Roger,
> Are you any where near Santiago?
No. I'm 750km away from Santiago. But I have relatives in Santiago, so
if necessary I can ask them to get me something from there. The problem
is that none of them have the faintest idea about technical things, so I
need to describe EXACTLY what I want!
> Are there import duties on things like Artic Silver?
Very low cost items, say below 50 dollars or so, usually get through
customs without paying duties. Everything above that pays import duties
and sales tax, amounting to about 27% over the total CIF. We do have
free commerce treaties with many countries, including the US, but the
paperwork required to get some import handled under these treaties is
usually too complicate for small items. Among other papers, they ask or
a signed certificate by the seller, printed on a specific form, stating
that indeed the item purchased was made in the same country from which
it is sent to Chile. I don't want to bother salespeople with such stuff,
and so usually I end up paying the full duties. And anyway, the free
commerce treaty only waives the 6% import duties, and NOT the 19% sales
tax nor the other minor charges!
Yesterday I picked up the parcel from Digikey, at the local post office.
The order was $219 in merchandise, plus $46 shipping, and I had to pay
$55 in duties and taxes at the post office. That's how each dollar on a
price tag magically turns into 1.46 dollars! :-(
And of course the credit card company also pockets a few percent
commission on every international purchase, making each dollar go up to
a nice round 1.50 dollars.
But I won't complain. In some other countries it's much worse!
> Can you purchase from New Egg
Apparently not. I just checked, and they don't seem to offer postal
shipping. Their only international shipping seems to be Fedex, and I
can't use that because it doesn't deliver to my place.
> or some of the other computer parts suppliers on the net?
I can probably find some, but I have to check them one by one. I can
only purchase from suppliers who can take international credit cards or
Paypal, and who are willing to ship by mail (Priority Mail preferred,
but First Class Mail and Express Mail are fine too) to a P.O.Box in
Chile. Some companies make absolutely no fuss with that, while others
just won't do it, and then some are willing to do it but only after
checking the credit card info, which is fine with me.
> At roughly 250C the doping starts to migrate. it normally doesn't
> cause an immediate failure, but starts the process of gradual
> degradation.
Then I should be safe with my oven. I don't ever go much over 200 C.
> *Most* modern devices should be able to tolerate 200 to 250C for a
> short time with out causing a failure. I definitely would not want to
> operate one at those temps.
I agree. For long life electronics, I prefer to keep my silicon below
100 C! For intermittent use, I set 150 C as absolute limit, but aim at
130 C or so. Beyond that is the range where trouble must be expected.
But a few minutes in an oven seem acceptable at 200 C and a bit higher,
and most modern parts are actually rated for it!
> The resolder technique is a good one although it's best done with the
> quick heat technique.
Which I don't have! Except for small SMDs, which I can solder quite
quickly with a hot air pencil, but there is NO control over that part's
temperature that way! I could easily be heating the body of a small IC
to more than 300 C before its pins and the board melt the solder! So I'm
very careful when doing that, trying to heat just the pins and the pads,
not the body, to the extent possible. Oven soldering seems safer in any
case.
A soldering iron is indeed much hotter than th emelting point of solder,
but by heating the pins one by one, the silicon stays much cooler.
> They also use a high intensity light...I don't remember the name but
> I'd been guns using these tubes used to solder copper water pipes in
> just a second or two.
Probably that's beyond the reach of an electronic DIYer. Even oven
reflow soldering is considered esoteric by most! ;-)
> If you are interested in a go/no go IE flow/no flow then the
> hysteresis should not be a problem as long as it kicks out soon
> enough.
That's my idea indeed. The basic switch kicks in at a flow of 3 liters
per minute, and there is another one that is more sensitive. One of
those should be fine, likely the 3 l/min one. The problem is that they
kick out at a flow level FAR below the kick-in level. If the water flow
for any reason gets small, but not zero, these flow switches would NOT
provide secure protection. I don't think it's very likely for the flow
to get small but nonzero, so the switches might be OK. But if I find
some that have less hysteresis, I would prefer them.
But then, I'm planning to use MOSFETs that cost 30 dollars for the whole
set, so even if I blow them up from lack of water flow, it wouldn't be
the end of the world! I might even provide no protection at all, and see
how well that works in practice! Not a good idea, sure, but we are
entitled to our bad ideas, right? ;-)
> Currently I might run as long as 10 minutes on a data transmission,
That's almost a bulletin!
> or 5 minutes on slow scan.
That I have done too. Another thing I used to do a lot was packet radio
at full power with a duty cycle of 70 to 80%, for hours. That was when I
ran a BBS, and had long automatic traffic forwarding sessions with other
BBSs. That was without an amp, though. Just my TS-450S, which is still
going strong. Only the power supplies kept blowing up, until I designed
and built my big switching power supply, which found its way into QST
and the ARRL's Handbook. That one was a definitive solution. It never
blew up, and is still going strong too, running 24/7 powering several
radios and accesories.
OK, I will switch off the brag mode now.
> Yes, I like the idea of being able to run the legal limit class A
> meaning at 1500 hout at say 30% efficiency I need to run between 5000
> and 6000 watts input. at 5000 in - 1500 out that is 3500 watts of
> plate dissipation figuring 30% efficiency.
I admire your commitment to signal quality. I suppose in the USA that is
warranted. Unfortunately down here in Chile it isn't. The bands are so
full of totally overdriven, absurdly distorted SSB signals, with
splatter covering hundreds of kHz, that there is just no point in
producing a particularly clean signal oneself. So I'm looking to produce
a signal quality that fullfills the regulatory requirements, hopefully
with some room to spare, but there my commitment ends.
