Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Interesting curves...

120 views
Skip to first unread message

Fred Bartoli

unread,
Nov 6, 2012, 8:54:39 AM11/6/12
to
See 2nd graph on page 4 (efficiency vs power)

http://www.macomtech.com/datasheets/MAGX-000912-125L00%20Prod%2018%20Aug%2011.pdf


--
Thanks,
Fred.

John Larkin

unread,
Nov 6, 2012, 11:19:31 AM11/6/12
to
On Tue, 06 Nov 2012 14:54:39 +0100, Fred Bartoli <" "> wrote:

>See 2nd graph on page 4 (efficiency vs power)
>
>http://www.macomtech.com/datasheets/MAGX-000912-125L00%20Prod%2018%20Aug%2011.pdf

How annoying. No Idss spec, no DC curves. RF people are like that:
just fiddle with the bias until it works.

Since the abs max range of gate voltage is -8 to -2, it's presumably
destroyed in shipping. Unless they ship it with a battery attached.

Abs max supply voltage is 48, but leakage is tested at 175.

The forward transconductance spec is clearly wrong.

Pitiful. We use phemts in pulsed time-domain applications, and usually
have to test them ourselves to get basic specs. Even the manufacturers
often don't know basic stuff about their parts. A Spice model, or even
basic DC curves, is rare for RF parts.


--

John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

Vladimir Vassilevsky

unread,
Nov 6, 2012, 11:34:39 AM11/6/12
to

"Fred Bartoli" <" "> wrote in message
news:5099169f$0$1921$426a...@news.free.fr...
> See 2nd graph on page 4 (efficiency vs power)
>
> http://www.macomtech.com/datasheets/MAGX-000912-125L00%20Prod%2018%20Aug%2011.pdf
>

Well, it is clear why. However the one who prepares datasheet doesn't seem
to have a clue.

VLV



John Larkin

unread,
Nov 6, 2012, 1:38:43 PM11/6/12
to
On Tue, 06 Nov 2012 08:19:31 -0800, John Larkin
<jjla...@highNOTlandTHIStechnologyPART.com> wrote:

>On Tue, 06 Nov 2012 14:54:39 +0100, Fred Bartoli <" "> wrote:
>
>>See 2nd graph on page 4 (efficiency vs power)
>>
>>http://www.macomtech.com/datasheets/MAGX-000912-125L00%20Prod%2018%20Aug%2011.pdf
>
>How annoying. No Idss spec, no DC curves. RF people are like that:
>just fiddle with the bias until it works.
>
>Since the abs max range of gate voltage is -8 to -2, it's presumably
>destroyed in shipping. Unless they ship it with a battery attached.
>
>Abs max supply voltage is 48, but leakage is tested at 175.
>
>The forward transconductance spec is clearly wrong.
>
>Pitiful. We use phemts in pulsed time-domain applications, and usually
>have to test them ourselves to get basic specs. Even the manufacturers
>often don't know basic stuff about their parts. A Spice model, or even
>basic DC curves, is rare for RF parts.

Which is another job idea: offer to proof-read peoples's data sheets.
That would be the ideal job for some older guy who knows everything
about everything.


--

John Larkin Highland Technology, Inc

jlarkin at highlandtechnology dot com
http://www.highlandtechnology.com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser drivers and controllers
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro acquisition and simulation

Bill Sloman

unread,
Nov 6, 2012, 6:05:32 PM11/6/12
to
On Nov 7, 5:38 am, John Larkin <jlar...@highlandtechnology.com> wrote:
> On Tue, 06 Nov 2012 08:19:31 -0800, John Larkin
>
>
>
>
>
>
>
>
>
> <jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> >On Tue, 06 Nov 2012 14:54:39 +0100, Fred Bartoli <" "> wrote:
>
> >>See 2nd graph on page 4 (efficiency vs power)
>
> >>http://www.macomtech.com/datasheets/MAGX-000912-125L00%20Prod%2018%20...
>
> >How annoying. No Idss spec, no DC curves. RF people are like that:
> >just fiddle with the bias until it works.
>
> >Since the abs max range of gate voltage is -8 to -2, it's presumably
> >destroyed in shipping. Unless they ship it with a battery attached.
>
> >Abs max supply voltage is 48, but leakage is tested at 175.
>
> >The forward transconductance spec is clearly wrong.
>
> >Pitiful. We use phemts in pulsed time-domain applications, and usually
> >have to test them ourselves to get basic specs. Even the manufacturers
> >often don't know basic stuff about their parts. A Spice model, or even
> >basic DC curves, is rare for RF parts.
>
> Which is another job idea: offer to proof-read peoples's data sheets.
> That would be the ideal job for some older guy who knows everything
> about everything.

