H
Henry Kolesnik
Guest
I like good test equipment and when I started in the 1950s good was
homebrew, war surplus and the local radio TV shop. Then in the RCAF it was
a Tek 545 and later at university it was a Tek 545B. (I worked my way thru
college as a TV repairman working 20 to 30 hrs per week!) After university
and when I had spare cash I graduated from homebrew and surplus to kits.
Then my first good bad scope was an HP 120. Nothing but problems with the
HV supply. Now unwllingly retired at 63 one of my passions is to build a up
a small lab for personal enjoyment by finding at good prices test equipment
made by the best mfgs such as Tektronix , GR, HP, Fluke, Boonton, and
others, but the others usually don't hold a candle to the top 4. Others
include Wavetek, Racal, Philips, Systron Donner, etc. There's a third class
that includes Heathkit, Eico, RCA, Sencore, Hickok, B&K and others who
furnish radio
and TV test bench eqpt and I have some that is very, very good. A fourth
class is foreign like Marconi, Meguro, Siemens, Rohde & Schwarz, all
excellent but that I generally shy away from them because information and
parts are in the unobtanium category.
Generally finding a very good deal at a swapmeet means there's something
wrong with it, either unknown or undisclosed by the previous owner.
Finding docs for the best is hard enough but an order of magnitude harder
for the others and that goes another order of mag for parts. Sometimes one
needs 2 or 3 junkers to succeed or give up.
I don't like to leave a swapmeet without some good deal that I can take
home and plug into the Variac to see what it does. Generally I'll look
inside to see if it's complete with no obvious missing parts and no char
babies. Then I turn the Variac up and watch the current and look and smell
to see if all the smoke the the factory sealed-in is still there. If it is
but it doesn't work I check power supply voltages. It never ceases to amaze
me that very few of the mfgs have figured out how to build a reliable power
supply. In my work, Telecom, most critical hicap stuff had dual supplies
so it could still operate if one supply failed and to top that off they
migrated to hot swapable supplies! (If you don't have good tires, carry 2
spares!)
My own theory on power supply
unreliability is that the companies assign newly graduated engineers to
design power supplies and those guys don't have enough experience to
understand the ravages of long term heat, continual cycling and the crazy
stuff that can come down the AC power line! One exception today may be PC
power supplies as I don't see boatloads at the meets like I used to but
maybe it's because I'm not looking!
First, I check the supply voltages and if they're down I check the
rectifiers and capacitors. Many times I'm lucky and find the problem right
away but other times if its a short it's an arduos process on a pcb. After
much
wasted time the bad cap is found, an Al electrolytic if there are no or few
tantalums but if there's lot of tants there's a very good chance it's one of
the tantalums with a dead short and not exploded. Partial shorts for high
I^2*R are best
because they smoke or end up as char babies!
It would be nice to find a set of bath tub curves on some of the old
tantalums. I don't have a clue as to how many mfgs there were and how many
times
they improved their designs as they experienced failures. I guess what we
need to know is which one fail the most. I salvage many parts and haven't
had any real bad experience re-using previously well-used tantanlums. Is an
good old time tested tantalum better than a new one?
If the fault isn't in the power supply or on a supply rail it soon
becomes a waste of time without a schematic.
To end this mild diatribe, obsolescence of test eqpt is a matter of
economics for business and sadly me! I still like my Boonton 260 Q meter but
I'd rather have an HP 4342A and I'd rather have an HP 8921A to replace my HP
141T, a newer Tektronix TDS to replace my Tek 465. Space, time, dollars,
and performance all suffer in my lab but it's there for my enjoyment and
entertainment, and most likely to be upgraded with smaller, higher
performing, newer, less obsolete gadgets. I perceive that eqpt with surface
mount, PICs and ASICs is generally impossible to repaiur at the componenet
level and board swap is an alternative if you can find spares but then there
are so many suffix updates!
73
hank wd5jfr
homebrew, war surplus and the local radio TV shop. Then in the RCAF it was
a Tek 545 and later at university it was a Tek 545B. (I worked my way thru
college as a TV repairman working 20 to 30 hrs per week!) After university
and when I had spare cash I graduated from homebrew and surplus to kits.
Then my first good bad scope was an HP 120. Nothing but problems with the
HV supply. Now unwllingly retired at 63 one of my passions is to build a up
a small lab for personal enjoyment by finding at good prices test equipment
made by the best mfgs such as Tektronix , GR, HP, Fluke, Boonton, and
others, but the others usually don't hold a candle to the top 4. Others
include Wavetek, Racal, Philips, Systron Donner, etc. There's a third class
that includes Heathkit, Eico, RCA, Sencore, Hickok, B&K and others who
furnish radio
and TV test bench eqpt and I have some that is very, very good. A fourth
class is foreign like Marconi, Meguro, Siemens, Rohde & Schwarz, all
excellent but that I generally shy away from them because information and
parts are in the unobtanium category.
Generally finding a very good deal at a swapmeet means there's something
wrong with it, either unknown or undisclosed by the previous owner.
Finding docs for the best is hard enough but an order of magnitude harder
for the others and that goes another order of mag for parts. Sometimes one
needs 2 or 3 junkers to succeed or give up.
I don't like to leave a swapmeet without some good deal that I can take
home and plug into the Variac to see what it does. Generally I'll look
inside to see if it's complete with no obvious missing parts and no char
babies. Then I turn the Variac up and watch the current and look and smell
to see if all the smoke the the factory sealed-in is still there. If it is
but it doesn't work I check power supply voltages. It never ceases to amaze
me that very few of the mfgs have figured out how to build a reliable power
supply. In my work, Telecom, most critical hicap stuff had dual supplies
so it could still operate if one supply failed and to top that off they
migrated to hot swapable supplies! (If you don't have good tires, carry 2
spares!)
My own theory on power supply
unreliability is that the companies assign newly graduated engineers to
design power supplies and those guys don't have enough experience to
understand the ravages of long term heat, continual cycling and the crazy
stuff that can come down the AC power line! One exception today may be PC
power supplies as I don't see boatloads at the meets like I used to but
maybe it's because I'm not looking!
First, I check the supply voltages and if they're down I check the
rectifiers and capacitors. Many times I'm lucky and find the problem right
away but other times if its a short it's an arduos process on a pcb. After
much
wasted time the bad cap is found, an Al electrolytic if there are no or few
tantalums but if there's lot of tants there's a very good chance it's one of
the tantalums with a dead short and not exploded. Partial shorts for high
I^2*R are best
because they smoke or end up as char babies!
It would be nice to find a set of bath tub curves on some of the old
tantalums. I don't have a clue as to how many mfgs there were and how many
times
they improved their designs as they experienced failures. I guess what we
need to know is which one fail the most. I salvage many parts and haven't
had any real bad experience re-using previously well-used tantanlums. Is an
good old time tested tantalum better than a new one?
If the fault isn't in the power supply or on a supply rail it soon
becomes a waste of time without a schematic.
To end this mild diatribe, obsolescence of test eqpt is a matter of
economics for business and sadly me! I still like my Boonton 260 Q meter but
I'd rather have an HP 4342A and I'd rather have an HP 8921A to replace my HP
141T, a newer Tektronix TDS to replace my Tek 465. Space, time, dollars,
and performance all suffer in my lab but it's there for my enjoyment and
entertainment, and most likely to be upgraded with smaller, higher
performing, newer, less obsolete gadgets. I perceive that eqpt with surface
mount, PICs and ASICs is generally impossible to repaiur at the componenet
level and board swap is an alternative if you can find spares but then there
are so many suffix updates!
73
hank wd5jfr