A shotgunning look into my random thought patterns, turning up surprisingly non-random, non-trivial connects on several hoopy wavelength. Your Mileage May Vary.
Ok, so things went well in the wee hours of the morning for the power supply. I'll let the videos speak for themselves:
I had to turn the camera off to capture the results because I couldn't find a safe way to hold my camera and take the measurement. Here's the after video:
The power supply is pretty much ready to run at this point. All that is left is packaging and hooking up the radio, There's just one last check to make, the one for ZAP!
The rest of the night was pretty interesting and I have some more testing to run before I'll be ready to post.
So Here's what I have been able to discover:
I'm still not 100% certain why the unit failed the first time. Initially I suspected the power supply, but I'm reinvestigating, I discovered another loose solder connection on one of the tube sockets. I'm not sure that's what did it, but I'm always suspicious of any changes I made. I changed tubes, the supply blew up. Thems the facts. Whether changing the tube caused a supply malfunction remains to be seen.
You can notice from the pic that I changed out the crystal holder part of the circuit. I did this to keep the metal case of the crystal WELL away from the supply voltage terminal, and because it looks better put together up and off of things as opposed to flopping around all dangly like. The bad news is my thumb is sore today because I think I poked it with a wire. If things don't improve soon, may have to go see the Dr about lockjaw! Heh!
how you 'spain that one to the XYL?
The second time the supply failed, I am 100% certain was because I used a severely underrated rectifier diode. I used a 1N4001 blindly following a misprint in the schematic I was using, instead of a 1N4004. So I'm beefing up to a 1N4005 just to make sure, and that's only IF I can't find any 1N4007's laying around the junque box. Next I will isolate the 150 B+ part of the circuit from the line, and instead of using one diode, I will go ahead and use four, building a full wave bridge rectifier. This is because I got diodes, and math is on my side...
From an awesome mess
Comes a beautiful circuit
Arise O Phoenix!
OK:
so you take a half wave rectifier like W3IRZ(sk) first designed into this rig, and you take the caps he had for smoothing, do some funny math, and using the max transformer rating of 1 amp. I'm not expecting to actually draw more than about .05 amp here, and yes, I know about derating the transformer current rating in a bridge rectifier, the actual rating is 1.2 Amp. Radio Shack sells a 3 Amp transformer too, it is part number 273-1511 you can ship it to my address from QRZ.com or hit the donate button in the top right hand corner of this webpage :)
anyway,
you do the math found at MIT: and discover that there is 24.1 Volts of ripple voltage when the load draws 1 Amp.
Now this is worst case scenario mind you, but still. at 50mA which is a lot closer to the actual current draw, it will be a lot less, like 1.205 volts ripple, which is less than 1% of peak voltage (150VDC). An acceptable number, but since the 470 microfarad caps needed to produce this performance are now BBQ in the trashcan, it's time for plan B.
I have some caps suitable for filtering, but not the exact values as before. I needed a way to lower the required amount of smoothing needed to get a usable DC voltage. I decided to build a bridge rectifier, and use a couple of caps left over from the earlier efforts. Using two 220 microfarad capacitors, I actually drop my ripple voltage a little bit, to like .95 volts at 50mA. I can handle that!
Now you can buy a bridge rectifier at the local shack, but this is an exercise in understanding, and I don't have any 1 Amp, 120Vrms AC bridge rectifiers in the Junque box. I do have 4 1N4005 diodes though, and knowledge, and a community of assorted skalawags, errrr, electrically inclined persons who can help me out.
This is the bridge, as I have constructed it. The alligator grippers are holding it at the position where I hook in the AC, one side to top, one side to bottom.
The bridge will be hooked the (originally primary) secondary of a 12.6VAC to 120VAC transformer.
Here's a closeup, showing the AC wires hooked into the circuit. I've intentionally left the 'ears' on the soldered circuit leads so I can do some testing later. You can see that the bands indicating the cathode of the diode are running from left to right. As it is layed out, the rectified positive voltage will be on the right, and the negative side is the left.
This is an overhead shot of the circuit so far, laid out on a potential housing mount. The pic is upside down from the way that I took it, but whatever.
you see that I have two identical transformers, they are Radio Shack "120-12.6VAC 1.2 Amp" available at most stores. I have rolled up and taped off the center tap of the transformer, because I don't plan on using it. The two outer leads have been tied together. I will take my 12 VAC filament voltage from this point. Hot and neutral wires will be hooked up to the primary transformer on the left, and the diode bridge is hooked up to the secondary transformer on the right. If all goes well, and I haven't done anything stupid, we will have pulsed DC ready for smoothing at the secondary end.
Here's a video showing how everything is currently laid out:
I won't be hooking AC up to it until I've got the unit planned out and tightened up a lil bit, namely I will be solving space issues, hooking up a switch and fuse set up, and hooking it up the way I will have it inside the box.
So what do you guys think?
Go or No Go?
Good news, I made my first QSO with the Twin tube 80!
Bad news, won't be making another one any time soon!
Yay smoke!
In Chronological order, the good news first:
I made my first QSO with this rig!
Yep, it's true, actually made my first two QSO's, one with W1PID in New Hampshire, and one with AK4O down the road from me in Tampa. I got so excited, that I forgot to video the QSO w/ W1PID, even though that was my intention. Here is a vid from right after the QSO,
I felt pretty good about it!
About three hours before 9pm (local time, 0100 UTC), I announced that I would be QRV at 0100 UTC on the QRP-L reflector, the NoGA yahoo group, and other places. About 20 minutes before starting, I went out to the shack, turned on the RX rig, and plugged in the TT80, I noticed a lil' spark right as I plugged it in, looking back I think this was the beginning of the end of the capacitors in this iteration of the power supply. At the time, I didn't think anything about it because hey, tubes were glowin', and there was no smoke!
After getting some of the kids in bed, and tucked in for the night, it was radio operating time. I started calling CQ, and it seemed like almost immediately, I got a return, notably, from W1PID, click his call sign and read his adventures, they are well worth it! He takes his radios all over the place :) Considering my antenna on 80m is a W3EDP with the counterpoise unhooked (is just over 1/4 wave on 80m), and worked against my station ground, I'm glad to have worked anyone, but feel especially good that my signal made it all the way to New Hampshire!
I then took to calling CQ again. Avoiding static crashes, and other QRN, and the ever present Digi QRM, it wasn't easy, especially using the Swan as a receiver. It says "CW" on the mode selector switch, but this is a lie :) it's really the USB SSB filter working, so you hear 2.7KHz of bandwidth all at once. I've had some success using an active audio filter, but I've got a lot of work to do in the shack before I get that particular aspect of the problem solved. I have seen people homebrew CW filters to install in the Swan 100MX, but I don't know if I will yet or not.
About 20 minutes after the first QSO, I copied and replied to AK4O who lives about 36 miles from me (as the signal flies), and also got a nice email from him. I called CQ some more after that, for about an hour, without any luck.
Then, I unplugged the TT80, and tuned around some with the Swan, on 80 and 40, and decided to start trying different tubes (same type) in the sockets. you know what they say about not messin' with a good thing?
And Then:
EGADS!
Whatever went on in there, wasn't good.
hmmm,
since the first thing I did was change out a tube, I checked the tube base, and I didn't see any obvious shorts, so I began to suspect the capacitor. I remembered the extra beefy spark when I first plugged in the power supply, and I knew I had a suspect!
I found pieces of the rectifier diode, and cleaned up the box and filament transformer with simple green. That's some awesome cleaner. Then I decided to function check the filament transformer. It passed. I tested the resistance across the power surge resistor between the two capacitors: it failed. No continuity!
I then rewired up a new power supply using my last 470 microfarad 250v rated electrolytic, and a new 220 microfarad 250v electrolytic. I went to a 50ohm 10watt power surge resistor (it was as close as Radio Shack could get). Now I am ready to test things out before hooking the TT80 back up. I'm just looking for 150V dc power. Whelp, as soon as I plugged it in the 470 microfarad blew up again.
dang.
I'm throwing this cord out, grabbing another, and trying to scrounge up a couple more caps.
Until then, this project is officially on hold!
I decided to open the power supply up, and clean up some circuitry. I wanted to make sure that things stayed put once I was ready to QRV. Here's how things looked before getting them all put together.
Innards of the power supply. You will see in the final pic that I replaced the B+ wire, the red and black wire in this picture
and here's the transmitter:
I secured the top to the the lid using some ultra heavy duty fastener called dual lock. I wanted a rock solid mount, and I think I got it!
I have two other plans for the transmitter that I haven't decided if I'm going to do yet or not. First, I'm thinking of replacing the clips for the crystal. I'm not really satisfied with how they connect but they are protected. Second, I'm thinking of breaking the rig down, and redoing the chassis all together. The lil' Altoids tin it's on now is showing its ten years of age.
In this pic you can see the transformer, two filter caps, ground connection, and B+ line out. Everything is ready to be buttoned up and fed AC! The caps are mounted to the aluminium plate using mounting tape, and actually rest upside down inside the Power supply. I actually bought a 50 Ohm 10 watt resistor to go between the caps, but decided that the 45 ohm 5 watt would work better because of size. Radio Shack doesn't carry a 45 Ohm power resistor in the store, they carry a 0.47 :/
if you look closely at the power cord, it has a piece of heavy gauge magnet wire wrapped around it, this is to prevent me from accidentally pulling the cord out of the power supply, placing stress on the solder connections, and generally increasing the potential for fail.
Now for a demonstration of the awesome power of hollow state:
My theory on increase power output is that the powerbuss solder connections are better now, and tubes are awesome.
seriously on the powerbuss connection though.
Originally, in the interest of using things over and making due, I had hooked up connections with some spare wire I had laying around that at one point in time was part of a microwave. I noticed when I was hooking things up that this wire was silverish and didn't take solder well. As I remember there was no solder for power buss lines in the microwave, it used crimped terminal connections. hmmmm, probably is made from something from the lowest bidder. Fortunately, I also had some real buss wire laying around, so I replace unknown material with copper.
More copper means less power wasted heating a wire.
Here's how everything looks set up and ready to play:
no QSO yet. Will announce on twitter, QRP-L, SKCC facebook group, and other places when I get QRV.
Now that I am reasonably sure of my polarity, and I'm done cussin' the feller who put this plug together, it's time to go over the transmitter itself, and see if there's any improvement to be made. Heh. My radio uses 2 6AQ5A's in parallel. The filament takes 6 volts, and are wired in series. Right off the bat I saw that I had I had the filaments wired in parallel. This was from a time when the only transformer I could find was a 120v AC to 6v AC. I certainly corrected that lil' problem. It also explained why the sucker glowed a lil' more than I expected the first time I lit it up. 12v on a 6v filament will do that. I'm glad I didn't blow the tubes! I also cleaned up some wire and leads underneath, and made sure my point to point all went to the right places. In the process, I discovered a cold solder joint, and found a better way to route some parts.
The Kind of Ugly that's a Pretty all of it's Own
It's not necessarily my best work, but hey, it will do!
And yep, that's three 33pf NP0 capacitors wired in parallel.
Time to do some testing.
first, go see Dave Benson's website, Small Wonder Labs, and order a rockmite :) also, check out his documentation section. Download a manual for the SW-40+ transceiver, and look in the testing section. In that section is a circuit you can make pretty easily, and use it to measure the power your radio is putting out. I built mine on some radio
Circuit, dummy load, and babyfood jar.
Prototyping board, specifically so that things would stay together. I connected the orange and yellow wire across the dummy load, and made sure that I hooked the yellow wire to the side that goes to the pin, and the orange wire to the side that goes to the ground side of the plug. At the measuring end of the plug are two posts made from surplus resistor leads, I hook my Digital VOM to them using probe hooks. The set up is hands free once you hook the probes onto the posts, and turn on the VOM.
Here's a little video about what I discovered when I measured the voltage across the dummy load:
Ok, so, according to theory, the voltage measured here is Peak-to-Peak voltage.
Knowing that, tell me how much power is getting to the dummy load.
Show your work!
Went to shack to take pics for post I'm working on, brought out radio. Took pics, did some prep work getting things ready for final packaging.
Turned on radio, tuned it up, tuned around...
Came time to choose, finish article, or play radio, it was time to play radio!
Worked F5IN at 05:40 UTC
Received a 569,
gave him a 579.
my power, Less than 100 watts.
Band was only meh tonight.
73!
I didn't want to button my power supply up before figuring out why the hot wire was white. Basically, I wanted to make sure this radio could work anywhere, and having a hot white wire was troubling. Like David suggested in his comment, I suspected the outlet, and then the wiring in the Mains box. Time to flip breakers and do some inspecting!
First, I flipped the breaker to the shack, and trundled back into it, hoping to find a black wire where a white wire should be. The Shack in the Back was built before we moved here, and is 'unfinished'. The guy who owned the place two owners ago built it with scraps left over from some of his contracting jobs I think. When I unscrewed the socket from the recepticle, everything looked, well, fine. I was impressed with the craftsmanship of the socket to tell you the truth, connections were tight, and in the right place. Time to check the mains box, but I knew it was going to look fine there too. It did.
Back to the thinking bucket.
Yon Thinking Bucket
I juiced the shack back up, grabbed my VOM, and began doing things my mother told me not to do, namely sticking metal things in electrical outlets. By metal things, I mean VOM probes. I always feel weird sticking things into electrical outlets, makes me make sure I do so with fear, and caution. That's not a bad thing, I suppose.
First, I checked the socket for power, Hot wire to neutral, and I got 120v AC. Good!
Then I checked for what should be the neutral wire to ground, and got no AC voltage, Good!
for the kicker, I checked what should be the Hot wire to ground, and got 120v AC, Good!
So the socket is good! Should have done that first instead of assuming it wasn't... hmmm.
Next test, the power supply strip I plug into the socket. Maybe there's something miswired there,
Just checked Hotside to ground, 120v AC. Good!
So the power supply is not the issue.
must be the power cord.
The Cord- Not Polarized
notice that in the picture of the plug end, the plug is not polarized. This plug and wire was salvaged from some piece of computer equipment, and at one point in time had a female chassis mount plug on it, the kind that looks like this:
Female Chassis Mount
On a polarized plug, the narrow slot/prong is hot, the wide slot/prong is neutral and the round prong is ground, my non polarized plug fits in the socket the way a polarized one would, but, was it wired like a polarized socket? Just where did the white wire go?
I made sure the plug was unplugged, clipped the cord out of the power supply, and I decided to idiot check the green wire first, green for ground right? Switching my VOM to measure resistance, I continuity checked the green wire to the round prong, and it was Good! Then I checked between the other two wires, and no other wire lead to the round prong, Green is good!
I then oriented the plug like I was about to plug it in.
Right is hot, left is Neutral...
As this picture shows, the prong on the right, goes to the hot side of the socket. That's the one I want to check.
First I check the black wire, no continuity.
Then I check the green wire, no continuity.
Finally, I confirm that the right prong, the prong that normally connects to the hot side of the plug is connected to the white wire.
The white wire is hot because it goes to the hot side of the socket.
Heh.
I have no idea why this cord is made this way.
I just know that I can safely button up the power supply knowing It's the Cord, not my shack!
This is part 1 of a series, a complete list articles in this series is available under "Ham Radio Master links" on the sidebar.
