50-ohm power resistors

The chimney effect is why an extruded heat sink works best with the long axis of its fins oriented vertically. It's also why the thermal resistance is a decreasing function of power dissipation.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
 Click to see the full signature
Reply to
Phil Hobbs
Loading thread data ...

That doesn't look like it's a wideband dummy load, with those Wilkinson splitters in it.

Jeroen Belleman

Reply to
Jeroen Belleman

rk> >>

te:

ote:

s end

king

or him.

yeh, you can see on the plots that it is for ~1GHz, at low frequency the four 50R resistors are in parallel

-Lasse

Reply to
Lasse Langwadt Christensen

Just in case it wasn't entirely clear my comment was aimed at Mr Sloman, who seems to have a bit of a primary schoolkid like attitude going on.

Engineering & business skills rather than maths & physics.

NT

Reply to
tabbypurr

Those who pay attention would know that it was Dr.Sloman, and that designing transformers for a specific job isn't a primary school activity, which is why John Larkin chickens out of it.

Twitting John Larkin for his anti-intellectual attitudes may be a bit frivolous but he's being twitted fro making Tulane look bad, and Tulane is a tertiary education establishment (more or less).

But one of the business skills is knowing how much math and physics to deploy to get the optimal product. One definition of an engineer is somebody who can do for a dollar what any damn fool can do for two.

Insisting on using off-the-shelf transformers when you can get something smaller and better wound specifically for the job looks very like spending two dollars when you don't have to.

--
Bill Sloman, Sydney
Reply to
bill.sloman

On Saturday, September 17, 2016 at 9:00:02 PM UTC-4, snipped-for-privacy@ieee.org wrot e:

cific job isn't a primary school activity,

Actually designing transformers is pretty simple as long as you are using s tock lamination sizes. It would be fairly simple to make a computer progra m to come up with designs for transformers say under 1 kva. Or could be do ne in a spreadsheet. Reference Data for Radio Engineers used to have two or three pages on designing transformers.

Dan

Reply to
dcaster

You can wind your own capacitors, too.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

If you have to use an air current detector to see such small currents, you don't have nearly enough air to make a significant difference in heat removal.

--

Rick C
Reply to
rickman

I think you are confusing regular convection with the chimney effect. If you don't have a a chimney, it isn't the chimney effect.

--

Rick C
Reply to
rickman

That sounds horrible. It can be done on a PIC in probably as much time, if the iteration is as naive as your description sounds...

Simply use the convergent with denominator closest to 2^N without going over:

formatting link
You are /guaranteed/ to get the closest fraction, out of all 2^2N possible fractions. The growth rate is guaranteed equal or better than the series for phi (which are Fibonacci numbers, so grow as a modest exponential), so for N bits you need O(N) steps. It's fast, and it's all integer multiplication and addition.

An 800MHz ARM should be able to do it real-time in the megasamples.

If applied to PWM, you can also guarantee the frequency within useful bounds (i.e., the denominator being between 2^(N-1) and 2^N - 1) by left-shifting the numerator and denominator until the denominator's top bit is set. In this way, you can get several bits improvement in average-case accuracy. Combined with dithering, there might even be a general-case solution, i.e. that allows one to fill in between all the denominator=MAXVAL codes.

Hmm. Dithering between different PWM frequencies and codes is an interesting problem. It's time-weighted, after all. I will have to think about that.

The Pade approximant also applies to functions, which is helpful for curve fitting and network analysis. Maybe not as useful for embedded and DSP because of the division, but that can still be worthwhile.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
 Click to see the full signature
Reply to
Tim Williams

Well, that's blatantly, on-the-face wrong. Perfectly still air is a fantastic insulator. We fill our homes' walls with the stuff!

Vacuum is a good number of times even better (but with an even greater tempco), but that's a luxury reserved for "keeping hot things hot and cold things cold".

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
 Click to see the full signature
Reply to
Tim Williams

ded

nt

nt.

g

A chimney just makes regular convection more effective. You get a longer co lumn, of hotter, less dense air above your heat source which creates a larg er pressure difference, and a higher flow rate across the heat source, than you would get without the chimney.

--

Bill Sloman, Sydney
Reply to
bill.sloman

ce

eeded

rent

vent.

ong

column, of hotter, less dense air above your heat source which creates a la rger pressure difference, and a higher flow rate across the heat source, th an you would get without the chimney.

it has to be properly sized an insulated to really work well

-Lasse

Reply to
Lasse Langwadt Christensen

ieee.org:

ance

needed

urrent

r vent.

long

r column, of hotter, less dense air above your heat source which creates a larger pressure difference, and a higher flow rate across the heat source, than you would get without the chimney.

Sure. It's an optimisation problem. You use the chimney to make a longer co lumn of hot air, and thus a larger pressure differential, but the air has t o flow through the chimney, which creates drag.

The more air you draw through the chimney, the cooler the air in the chimne y - you only have a fixed amount of heat to be dissipated, and the more air you spread it across, the cooler the air column.

Lots of variables.

--
Bill Sloman, Sydney
Reply to
bill.sloman

It's not logical that an algorithm would run in the same time on a PIC and an 800 MHz full ARM with hardware FP.

From his verbal description, that's probably what he did.I think he used double floats, which I think are faster on the ARM than using ints.

Average looks to be below 10 us, but occasionally 100s of usec, maybe even more. We could budget a couple hundred usec maybe.

We don't want period jitter, so dithering is out. This is being considered as an alternate to DDS, which has jitter and other issues.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

rote:

specific job isn't a primary school activity,

g stock lamination sizes. It would be fairly simple to make a computer pro gram to come up with designs for transformers say under 1 kva. Or could be done in a spreadsheet. Reference Data for Radio Engineers used to have two or three pages on designing transformers.

Fewer variables to manipulate. Your chances of being able to buy pretty muc h exactly what you need are a whole lot higher with capacitors than they ar e with transformers. But I have used a gap in a printed circuit track as a capacitor.

--
Bill Sloman, Sydney
Reply to
bill.sloman

Yup, and slowly moving air is right behind still air for insulating. Read nearly any spec sheet on an air cooled device. It will give thermal ratings for various ft/s which are far above speeds that require tissue paper to measure.

--

Rick C
Reply to
rickman

You say "just", but that is the point. A chimney is much more effective than fins without a chimney. That is the point.

--

Rick C
Reply to
rickman

You don't "draw" the air without heating it. The heat is what creates the movement. So talking about moving more air with a fixed about of heat makes no sense.

--

Rick C
Reply to
rickman

ance

needed

urrent

r vent.

long

.

ger column, of hotter, less dense air above your heat source which creates a larger pressure difference, and a higher flow rate across the heat source , than you would get without the chimney.

r column of hot air, and thus a larger pressure differential, but the air h as to flow through the chimney, which creates drag.

imney - you only have a fixed amount of heat to be dissipated, and the more air you spread it across, the cooler the air column.

It doesn't make any sense to you because you can't do quantitative thinking .

Twice the amount of air at half the temperature takes up the same amount of heat. Because it's at half the temperature you need twice as long a chimne y to get the same pressure difference to drive the flow, and a root two lar ger diameter chimney for that pressure difference to generate twice the vol ume flow rate through the chimney.

Perfect sense ...

--
Bill Sloman, Sydney
Reply to
bill.sloman

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.