When in use, transistors do age. It's caused by the migration of the
majority carriers migrating across the junction causing the majority
carriers on one side to cause an increase of minority carriers on the
other side. This is directly related to temperature at the junction.
This was a big problem with early transistors, but the impurities in
today's transistors are measured in the parts per trillion, where they
were in parts per billion. Still, heat is the enemy of the transistor's
life and much sooner than with tubes
In the "old days" I ran CPUs at 80C to 90C with those little screaming
90 mm fans. Today, 40C is typical for quad and 8 core CPUs running
between 4 to 4.5 GHz, under load. I have two CPUs that will hit 5 GHz
when overclocked. Normal speed is 4.7 GHz, but these require liquid
cooling.
I have a couple older, SS rigs that the finals get too hot to touch.
Around 50C or 122F is where for most of us, things become too hot to touch
Yes, we can purchase parts looking toward the future, but we shouldn't
need to do that. It's highly unlikely any company is going to stockpile
transistors that are likely to go out of production. For high power
devices that amounts to a lot of money tied up with the likelihood of a
very slow payback. Add to that, most states tax inventory much like
property tax which is why you see so many "year end" sales. It's an
attempt to reduce inventory to a minimum. Here I believe it's 7.5%. At
that rate the cost of that inventory doubles around every 10 years. It's
not a good incentive for keeping slow moving inventory and would quickly
increase the selling cost well beyond what hams would be willing to pay.
73, Roger (K8RI)
On 6/5/2017 Monday 5:18 PM, Bill Ashby wrote:
Good Afternoon Roger!
With the Solid State Amps, which I'm Thinking of getting one, I too
believe that the parts May begin to be "More Available" as they "Age".
I would Hope that Someone would "Pick up" the "Slack" and "Stock pile"
the Drive Transistors so that When people Need to Replace the Drivers,
they would be Available.
On the other hand, I think it would be a Great Idea to Get a Few
"Driver Transistors" to Have, while they are still Available and Have
them "On Hand" for When the Original Set "Die".
I Also Agree that Tubes Are very "Forgiving" when they are "Over
Driven". Tubes Can and Do last a Lot Longer, just by the way they are
Made!
I currently have an Ameritron AL-80B, which I Absolutely Love! I'm
"Considering" a Solid State Amp, and May "Take the Plunge" one day
Soon, but Not to Replace the Amp that I currently have, but as an
Alternative Amp, for the Power Requirements needed to Drive the Solid
State Amps, being a bit Less than that of my Tube Amp! I may be
incorrect in my assumption that the Solid State Units being less of a
Draw regarding the Power Requirements, but I need to do some more
Research on the Differences between the Requirements of All of what's
available Currently!
The Tube units have a Solid History, whereas the Solid State units
haven't been in use for the same period as the Tubes, and may Never
get to that point!
I know that the Transistor "Should" last for around 100 years,
Theorists have said, but being as the Transistor has Not been in
Production for Really That Long, compared to the Tubes, we'll just
have to "Wait" and see!
At this point, I'd Trust a Tube rather than a Transistor, at least for
the time being, but I may find out that the Solid State units Might
just Surprise me!
Again , Thanks for your Input!
-73-
Bill
N6FFC
On Mon, Jun 5, 2017 at 13:48, Roger (K8RI) <k8ri@rogerhalstead.com
<mailto:k8ri@rogerhalstead.com>> wrote:
Add to that, the lack of knowledge in electronics, building, and the
general age of "old timers", module / pallet replacement seems
even more
likely. Most of today's rigs use an output module.
It's probably going to mean a lot more work (and money) for those
doing
repair work.
The old Collins and Hallicrafters rigs were relatively easy to repair
and align. Very few of us have the knowledge and fairly expensive
equipment to repair today's rigs so it's back to the factory, or
authorized repair center with shipping costs as much, or more than the
cost of repairing the old rigs.
As for device / pallet, replacement we are kinda getting back to
requirements similar to bipolar substitution, although I think gain of
the LDMOS will be less critical than with bipolars as long as it's
sufficient (and fits). The Yaesu Quadra is an example of an amp that
would be easy to upgrade and make much more rugged except for the
device. Examples do exist with the 150 W devices being replaced
with 300
W devices successfully, but the conversions were neither simple or
straight forward and required a much more rugged Power Supply,
making it
a major rebuild.
73, Roger (K8RI)
On 6/5/2017 Monday 8:28 AM, Joe Subich, W4TV wrote:
> On 6/5/2017 3:46 AM, Roger (K8RI) wrote:
>
>> The current crop of new devices is likely to be around much longer
>> than devices made just 10 years ago.
> Given the trend to pallet based amplifiers, I doubt the device shelf
> lifetime will be very long. Instead, rather than replace an LDMOS
> device on a pallet, the entire pallet (input transformer, bias,
> output transformer, spreader, etc.) will be replaced in the event
> of a device failure much like an entire tube is replaced in a tube
> amplifier.
>
> It may not matter what device (or devices) is used on a given pallet
> so long as the I/O and form factor are the same and the new pallet
> equals or exceeds the PO, compression point and IMD specs of the
old.
>
> 73,
>
> ... Joe, W4TV
>
>
> _______________________________________________
> Amps mailing list
> Amps@contesting.com <mailto:Amps@contesting.com>
> http://lists.contesting.com/mailman/listinfo/amps
>
---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus
_______________________________________________
Amps mailing list
Amps@contesting.com <mailto:Amps@contesting.com>
http://lists.contesting.com/mailman/listinfo/amps
--
God Bless...
Bill
N6FFC
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|