Other Designers' Designs

Gentlemen,

Has anyone ever been baffled by the intended purpose or function of a certain sub part of a schematic? Something that defies your attempts to discern why it was included, notwithstanding that it was used by a manufacturer of unquestionable repute? I get this quite frequently, but then I'm just a hobbyist who never made any formal study of electronics. But what about our pro designers here? Just curious....

CD

Reply to
Cursitor Doom
Loading thread data ...

A schematic is not intended to convey the information that you're looking for. You can only intuit.

RL

Reply to
legg

Sometimes manufacturers include circuitry that has no function, but which can still cause problems. Ubiquiti make a range of WiFi access points. One intended for educational use has some extra circuitry for simulating school bells and alarms. They left the components in place for the ordinary version (UAP-AC-PRO). Unfortunately, one of the ceramic capacitors must have been badly specified and many units failed because leakage in that capacitor caused the PoE negotiation thresholds to change. I have repaired one by just removing the capacitor. John

Reply to
John Walliker

Sure! It's hard to guess what went on in the orginal designer's mind. Usually it's something smart that I did not understand, sometimes it's something silly that the designer did not understand.

Often, simulating the mystery circuitry helps to clear things up. Sometimes you have to grind the math.

Circuitry that looks superfluous may be there to deal with special conditions, such as power-up or down, or surge protection. It may be there to improve linearity/distortion, for temperature compensation, or it may be a stability or bandwidth tweak. Many other reasons are possible.

In patents or proprietary circuitry, it may be there to confuse or mislead.

Jeroen Belleman

Reply to
Jeroen Belleman

If I posted my schematics, I'd totally be guilty of this :D

(Although that being said, mine are very simplistic much of the time, so there's not very much to "get wrong" anyway)

Reply to
Dan Purgert

In my misspent youth, one of the least useless things I did was to go through app notes, assume that all the circuits were junk, and figuring out what was wrong with them.

Lots of them were perfectly functional, but I could usually find valid criticisms. Some had lock-up states, or needed protection circuits, or used the wrong parts just because they happened to sell them.

So it’s a pretty good way to learn about circuits.

Cheers

Phil Hobbs

Reply to
Phil Hobbs

Nice to know I'm not alone! To take just the latest example of the phenomenon, I've been working on a vintage oscilloscope which was DOA:a Tektronix 7000 series from 1970. The PSU section was faulty. I've got it working now, but not through understanding exactly how it works; just old-style trouble shooting. Anyway, I was curious about the +5V regulator section. Here's the regulator board in its totality:

formatting link
And here's a close-up of the 5V section:
formatting link
There's a pair of transistors contained within a monolithic array of five in all. U973E and U973D and 'D' had blown and the output was sitting at just over 11V instead of the correct level of 5V. Upon replacing 'D', the output reverted to +5V. That was the last of the outputs requiring attention and the PSU now works, but I still don't understand the purpose of the 'E' and 'D' pairing.

Reply to
Cursitor Doom

Yes. Also I've run into some that are design flaws. Ed

Reply to
ehsjr

Shame you locked them behind whatever "yandex" is.

Reply to
Dan Purgert

Those links require a login to view. But if this scope is the TEK7623A you asked about in May 2023 then U973D and U973E form a long tail pair differential amplifier (in essence an op amp).

The 5V sense line is divided by 4kohm and 40kohm against the -50V rail so that when in regulation the feedback voltage is zero, that is the input to U973E base. The other half of the long tail pair U973D is doing the comparison to zero (it's base is connected to zero volts by a resistor).

U973A and B form another diff amp for the current limit function. They used similar structures for the -50V +50V -15V and +15V regulators.

piglet

Reply to
piglet

Shouldn't be an issue. I've posted this link before a few months ago and no one had any problem accessing it - or to be precise, no one mentioned any issues accessing it. Perhaps the access expires after a period. Try this:

formatting link
No way does this need a password.

Reply to
Cursitor Doom

+1. There are a lot of circuits designed by stringing together app notes, selected by keyword searches.

Similar to coding out of Stack Overflow.

Cheers

Phil Hobbs

Reply to
Phil Hobbs

Basic diff amp. +5V is proportional to -50V rail.

RL

Reply to
legg

We add strategically-placed 0-ohm jumpers in places that might need a bead or anti-snivet resistor.

Otherwise, pop options are mostly power supply things or range selection.

Cheers

Phil Hobbs

Reply to
Phil Hobbs

Well, it's very tempting as those app note fragments often do the trick. All other things remaining the same, of course.

Reply to
Cursitor Doom

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.