John Larkin furthers his claim to be our most out-of-touch-with-
reality poster.

I've never made any claim to be particularly expert with RF parts.
I've certainly used wide-band transistors at frequencies close to
1GHz, but strictly as wide-band amplifiers, and I'm aware that there
are lots of fun aspects to their narrower-band use where I know very
little - possibly even less than John Larkin.

--
Bill Sloman, Sydney

Jamie

unread,
Nov 6, 2012, 7:20:10 PM11/6/12
to
John Larkin wrote:

> On Tue, 06 Nov 2012 14:54:39 +0100, Fred Bartoli <" "> wrote:
>
>
>>See 2nd graph on page 4 (efficiency vs power)
>>
>>http://www.macomtech.com/datasheets/MAGX-000912-125L00%20Prod%2018%20Aug%2011.pdf
>
>
> How annoying. No Idss spec, no DC curves. RF people are like that:
> just fiddle with the bias until it works.
>
> Since the abs max range of gate voltage is -8 to -2, it's presumably
> destroyed in shipping. Unless they ship it with a battery attached.
>
> Abs max supply voltage is 48, but leakage is tested at 175.
>
> The forward transconductance spec is clearly wrong.
>
> Pitiful. We use phemts in pulsed time-domain applications, and usually
> have to test them ourselves to get basic specs. Even the manufacturers
> often don't know basic stuff about their parts. A Spice model, or even
> basic DC curves, is rare for RF parts.
>
>
Don't feel bad, the other day I discovered that we have a tube of power
mosfets 18 amps, 500V out of a batch we order with a different date code
on them that does not pass the 500V break down test, they're good for
only ~250v before they start to break down. These were purchased all at
the same time but this particular tube seems to be older stock.

Not only that, I notice the package, although the same type, looks
like they were made else where on the planet, they don't seem to have
the same exact mold shape. I wonder if I got rebranded or defective parts?

It's not a bug deal, we just put those assigned with a note on them,
they now are test fets, but these fets are not cheap and it's a good
thing I didn't pay for them..

Jamie

John Larkin

unread,
Nov 6, 2012, 7:28:57 PM11/6/12
to
I keep making great suggestions about fun and useful things that
geezers could do, and geezers pop up and tell me what they don't know
and can't do.

Fred Bartoli

unread,
Nov 6, 2012, 8:35:51 PM11/6/12
to
Vladimir Vassilevsky a �crit :
I meant the small hook at the curves upper right extremity.

I'd be curious to know what so obvious about these...

--
Thanks,
Fred.

Vladimir Vassilevsky

unread,
Nov 6, 2012, 9:06:06 PM11/6/12
to

"Fred Bartoli" <" "> wrote:
> Vladimir Vassilevsky a écrit :
>> "Fred Bartoli" <" "> wrote:
>>> See 2nd graph on page 4 (efficiency vs power)
>>>
>>> http://www.macomtech.com/datasheets/MAGX-000912-125L00%20Prod%2018%20Aug%2011.pdf
>>>
>> Well, it is clear why. However the one who prepares datasheet doesn't
>> seem to have a clue.
>>
> I meant the small hook at the curves upper right extremity.
>
> I'd be curious to know what so obvious about these...
>

On the previous plot, they have output power vs input power. The output
power decreases after passing maximum due to overexitation. They just put
sequential numbers of the output power from that plot into drain efficiency
calculation. So, after maximum reaching maximum power, the graph makes a
hook.

Vladimir Vassilevsky
DSP and Mixed Signal Consultant
www.abvolt.com


Robert Baer

unread,
Nov 6, 2012, 9:09:19 PM11/6/12
to
Fred Bartoli wrote:
> See 2nd graph on page 4 (efficiency vs power)
>
> http://www.macomtech.com/datasheets/MAGX-000912-125L00%20Prod%2018%20Aug%2011.pdf
>
>
>
Something wrong . . look carefully at the last graph on page 4.
One cannot have _two_ efficiencies at a given output power.

Robert Baer

unread,
Nov 6, 2012, 9:10:46 PM11/6/12
to
Fred Bartoli wrote:
> See 2nd graph on page 4 (efficiency vs power)
>
> http://www.macomtech.com/datasheets/MAGX-000912-125L00%20Prod%2018%20Aug%2011.pdf
>
>
>

Robert Baer

unread,
Nov 6, 2012, 9:13:19 PM11/6/12
to
Maybe counterfeit parts..a lot of that going on.

Robert Baer

unread,
Nov 6, 2012, 9:17:43 PM11/6/12
to
Maybe they were shot down by those big companies that need expertise.
HR takes one look at them and instantly determine they are 40 or
older and kick them out as being "overqualified".
The companies would rather hire two greenies fresh out of college and
$$pend even more money in training for a year or so to get them to be
useful..

k...@att.bizzz

unread,
Nov 6, 2012, 11:36:17 PM11/6/12
to
Absurd. It's just the opposite, in fact. Recent grads are living in
mommy's basement, playing the XBox.

John Fields

unread,
Nov 7, 2012, 8:14:30 AM11/7/12
to
JF

Bill Sloman

unread,
Nov 7, 2012, 9:55:01 AM11/7/12
to
On Nov 7, 11:29 am, John Larkin <jlar...@highlandtechnology.com>
wrote:
It's the contrast between ill-informed fantasies and objective
reality. The US electorate seems to have preferred the latter.

--
Bill Sloman, Sydney

legg

unread,
Nov 7, 2012, 10:18:26 AM11/7/12
to
On Tue, 06 Nov 2012 10:38:43 -0800, John Larkin
<jla...@highlandtechnology.com> wrote:

>On Tue, 06 Nov 2012 08:19:31 -0800, John Larkin
><jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>
>>On Tue, 06 Nov 2012 14:54:39 +0100, Fred Bartoli <" "> wrote:
>>
>>>See 2nd graph on page 4 (efficiency vs power)
>>>
>>>http://www.macomtech.com/datasheets/MAGX-000912-125L00%20Prod%2018%20Aug%2011.pdf
>>
>>How annoying. No Idss spec, no DC curves. RF people are like that:
>>just fiddle with the bias until it works.
>>
>>Since the abs max range of gate voltage is -8 to -2, it's presumably
>>destroyed in shipping. Unless they ship it with a battery attached.
>>
>>Abs max supply voltage is 48, but leakage is tested at 175.
>>
>>The forward transconductance spec is clearly wrong.
>>
>>Pitiful. We use phemts in pulsed time-domain applications, and usually
>>have to test them ourselves to get basic specs. Even the manufacturers
>>often don't know basic stuff about their parts. A Spice model, or even
>>basic DC curves, is rare for RF parts.
>
>Which is another job idea: offer to proof-read peoples's data sheets.
>That would be the ideal job for some older guy who knows everything
>about everything.

It doesn't pay.

A typo in the hand is worth two proofs in the bush.

RL

Phil Hobbs

unread,
Nov 7, 2012, 10:27:53 AM11/7/12
to
Sure you can. There's nothing inherent in the idea of an amplifier that
prevents the output power from going down as you increase the input
power, past a certain point. All you need is a bias shift.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

legg

unread,
Nov 7, 2012, 10:31:23 AM11/7/12
to
Can't blame the spreadsheet software.....

Actually, Excel used efficiency as the dependant coordinate for
drawing the line connection sequence. Efficiency should drop with
over-exitation, if input drive power is included in total power
consumed.

There are valid bistable graphical plots, but this isn't one.

RL

Jamie

unread,
Nov 7, 2012, 8:19:42 PM11/7/12
to
You'd make a perfect politician, never getting the facts right and
always twisting the truth, if you really knew what the truth was.

Jamie

Robert Baer

unread,
Nov 7, 2012, 9:05:11 PM11/7/12
to
..here come de puns..
Well, it was obvious as to the why and what, but the giraffe had its
neck broken by the plodding idiot by interchanging the axis.
Just rotate the giraffe 90 degrees so it will be standing on its legs
and thus have no broken neck..

Robert Baer

unread,
Nov 7, 2012, 9:07:19 PM11/7/12
to
I was talking about the job market before His Changeness.

Robert Baer

unread,
Nov 7, 2012, 9:09:26 PM11/7/12
to
Phil Hobbs wrote:
> On 11/06/2012 09:10 PM, Robert Baer wrote:
>> Fred Bartoli wrote:
>>> See 2nd graph on page 4 (efficiency vs power)
>>>
>>> http://www.macomtech.com/datasheets/MAGX-000912-125L00%20Prod%2018%20Aug%2011.pdf
>>>
>>>
>>>
>>>
>>>
>> Something wrong . . look carefully at the last graph on page 4.
>> One cannot have _two_ efficiencies at a given output power.
>>
>
> Sure you can. There's nothing inherent in the idea of an amplifier that
> prevents the output power from going down as you increase the input
> power, past a certain point. All you need is a bias shift.
>
> Cheers
>
> Phil Hobbs
>
Yea, yea, yea. Bot the presentation is absurd; rotate the graph 90
degrees..

k...@att.bizzz

unread,
Nov 7, 2012, 11:33:47 PM11/7/12
to
On Wed, 07 Nov 2012 18:07:19 -0800, Robert Baer
It's been true for some time. It's hard for script-kiddies to find
jobs. Experienced design engineers aren't having a lot of trouble
finding work. While I have experienced some age discrimination, it's
just as often been reverse.

