Diodes and capacitors

"High confusion" isn't a desirable condition. But one common feature in my design process is working up a particular design to the point where it migh t work, but starts looking too messy or too complicated, and moving on to r ather different approach. There are usually a number of different ways of g etting something to work, and the first thought that comes to mind isn't al ways the best solution.

--
Bill Sloman, Sydney

:

re

n.

ue be

y what

t. It

n

the

.

ley

e

unch

??If we

ause

/lens

ng

ly

e

a new

lly

w the

other

all

ng

a

to

n.

The mark of a good troubles shooter is comments along the line of "that wou ld explain most of what we are seeing, but not this or that".

--
Bill Sloman, Sydney

Some designs stalled there:

Best regards, Piotr

te:

are

e

ion.

alue be

e.

by what

n't. It

ven

s

s?

l

d the

ad.

ooley

ude

launch

??If we

de

ecause

ng/lens

king

ally

ome

th a new

eally

How the

me other

in all

ot

t

hing

n

g a

s to

e w

ion.

ould explain most of what we are seeing, but not this or that".

Does google groups have a plonk function? George H.

Is that a ROM? Too many wires to be a ram.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

Nice, should be able to store a telephone number in there! Or 1/10,000th of an mp3 file

Guess so. There are no diodes either, so it can't be a magnetic logic module. Seems to be a transformer-based ROM with a creative wiring. But I don't understand the purpose of the doubled cores, the cross section shouldn't be a problem in this application.

Best regards, Piotr

This is small and compact enough to be a CPU sequencer ROM. You don't need high capacity there, but extreme parallelism and speed are of crucial importance. The transformer-based approach shines there and is extremely robust. But I can't imagine the production specification of such a beast. Most likely hand-made.

Weaving that for eternity could be a good punishment in the EE hell, if there is one.

Best regards, Piotr

One more remark: if that's a ROM, then the capacity can be surprisingly high, much higher than your estimate. In the case of ROMs the information is in the wiring, not in the cores. The number of cores typically is equal to the number of the data bus width in bits.

Great technology for generating hundreds of jitter-free enables with a single current pulse.

Best regards, Piotr

+2 Keeps HF coupling intact on unearthed radio receivers even during zero crossings when diodes become non-conducting.

piglet

Here's a 304 bit core memory:

and a 4K bit one

I designed a CPU once, from TTL gates - it had three instructions - and designed a magnetic-core ROM to store the program. This was a shipboard bell logger. Just before we did the magrom, a new fusible-link bipolar rom came out, so we used that instead. Pity... the magnetic thing would have been fun.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

On Jul 24, 2018, Piotr Wyderski wrote (in article ):

It seems to be Soviet..

One big advantage of magnetic memory is extreme radiation hardness.

Joe Gwinn

They gotta keep a log every time you ring the bell? Man they don't let nobody have any fun.

When the throttle is moved in the bridge, a bell rings in the engine room until the engine room crew aligns their machinery to the order. They don't stand around looking at the engine order telegraph, or they may be taking a nap.

The bell logger records any change of throttle position, and the response, and the actual RPMs, and maybe some other stuff like compass heading. It's a legal requirement. The logs used to be kept by hand, and we were sort of pioneers in making automated bell loggers.

The bell log is the maritime version of a black-box crash recorder.

More modern ships let the bridge crew directly control the machinery, but it's still logged.

I enjoyed working on ships. It's a whole 'nother world with its own language.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Thanks for the replies.

It looks like the consensus is that it's either for RF reasons or diode switching noise reduction.

Would there be any reason to put a capacitor across a schottky diode? Other than for RF bypassing reasons?

Maybe to stabilize the nonlinear capacitance, by swamping it?

Schottky capacitance can be so large, it ends up looking very much like reverse recovery, in circuit.

Tim

-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Design Website: