PP Amplifier output impedance matching

[Diabolical Artificer]

When I was struggling with this, not knowing much I used to worry about minor details like the one your dealing with, I found just cracking on, building a circuit on the bench, then testing it by putting different loads on the secondary taught me more in half hour than hours slogging through theory.

"there are no correct answers in all branches of engineering" If you read any thread, ask any question you'll get several answers and probably get more confused, best not to worry and do what you want anyway.

Andy.

[Amie]

Hi guys

Going through this has been helpful in understanding the process of what's actually happening in the circuit.

The theory at least heads you off in the right direction!

Dad used to say not to worry about too many decimal places, he said that one of his tutors took his very expensive new calculator away from him (late 1970's) and gave him a little six inch slide rule, telling him it would be as accurate as he would ever need for in electronics! (Boots ringplan, I still treasure it).

Amie.

[ppppenguin]

An ounce of theory often saves a ton of practice but there's often no substitute for trying something. Part of the art of engineering is recognising what mix of theory and practice will get you where you want to be.

Your dad was very wise. Never confuse precision with accuracy. Slide rule accuracy (2 or 3 signifcant digits) is good enough for most purposes. The slide rule also teaches you about orders of magnitude since you have to do those by hand. When using a calculator it's very easy to out by a factor of 10 or 100, simply by miskeying. "It must be right, that's what it says on the calculator". If you've used a slide rule or mental arithmetic then you will gain a feel for a rough answer and spot those errors.

[Terry]

Your dad was a good teacher.

Much of the stuff we discuss here was made in the days when 20% tolerance resistors were the norm and all meters were analogue. Parts of the resistance scales of these meters could be very cramped, thus limiting the accuracy with which they could be read.

Fast forward to the 21st Century and someone pops up here or one of the other vintage forums. They are armed with a nice accurate DVM and post something like: "I changed R27 because it had increased in value from 220k to 238k ..."

Of course, it made no difference because it could have been anywhere from 176k and 264k and still have been within tolerance! It probably hasn't changed in value from the day it was manufactured.

I created a chart to illustrate the ranges covered by the various popular tolerances which you might find of interest. I'm sure I've posted it here before but can't find it, so here it is again:



You probably already know this already but the E6, E12 & E24 ranges denote the number of values in each decade.

[Kalee20]

Dad used to say not to worry about too many decimal places, he said that one of his tutors took his very expensive new calculator away from him (late 1970's) and gave him a little six inch slide rule, telling him it would be as accurate as he would ever need for in electronics! (Boots ringplan, I still treasure it).

Largely true, but you do need to be careful.

I quite often measure power going into (volts x amps) and out of (different volts x different amps) a transformer. usually the powers are nearly equal; subtracting them gives the losses in the transformer.

When you subtract two nearly-equal numbers, a small change in either can make a massive change to the answer. Example, 990W out, 1000W in, so 10W lost. But make a 1% uncertainty in input power (which is about the limit of a slide rule), thus I may be from 990W to 1010W in, and the losses could be from zero to 20W, a +/- 100% change! And this could mean the difference between needing a fan to cool it, or not needing a fan.

Accurate calculators are great, and I have access to as many figures of accuracy as I need, which wouldn't be the case with a slide-rule. (Though, I still know how to work out with pencil and paper)!

[BusyBee]

Maybe not relevant here but with computers (there are ancient digital ones) the binary arithmetic and hex etc, if you get down to machine level, needs to be exact. No slide rule there. Yes computer math can be an exception but there may be others. With lower resolution ADC etc one binary digit may mean appreciable error, and always something to be aware of. I must admit though that I have not used hex arithmatic or conversion for some years. What little I do is straightforward in my head.

Tracy

[Terry]

Largely true, but you do need to be careful.

Yes, but Amie has already demonstrated a determination to learn and obviously knew quite a lot before she started posting here. I'm quite sure that, if the needs to tackle the sort of problems raised by you and Tracy her natural instincts will warn her when absolute accuracy is vital.

For now, though, I don't think she needs to worry!

[boater sam]

Amie, you have disabled PMs.

I have not seen the cheque yet.
Sam.

[pwdrive]

ECL80, the old Mullard push pull article from the Valves, Tubes & Circuits publications (No:2)


  mullard VT&C Push pull ECL80 amp.pdf (Size: 1.49 MB / Downloads: 14)

Lawrence.

[Amie]

Amie, you have disabled PMs.

Sorry, mum won't allow PMs normally. I will ask if I can turn it on again.

I have not seen the cheque yet.

Posted it on the 16th, second class, but should be there by now.
I Will get another one sent you, please PM address again.

Sorry,
Amie.