing on some clothes), I'm greeted by the smell of burnt somthing-chemical. It was a familiar smell, but I couldn't place whether it was batteries, capacitors or transformers. I knew it was one of those, though. The odor was coming from the UPS room. I headed upstairs and saw that everything was down, with the exception of the phone system and supporting hardware, which is in series with another UPS. (Stupid, but saved them this time.)

We find out that something had tripped the utility breaker, the ATS output breaker and the UPS input breaker. We pulled the emergency maintenance breakers on the UPS to isolate it from power and got our hardware up bypassing the UPS. The ATS flipped back to utility ten minutes later (nothing like watching 480 V 3 phase arc under load on a giant switch that throws itself.)

When the electrician got there, he opened up the battery cabinet. One of the cells had a three inch diameter hole in it. There was soot everywhere.

We're not sure if the battery explosion was cause or effect yet. APC is coming to take a look tomorrow.

So, I got home around eight this evening. I made some dinner and have just been chilling out with the laptop and music going. I gotta be in the office before eight tomorrow morning.