Zener diodes in TO220 or similar?

[Joerg]

Actually, a lot longer than that. They were still called that long
after they changed color and became twisted pair. ;-)

At a client in Arizona they announced the imminent arrival of the lunch
van this way" "Attention, roach coach will be here in five minutes".

 

[legg]

Those kinds of designs usually have only very tiny stuff on the circuit
board. TSSOP, MSOP, 0402 or smaller, SC-75, but tons of it. All the
heavy parts such as TO-220 are off the board because they must be bolted
to a chassis.

Actually, limitations for wave-side smd include both upper (body
thermoplastic thickness) and lower (lead spacing) limits.

I understood that bolting to220s to metal was anticipated here.

Ideally, the aim is to avoid process steps by selecting smd parts that
can all be included in the single wave operation, however in higher
density work, a 'component-side' smd reflow, or dropping of wave
altogether (double-sided reflow only) is finally enforced. TSSOP and
0402 are impractical to adhere and wave - are usually restricted to
reflow only.

You claim a real estate problem in higher density mixed circuitry with
power components. Just pointing out options here.

RL

RL

 

[John Larkin]

Phenolic terminal strips and 12AT7s.

Cheers

Phil Hobbs

I just got a couple of bags of phenolic terminal strips, the kind that
old TV sets used. They are great for breadboarding. They are getting
hard to find these days.

John

 

[John Larkin]

Actually, machining a semiconductor package down as you suggest is a
FAR more risky move reliability-wise.

This

http://www.highlandtechnology.com/DSS/J720DS.html

starts with a laser in a big sugar-cube ST-connectorized housing. Then
we turn it down...

John

 

[Jim Thompson]

At a client in Arizona they announced the imminent arrival of the lunch
van this way" "Attention, roach coach will be here in five minutes".

In Arizona, that's standard nomenclature for such vans ;-)

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

I love to cook with wine Sometimes I even put it in the food

 

[Joerg]

I just got a couple of bags of phenolic terminal strips, the kind that
old TV sets used. They are great for breadboarding. They are getting
hard to find these days.

For sauerkraut-wiring? Cool! I am down to about 2-3. You can be lucky if
you find NOS of this stuff and the silver on the dual solder-lugs isn't
completely black.

 

[Joerg]

Actually, limitations for wave-side smd include both upper (body
thermoplastic thickness) and lower (lead spacing) limits.

I understood that bolting to220s to metal was anticipated here.

Ideally, the aim is to avoid process steps by selecting smd parts that
can all be included in the single wave operation, however in higher
density work, a 'component-side' smd reflow, or dropping of wave
altogether (double-sided reflow only) is finally enforced. TSSOP and
0402 are impractical to adhere and wave - are usually restricted to
reflow only.

I can't remember any of my designs ever being wave-soldered. I usually
discourage clients and recommend reflow. We did that even in the 80's
although back then most others thought it was voodoo. When setting
temperature profiles you were pretty much on your own. Well, except
maybe for Jose Cuervo

You claim a real estate problem in higher density mixed circuitry with
power components. Just pointing out options here.

Thanks. Yes, that's my problem, real estate. So it'll all be the most
tiny parts I can find. And my eyes ain't getting better with age :-(

 

[Tim Williams]

For sauerkraut-wiring? Cool! I am down to about 2-3. You can be lucky if
you find NOS of this stuff and the silver on the dual solder-lugs isn't
completely black.

I have a baggie of nice and small ones, like 0.25" pitch or something like
that. They're great for high density stuff.

http://webpages.charter.net/dawill/tmoranwms/Elec_Compound9.jpg

Not the greatest up-close, but you can get some idea from the resistors
going into them. Made this class D tube amplifier a snap, plenty of pins
available.

Tim

 

[Joerg]

I have a baggie of nice and small ones, like 0.25" pitch or something like
that. They're great for high density stuff.

http://webpages.charter.net/dawill/tmoranwms/Elec_Compound9.jpg

Not the greatest up-close, but you can get some idea from the resistors
going into them. Made this class D tube amplifier a snap, plenty of pins
available.

Mine are flat, basically >1" wide strips of phenolic with lugs on there.
The advantage is that you get two lugs per contact so you don't have to
cram wires so much. The disadvantage is that you can't have multi-watt
resistors and stuff come too close or there will be a "burnt toast whiff".

 

[Fred Bartoli]

Joerg a écrit :

Mine are flat, ...

Really?

