strange complementary pair

CCS-loaded P-fet with bootstrapped drain load on the input FET.

Will there be feedback around that? The gain as-is isn't well defined I don't think

Seriously. Siliconix and International Rectifier published those topologies 30 years ago.

Then it just shows how clever I am to re-invent them.

Or you just reinvented them after forgetting you had glanced at the ap notes sheets all those many years ago, like we all do...

Hmm, would that error - forgetting something - disqualify you for being considered to be able to run for President?

John ;-#)#

No. I invent circuits.

It surprises me how many people can't.

Or think you do.

Some people invent stuff which is novel enough to patent. John Larkin doesn't seem to be one of them.

I'm looking forward to him inventing my current mirror variation on Baxandall's class-D oscillator (which nobody has bothered to patent, but I've not seen anywhere either).

I had a boss like that once, only it took him as little as 7 days to forget that some ideas weren't actually his own.

RL

The only boss I ever had was just the opposite. I won an argument with him over a technical design point. He was a Ph.D. in biomedical electronics. I was 19 and had taught myself electronics over the past 1.5 years. When the internet came to my region 28 years later, I found his email address and contacted him. One of the first things he said in his reply was to mention that project and went on to say "You taught me a good lesson that day......"

Read The X-Chapters. It has some interesting circuits.

I've known people who think that all circuits come from books. They don't explain how those circuits got into books.

This is an electronic design discussion group. I figure that a couple of per cent of the people who post here can actually design (or even understand) circuits, which is why the rest are so active in off-topic blather, where their rightness or wrongness is un-provable and their responses to ideas are mostly childish insults.

The upper fet is not a constant-current source.

The general term for that was, in my experience,'insubordination'.

Not always. I managed to get around the mechanical boffin once, getting my department head to forward rough sketches to a vendor, for what mech had assured us was a dead end. The (cheap!)quote and free, fully-functioning samples came back in less than 5 business days. . . . . and then the mech guy got his hands on the sketches, to enter them into the CAD system. Although the results did actually function (well, mostly - with a lerger temperature rise, oversized profile and 30% more dollar-per-inch 3xfep litz),the additional details, ovesized dimensioning (for tolerance) and general messing about made the part an embarrassment to introduce into production.

RL

ologies 30 years ago.

There are a lot of good circuits in books, and - if the book is well writte n and properly indexed - you can usually find something useful a lot faster than you can invent it. Most people do understand that somebody invented e very last one of the circuits that get into books - and that most them are now obvious to those skilled in the art.

I would like to know how John Larkin thinks he figured that out. He's much more interested in presenting his own circuits than he is in commenting on other people's. Presumably he figures that people who don't admire his cir cuits can't understand them, and neglects the cases where they are perfectl y comprehensible but less than admirable.

htness or wrongness is un-provable and their responses to ideas are mostly childish insults.

John Larkin's off-topic opinions about climate change do happen to be wrong , and this is demonstrable, but not in any way that he can recognise, and h e does get testy when people point this out.

if it takes so long to make that the patent is expired before it is done why bother ;)

In science, there is always one correct theory, so scientists are motivated to find it and scorn others. They tend to brutally mock a colleague who may be in error; I've seen it happen in person several times.

Scientists also tend to believe that having an advanced degree makes them smarter than everyone else.

There are more functional circuits you can design from Digikey parts than there are photons in the universe, and any non-trivial electronic problem has many solutions, and it's not provable which is best.

So scientists tend to be mediocre circuit designers. Read The Review Of Scientific Instruments for amusing examples.

Right. As people copy your designs, just stay a generation ahead.

Designing is fun. Patent applications aren't fun.

I've been very fortunate in managers. My first one was a fine microwave engineer named Josef Fikart, who escaped from Czechoslovakia in 1968 and proceeded to mentor three generations of young radio engineers including yours truly.

When I left there and went to grad school, I didn't always get on with my thesis adviser Gordon Kino, but he taught me a great deal and always fought my corner, something that does not grow on trees.

At IBM I had one boss who was too chicken to take even smart risks--a nice enough old buffer but frustrating to work for. I shopped my project around to other departments and eventually wound up in Packaging Research, which was a strange place to be doing antenna-coupled MIM tunnel junctions, but hey, they were nice guys who signed my purchase reqs. ;)

I've only ever had one really bad one, also at IBM but a decade earlier. He was the guy who I worked for as a postdoc when I was 28 years old. My postdoc project was weird--it was to take the interferometric-readout attractive-mode atomic force microscope and make a product out of it. It was called the SXM, for "scanned anything microscope", and wound up being sold by IBM for some years and then by Veeco for another 20 years or so.

The root of the trouble was that my expertise overlapped my boss's, specifically in the area of measurement physics and optics. He really wanted to micromanage the project, even though I had two years' product design experience and was a circuits guy from way back. He was a very theoretically-oriented EE with zero product experience. His optics was okay, but his circuit ideas were, speaking charitably, naive.

He was also a political operator and rather vain, so to get his way, he set up a group meeting and presented his ideas on how my gizmo should work. This left me with two choices of how to fail. First, I could knuckle under and build something that I knew wouldn't work, or second, I could take his proposal apart in public and not have my post-doc renewed. (I had a wife and daughter at this point.)

I naturally chose #2.

The following week, he did it again, and I did it again. That was when he stopped coming round to my lab at the end of the day to see how things were going, so I realized I needed to switch groups fast. Once the SXM was transferred to the product group in Boca Raton, I jumped over to another group run by a very smart guy named Sam Batchelder, who's been a friend from that day to this. He got my postdoc renewed and found the right project for me to work on to get me hired as a research staff member. (Sam is really one of the ones.)

Cheers

Phil Hobbs

I think it was mighty gracious of my former boss to acknowledge that he could learn something from a self-taught teenager. He was a New Zealander working in the US and was on loan to the institution. It was he who invited me to work at the medical institution when I was a patient there, some 3000 km from home. We had some good times together.

We built a prototype for a new type of artificial pacemaker. I helped design a digital patient monitoring system. He mentioned those things too when we briefly renewed contact after nearly 30 years. I did some work of my own for the labs and for the Wellcome Research wing which was headed by Dr.James Baker, an American. I was in the midst of designing a heart-lung machine when I left.

At the same time, I was also responsible for the maintenance of practically all the electrical and electronic equipment in the institution which was then one of the biggest in Asia. When I complained that my workload was a bit too much for one person, my boss said that it was too much for five people.

It depends on the load impedance. If there's no load there will be voltage amplification but obv. there can't be any difference in current between the output devices, there's only one current path available, assuming the strapping resistor is relatively large.

Yikes, it doesn't even need the p-fet.

Thanks, can you show that too? It isn't actually a "follower" as an increasingly positive input makes a falling output.

piglet