If I wanted to produce ultra high signal quality, instead of the
brute-force method of running class A, I would try to achieve it using
class B or even a higher class, but combined either with strong negative
feedback, or with active envelope control. It's more difficult to do,
more challenging, but should lead to a less expensive, much more
efficient, and simply better amplifier!
> dissipation although in SSB at a 20% duty cycle it is considerably
> less. At 5KW PEP in that is only 1000W of dissipation.
You are forgetting that a class A amplifier has constant power
consumption. If you make a 1.5kW class A amp that has 30% efficiency,
leading to 5kW input power, then the dissipation is 3.5kW when producing
peak output, and is 5kW when producing no output! In SSB at 20%
modulation duty cycle, it would be producing 4.7kW of heat!
> Class A may only be 25% efficient which would be 6KW in, but at a 20%
> duty cycle that is only 1200 W dissipation,
Nope! At least as long as you have the PTT pressed, the 25% efficient
class A amp will be drawing 6kW DC input power! You can play with duty
cycle only regarding TX/RX times. In class B instead you have the
benefit of reduced input power at every instant you are not producing
full output power.
So, a 60% efficient 1.5kW class B amplifier, at 1.5kW will be
dissipating 1kW; at 500W output it will be dissipating 943W (note that
the efficiency is down to 35%), and at 100W output it will be
dissipating 545W (a meager 15.5% efficiency), without considering the
power dissipated by the idling current if it is a class AB amp, like
most are. So, during SSB operation, the dissipation will average maybe
700W. The 25% efficient class A amplifier instead will have an average
dissipation of 5700W under the same operating conditions! It's a tall
difference.
And a class E amplifier with EER would be maybe 80% efficient at full
power, and would maintain this efficiency almost intact at lower
outputs! That would lead to only about 75W dissipation in SSB operation
at 1.5kW PEP output! That's why I would so much like to make an EER
class E amplifier. It would need no big heatsink, no water cooling,
would me small, lightweight, and so on. But I haven't been able so far
to solve the problem with the device capacitances. And apparently nobody
has. All legal limit class EER amplifiers I have heard of, have some
issues with this.
> Not at all. The FU728F will run just as efficient at 1500 out as it
> will running full tilt 2500 out. it just needs to be tuned up for
> that load impedance.
OK, yes, you are mostly right in that, as long as the tank components
allow to tune it up for lower power! Specially on the highest and lowest
bands, some might not allow that. And don't forget the power used by
filaments, blower, bleeder resistors, power supply core loss, and so on,
which will all be larger for a 2.5kW amp than for a 1.5kW one.
> Although still pricey you can use Chinese FU728F sockets. You do have
> to turn down the center locating pin by about 0.003" to get it to
> fit or carefully ream the hole in the 4CX1500B
If I ever get to use those 4CX1500B's, it will be either with homemade
sockets, or with sockets I can get for free or very cheaply, from junked
equipment. I just don't see myself paying the kind of money typically
charged for such tube sockets! I don't know what those Chinese sockets
cost, but I have seen some prices of US-made sockets, and found them
absolutely shocking and unexplainable!
> I'll be glad to trade you a couple 1 oz tubes of Artic Silver for
> those two 4CX1500Bs as they are my standby tubes for the FU728Fs.
> <:-))
Thanks, but I can't accept that trade, since I'm not positively sure
that my 4CX1500B's are actually good! ;-)
> I had a solid state amp on HF and it was a great amp, but the
> equipment here stays hooked up 24 X 7 and it's rare to unhook it even
> during our many thunderstorms.
I have all my HF antennas fed through a single feedline. When a
thunderstorm comes up, or when I leave the house for more than a few
hours, I unplug that single coax cable and throw it several meters away
from the house. Let those lightning bolts come! Actually I'm thinking
about automating that procedure.
And my VHF and UHF antennas are in the attic. The roof has a lightning
arrestor system on top.
> It just wasn't worth the risk considering my main tower has been hit
> at least 17 times. (with no damage)
Wow! My mast hasn't been hit at all so far. The tall trees around it
seem to be a good protection. Several of them have cracks, and a bit up
the hill one tree caught fire when hit. At least that's what the
neighbor says. It happened before I moved here. The tree is dead,
broken, and indeed looks charred at the surface, but not in the cracks.
> Here's a shot of the top of the main tower
> http://www.rogerhalstead.com/ham_files/Tower29.htm
Looks good! Mine isn't that impressive.
> We don't follow that technique. We start with a blob of the transfer
> compound in the *center* of the heat sink. With spring tension
> applied the heat sink is rotated back and fourth "with in tolerances"
> until the compound comes out the edges.
I use a similar technique: I apply a single nice strip of compound on
the mounting surface of a transistor, and if it has holes, I add a ring
of compound around each hole. Then I put it on the heatsink and apply
strong pressure. That makes the compound flow out evenly and filling ALL
gaps. If the compound is soft enough, I find it unnecessary to rotate or
move the part a lot, to get the excess out.
I definitely don't believe in any method that suggest to apply "a thin,
even layer" of thermal compound. Such a layer is always irregular, and
the surfaces are irregular too. When putting a part with such a thin
layer of compound on a heat sink, inevitable air will be trapped, and it
will be very hard to get it out! The thinner the layer was applied, the
less compound is there to move out and carry any air bubbles with it!
Manfred.
========================
Visit my hobby homepage!
http://ludens.cl
========================
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|