Recently I've been wanting to clear various projests out of my project queue. Some of them are things I've been working on, and needed specialized parts to complete it. Others were not working because of technical issues. Some are because I'm afraid of drilling holes into some things and potentially ruining them :-) This last week I've been working on a radio that has vexed me for for ten literal years. The radio is the NoGA Twin Tube 80 schematics and instructions available at that link. I built it as it appeared in the NoGA Compendium available here. The kit itself is no longer avaible, but the parts to build it are out there. Mike Branca W3IRZ (sk) helped me get the kit working when I was having issues with it at my house. My goal was to put it on the air for the NoGA CW net on Tuesday nites. Well, Mike gave me a lot of great advice about building radios on that trip to his shack, and at the end of the day I had a working radio. On the drive home, something affected the æther stored in the radio, and while after plugging the radio in, I got no smoke out, neither did I get any RF! Alas ! Every trick I tried played out, and I set the little tube transmitter aside for other projects. Somehow, I knew the problem was in the power supply. I suspected the large value capacitors in the rectifier circuit.
Fast forward to the present.
Every time I would look at this radio I would get sad. Sad for the loss of a friend (W3IRZ sk in 2003), and sad that I couldn't seem able to figure this thing out. I decided it was worth a try. Time to stop being sad, and get back to slingin' dits and servin' dahs.
Since I've always suspected the powersupply in this radio, I decided to just hammer away at the power supply until it was working. The way this particular rig works is that you have a 12 v ac line for the tube filaments, a 150 v dc line for the plate, and a ground that you ground through the plug. I got out my caps, diodes, resistor, plug, and began the project for the third time. After soldering everthing together, making sure to follow instructions, I took the leap and plugged it in. No smoke! I tested the 12v ac circuit. I got 13v, that's ok, the transformer output is actually rated at 12.6v ac, so no big surprise. The 150v dc line, no dice.
Hmmmmm, this is where I left off ten years ago. I went to bed dejected and tired. I'd stayed up until 3:00am for nothing. I hadn't even learned anything yet. Course all that was going to change the next night.
I do my best to junk out old stuff so I can have strange parts when I need them, you can't buy 220 microfarad 250volt rated capacitors at radio shack. You have to go online and look them up on Mouser or Digi-Key. Fortunately, one of my recent junking exercises produced the caps I needed. This is recycling! The next night I replaced the caps and diode (just in case) double checked my wiring, plugged it in, and nothing! Still no B+ voltage. Dejected I tuned around the band hoping to find something interesting. I began thinking of what the issue could be and applied a little problem solving.
Grabbing the VOM, I hooked the ground wire up to the black probe and began poking things with the red. Positive terminal of 470 micro farad cap, nothing, positive terminal at 220 microfarad cap, nothing. Bottom side of 1N4001 diode, nothing. Top side of 1N4001 diode, I switch to AC because I'm connected to the ground and the black (hot) wire directly, nothing. Wait, whut?
Nothing?
I have 13v ac at the transformer, between the white (neutral) and black (hot) wire I have 120v ac, but between green and black I got nothing?
Then it clicked. The black wire wasn't hot, the white wire was! Two snips and a dab or two of solder later, and I have the white wire hooked up as if it were hot, apply juice, check voltages, and we are good to go:
How's that for live action?
It's also why every shack needs a 'chicken stick' of some sort. Link takes you to a qrz.com site all about stories of chicken sticks.
So, I have power at my supply, but It's coming from a funny wire. Before buttoning everything up, I needed to figure out Why the White Wire is Hot.
That's Chapter 2!
Everybody that reads me has probably already found this guy,
but!
I need to tell you about M1KTA's QRP Ham Radio Blog because it is awesome.
Lots of projects
Great ones like his take on a KD1JV inspired PTO (Permeability Tuned Oscillator)
Great info on the W3EDP antenna (!)
All around, he seems like an ok fella!
I've got a project in the pipes I'm working on, and promise another update real soon.
73! (and 72)