 

[Joerg]

Joerg a écrit :

Really?

)

 

[Rich Grise]

It could but in this app that would likely be shot down as a
non-approvable kludge ;-)

Actually, machining a semiconductor package down as you suggest is a
FAR more risky move reliability-wise.[/QUOTE]

Who said anything about machining? I just meant solder the thing
directly between the C-B leads, as if they were huge SMT pads.

He is talking about a TO-220, isn't he?

Thanks,
Rich

 

[legg]

I can't remember any of my designs ever being wave-soldered. I usually
discourage clients and recommend reflow. We did that even in the 80's
although back then most others thought it was voodoo. When setting
temperature profiles you were pretty much on your own. Well, except
maybe for Jose Cuervo

My experience, devolving from power circuits, was the reverse. The
adoption of smd was gradual, with wave-side being the first option as
it required only (!) careful design and the purchase of placement and
adhesive dispensing hardware. Thermal profiles are actually more
easily controled in a wave situation. Reflow was used eventually for
higher-density control and interface circuits.

RL

 

[legg]

Big zeners are rarely used these days, at least by me. Need to dump
several watts this time. Other than the ghastly DO5 "long bolt" style,
are there any in modern heatsink-able packages such as TO220?

There's no reason why the power need be dissipated in the
semiconductor. A small signal zener/active reference and less beefy
semiconductor can dump the energy into passive components, over a
certain range, if the response requirement isn't too critical.
Capacitors can assist the latter limitation.

RL

 

[Joerg]

My experience, devolving from power circuits, was the reverse. The
adoption of smd was gradual, with wave-side being the first option as
it required only (!) careful design and the purchase of placement and
adhesive dispensing hardware. Thermal profiles are actually more
easily controled in a wave situation. Reflow was used eventually for
higher-density control and interface circuits.

Back then I worked for a US company but in Europe. Over there the
transition to SMT happened earlier and faster. Basically my very first
design on the job that I started in 1986 was a huge board, nearly 100%
SMT. The only non-SMT parts were a few electrolytics because I had no
faith in tantalums. Turns out that tantalum paranoia had a very good
reason and was a good thing ...

We had to purchase our own reflow equipment because providers were just
coming on line and they had some difficulties tackling our huge boards.
I was anticipating lots of problems but to my surprise our production
manager who had no prior SMT experience got all this licked without a hitch.

There were some power circuits on there but probably not the level you
are dealing with. Things like a few dozen pulsers with a couple of watts
each. End of 1989 the German engineering was closed and self-employed I
did design numerous power modules, hundreds of watts. Those were also
all reflow soldered. Actually the first SMT wave process I ever
encountered was in the US, never saw one over there.

 

[Joerg]

There's no reason why the power need be dissipated in the
semiconductor. A small signal zener/active reference and less beefy
semiconductor can dump the energy into passive components, over a
certain range, if the response requirement isn't too critical.
Capacitors can assist the latter limitation.

Yeah, looks like I am dumping it into a FET. Looks good on the sim so
far. More parts, but works.

 

[Joerg]

So I guess you've never seen dip soldering?

Not in a developed country

Wave soldering, yes, I've seen that but rarely for SMT although I am
sure it must have existed for that even in Europe. Reflow is so much
nicer and cleaner. Boards get loaded, big mushroom button is pressed, a
gentle hiss and they all trudge through an incubator-like glass tunnel.

 

[John Larkin]

A quick dip in dilute HCl shines them right up.

I have a baggie of nice and small ones, like 0.25" pitch or something like
that. They're great for high density stuff.

http://webpages.charter.net/dawill/tmoranwms/Elec_Compound9.jpg

Yeah, mine are like that. They are great for, say, the power
connections into a hacked-copperclad breadboard. Solder the ground
tabs directly to the copper.

John

 

[John Larkin]

Joerg a écrit :

Really?

Old Tektronix ceramic strips are cool too. I scored a bunch at a flea
market. Here's some of both:

ftp://66.117.156.8/99A260A3.JPG


John

 

[krw]

Not in a developed country

We have a selective solder machine that gets used for PTH parts. It's
essentially a solder fountain on a X-Y stage.

Wave soldering, yes, I've seen that but rarely for SMT although I am
sure it must have existed for that even in Europe. Reflow is so much
nicer and cleaner. Boards get loaded, big mushroom button is pressed, a
gentle hiss and they all trudge through an incubator-like glass tunnel.

A lot easier before RoHS.