Bill Sloman

unread,
Nov 7, 2012, 11:53:59 PM11/7/12
to
On Nov 8, 3:33 pm, k...@att.bizzz wrote:
> On Wed, 07 Nov 2012 18:07:19 -0800, Robert Baer
>
>
>
>
>
>
>
>
>
> <robertb...@localnet.com> wrote:
> >k...@att.bizzz wrote:
> >> On Tue, 06 Nov 2012 18:17:43 -0800, Robert Baer
> >> <robertb...@localnet.com>  wrote:
You've persuaded other people that at least some of the elderly should
be discriminated against?

I'd prefer that you hadn't gone to the trouble, particularly in your
posts here.

--
Bill Sloman, Sydney

Bill Sloman

unread,
Nov 8, 2012, 5:50:08 AM11/8/12
to
On Nov 8, 12:12 pm, Jamie
<jamie_ka1lpa_not_valid_after_ka1l...@charter.net> wrote:
> BillSlomanwrote:
>    You'd make a perfect politician, never getting the facts right

Find the fact I've got wrong.

> and always twisting the truth, if you really knew what the truth was.

I've probably got a better idea of what is true and what isn't than
you and John Larkin put together (and that's a depressing image).

--
Bill Sloman, Sydney

k...@att.bizzz

unread,
Nov 8, 2012, 10:22:10 AM11/8/12
to
A sane person couldn't come to that conclusion with a dart.

>I'd prefer that you hadn't gone to the trouble, particularly in your
>posts here.

Slowman, NO ONE cares what you'd prefer. Really!

Phil Hobbs

unread,
Nov 8, 2012, 11:01:07 AM11/8/12
to
It's a strange graph, for sure, but surely plotting efficiency on the X
axis would be much stranger, because that isn't the independent
variable. It would look better plotted vs input power, but then people
might miss the fact that the output power goes down as you crank up the
input power past a certain point.

John Larkin

unread,
Nov 8, 2012, 11:33:18 AM11/8/12
to
No wonder nobody will hire you. You delight in being repulsive. And
you're pretty clumsy at it.

Go work on that oscillator for a few more years.

I just finished the hardware design for a USB-interfaced magnetic
field mapper. It uses up to 32 tiny sense coils in a motorized
fixture, surrounding a NMR/MRI pulsed-gradient coil, to pick up
internal and external (leakage) fields. All it needs now is ARM code.
I won't do that, but I will document the algorithms, pseudocode, and
timing requirements, and leave all that in the user manual for future
reference.

This is the sort of thing you could do, if you cared to.


--

John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

Bill Sloman

unread,
Nov 8, 2012, 4:27:00 PM11/8/12
to
On Nov 9, 2:22 am, k...@att.bizzz wrote:
> On Wed, 7 Nov 2012 20:53:59 -0800 (PST),BillSloman
Which sane person did you consult to acquire that - erroneous -
opinion?

> >I'd prefer that you hadn't gone to the trouble, particularly in your
> >posts here.
>
> Slowman, NO ONE cares what you'd prefer.  Really!

krw, I'm not in the least surprised that that's your opinion. That
you've bothered to remind me - and the rest of the user-group - that
you are still a psychopathic idiot just confirms your status as a
total waste of bandwidth.

--
Bill Sloman, Sydney

Bill Sloman

unread,
Nov 8, 2012, 4:46:47 PM11/8/12
to
On Nov 9, 3:33 am, John Larkin
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Thu, 8 Nov 2012 02:50:08 -0800 (PST),BillSloman
A habit I've probably learned from you.

>And you're pretty clumsy at it.

Dream on.

> Go work on that oscillator for a few more years.

It shouldn't take that long. I've not yet been working on the current
version for a whole year yet, and I ought to be able to get the
hardware together well within the year, despite having changed
countries in mid-year. As usual, you've got your facts wrong.

> I just finished the hardware design for a USB-interfaced magnetic
> field mapper. It uses up to 32 tiny sense coils in a motorized
> fixture, surrounding a NMR/MRI pulsed-gradient coil, to pick up
> internal and external (leakage) fields. All it needs now is ARM code.
> I won't do that, but I will document the algorithms, pseudocode, and
> timing requirements, and leave all that in the user manual for future
> reference.

Who commissioned the design? And did they tell you why they needed 32
sense coils? Six would be the canonical minimum.
Three orthogonal coils at each point would boost that to 18. Eight
more sets of coils would fill in the corners of the cube, but that
would be 42 orthogonal coils. 32 is a strange number in that
application.

> This is the sort of thing you could do, if you cared to.

And had any way of beating you to your customer, who have to be mad to
prefer me - working out of our spare room - to you working out of your
established business. You do make a habit of posting truly moronic
assertions.

--
Bill Sloman, Sydney

Jamie

unread,
Nov 8, 2012, 6:48:40 PM11/8/12
to
A false image you mean.

Funny how you place yourself as the model to judge others. You seem to
think you're above it all when in fact, it is your kind that is causing
chaos.


Jamie



Bill Sloman

unread,
Nov 8, 2012, 8:11:31 PM11/8/12
to
On Nov 9, 10:42 am, Jamie
<jamie_ka1lpa_not_valid_after_ka1l...@charter.net> wrote:
> BillSlomanwrote:
> > On Nov 8, 12:12 pm, Jamie
> > <jamie_ka1lpa_not_valid_after_ka1l...@charter.net> wrote:
> >>BillSlomanwrote:
> >>>On Nov 7, 11:29 am, John Larkin <jlar...@highlandtechnology.com>
> >>>wrote:

<snip>

> >>>It's the contrast between ill-informed fantasies and objective
> >>>reality. The US electorate seems to have preferred the latter.
>
> >>   You'd make a perfect politician, never getting the facts right
>
> > Find the fact I've got wrong.

Jamie doesn't seem to have noticed that requirement.

> >>and always twisting the truth, if you really knew what the truth was.
>
> > I've probably got a better idea of what is true and what isn't than
> > you and John Larkin put together (and that's a depressing image).
>
> A false image you mean.

It's obviously false - jokes work that way.

>   Funny how you place yourself as the model to judge others. You seem to
> think you're above it all when in fact, it is your kind that is causing
> chaos.

You claimed to judge me by saying that "I never get the facts right",
but when challenged to find one fact that I've got wrong, you become
strangely reticent.

That makes you a liar, unfitted to judge anybody - though you have
managed to condemn yourself as an irresponsible bearer of false
witness. It's people like you who create chaos, by spreading
unsubstantiated claims, and failing to realise that they are libel-
mongering trouble-makers.

--
Bill Sloman, Sydney

Jamie

unread,
Nov 8, 2012, 9:01:30 PM11/8/12
to
The only claims I made are about your conduct and that requires no
proof or references to back up. You do well enough on your own exposing
yourself, no help is needed to make you publicly known as a
lunatic.

With out pulling names into this, others have confirmed that you can't
walk the talk, even though you try very hard to disillusion people of
that fact.

As for US politics? I would tell you this but I really can't with out
it being inaccurate, "Don't quick your day Job", actually I do believe
you may have forgotten what it really means to have a real Job. Sitting
back and being some one's pigtail advisor does not count.

Mushroom management comes to mind.

Jamie

Bill Sloman

unread,
Nov 8, 2012, 9:19:45 PM11/8/12
to
On Nov 9, 12:55 pm, Jamie
<snipped irrelevant comment>

>   Mushroom management comes to mind.

Jamie claims that "I never get the facts right" but can't come up with
a single fact that I've got wrong, and instead posts irrelevant
garbage. He thinks that we are the mushrooms who will put up with
being kept in the dark while being exposed to his daft delusions.

--
Bill Sloman, Sydney

Jamie

unread,
Nov 8, 2012, 9:48:59 PM11/8/12
to
what's the matter Slow_Man, am I hitting a little too close to home?

Hope you have your meds near by, it seems your blood pressure is elevated.

Jamie


Bill Sloman

unread,
Nov 9, 2012, 8:30:18 AM11/9/12
to
On Nov 9, 1:42 pm, Jamie
<jamie_ka1lpa_not_valid_after_ka1l...@charter.net> wrote:
> BillSlomanwrote:
>   what's the matter Slow_Man, am I hitting a little too close to home?

No, I'm just letting you demonstrate that you make claims that you
can't back up.

>   Hope you have your meds near by, it seems your blood pressure is elevated.

Find me a fact that I've got wrong, and I may faint in surprise, but
that's as much as you can hope to do to my blood pressure.

Contempt and disdain aren't emotions that raise the blood pressure.
Pointing out that you are an inept twit is just one of those
responsibilities that adults have to attend to from time to time.
Nobody whose opinion I value is going to take you seriously, but I do
need to remind lurkers that s.e.d isn't kind to the mentally
challenged.

--
Bill Sloman, Sydney
0 new messages