Re: [time-nuts] Thunderbolt - any negatives ?
Didier
Your comments regarding SMPS are very valid, but there are SMPS designs and
then some !
I bought a Wavetek 395 Function Generator with a totally blown up SMPS for
approx $100. I attempted to repair this unit but the PCB was badly
damaged/burned and one 8-pin Dil device was blown to bits - plus burnt out
resistors. I tried to obtain a replacement from the manufacturers without
success. After hooking up external supplies and proving that it was
otherwise OK, I contemplated building a linear supply within the box - it
would have been heavy and difficult to fit. This latter point reinforces
your comment about size and power dissipation. The original unit was
identical to the average small PC PSU, but with very different outputs.
After about 10 months of looking for a solution - I thought I recognized a
Wavetek 395 Function Generator in Bob Mokia's Lab. photo, and ask him if he
could find me a spare PSU, after a short while he came back with a
replacement for around $50 - presto I now have a fully working 395 for $100,
which I notice sells for $1500 in the US.
Footnote:
No axe to grind, but like others in the Group,I have found Bob Mokia to be a
"straight" and helpful dealer.
Roy
----- Original Message -----
From: "Didier Juges" didier@cox.net
To: "'Discussion of precise time and frequency measurement'"
time-nuts@febo.com
Sent: Tuesday, June 02, 2009 3:24 AM
Subject: Re: [time-nuts] Thunderbolt - any negatives ?
Most likely failures on power supplies are with the power components.
Failure of the pass transistor in a linear supply is likely to result in
overvoltage at the output, while failure of the switch on a switchmode
supply will blow the fuse instantly.
It is been my experience (after 30 years in the field) that a properly
designed switchmode supply is at least as reliable as a linear supply of the
same output power, if for no other reason than the lower dissipation and
resulting reduced failure rate.
By using integrated controllers with lots of protection features built-in,
switchmode supplies tend to be smarter than linear ones, and their failures
tend to cause fewer damage to other circuits.
Of course, your mileage may vary...
Didier KO4BB
-----Original Message-----
From: time-nuts-bounces@febo.com
[mailto:time-nuts-bounces@febo.com] On Behalf Of Robert Atkinson
Sent: Monday, June 01, 2009 2:46 PM
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] Thunderbolt - any negatives ?
Most switch mode power supplies actually run a voltage
doubler on the input when running on 110V. This puts over
300V across the transformer and switch. Also the regulation
loop crosses the isolation barrier introducing more failure
points that can result in overvoltage.
Robert G8RPI.
--- On Mon, 1/6/09, SAIDJACK@aol.com SAIDJACK@aol.com wrote:
From: SAIDJACK@aol.com SAIDJACK@aol.com
Subject: Re: [time-nuts] Thunderbolt - any negatives ?
To: time-nuts@febo.com
Date: Monday, 1 June, 2009, 6:17 PM
Hi there,
A switcher has much more stresses on the components, since
it usually
switches the primary side rectified 110/220V high-voltage across a
transformer.
Thus the switching FET has to be very high voltage capable (about
~170V DC in the US), and the second component under stress is the
primary high voltage capacitor, because it sees a very fast AC
switching current on it (current draw is on when the FET is on, and
off when the Fet is off). Also there has to be a fast
snubber network
to prevent the back-emf from destroying the primary Fet with
over-voltage.
A linear supply has none of these fast current/voltage
transients on
it, only a couple of diodes switching the 60Hz secondary onto a
capacitor at low voltage.
A secondary concern is thermally induced stress, switchers will
usually be packed into a very small enclosure with very high power
capability/density.
This is not possible for linear supplies, since the
transformer size
will usually determine overall sizing. Compare a Laptop
power supply
size (usually these have between 40W and 90W rating!) to a similar
rated linear supply.
bye,
Said
In a message dated 6/1/2009 09:48:29 Pacific Daylight Time,
hmurray@megapathdsl.net
writes:
Is there something I don't understand in this area? What makes a
linear supply more reliable than a switcher?
My first guess would be a switcher would be more reliable
because it
would run cooler.
That's probably assuming the same amount of design effort which is
probably not a valid assumption if I'm comparing a brand-X linear
with a brand-Z switcher. A quick glance at the general
construction
might give a better answer.
time-nuts mailing list -- time-nuts@febo.com To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com To unsubscribe,
go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
You could always transform this from a hardware problem to a software
problem. Take the output of your GPSDO, divide it down to somewhere
inside the audio band, feed it to a spare input on your USB sound
card and have software track this reference and correct the received
signal.
JDB.
LART. 250 MIPS under one Watt. Free hardware design files.
http://www.lartmaker.nl/
Lux, James P skrev:
-----Original Message-----
From: time-nuts-bounces@febo.com
[mailto:time-nuts-bounces@febo.com] On Behalf Of Rex Moncur
Sent: Monday, June 01, 2009 3:00 PM
To: 'Discussion of precise time and frequency measurement'
Subject: [time-nuts] Sound Cards for locking to GPSDO 10 MHz
references
Hi all
Does anyone have any experience of locking a USB external
soundcard to a GPSDO 10 MHz reference.
I am interested in advice on any good quality soundcards that
can be readily locked to either 10 MHz or if necessary to
some other frequency that we can derive from a GPSDO source.
I have done some tests with the SignalLink soundcard that
uses a Texas Instruments PCM2904 chip and requires a 12 MHz
lock frequency. This requires some cutting of tracks to
remove the internal oscillator feedback and insert the
locking frequency. 12 MHz is readily derived from 10 MHz but
I have not been able to get it to lock. The Texas
instruments data sheet suggests that it is possible to use an
external refernce but also says this is not recommended.
With this expereicne I would rather find a sound card that is
designed for external locking that does not require the
cutting of tracks.
For info the purpose of this request is that we are looking
at using very narrow bandwidth modes at less than 1 mHz for
light wave communcation. To date using LEDs and cloud
reflection we have worked over 200 km with WSJT but we should
be able to do 20 dB better if we can get down to milli-Hz
bandwidths (at the expense of spending all night to complete
a QSO). Our expereince to date is that standard sound cards
are just not stable to better than 5 milli-Hz at 1000 Hz
which should be readily solved by GPS locking let us get down
to sub milli-Hz levels.
Rex VK7MO
Some of the "pro" sound interfaces have a "word clock" input.
There are a variety of things that take a external input and generate a S/PDIF that's properly timed, as well. Lots of boxes will take a S/PDIF sync input (e.g. the Edirol FA-66 which was used by lots of Flex-Radio folk), so maybe that's something you could easily generate from your 10MHz.
A chart at Cakewalk shows that MOTU has a USB interface (828MkII) which has a word clock sync. It's going to be a pricey beast though, with 8in/8out ($800?)
Even if you have a word clock input, you're going to have to
synthesize that from the 10 MHz. Maybe it's easier to just
make a S/PDIF which is a MUCH more common sync signal. ( I
think S/PDIF is something like 3 MHz)
S/P-DIF [iec60958-3] has a baudrate which is 128 x sample rate and a bit
rate which is 64 x sample rate, which is inherited properties from
AES/EBU [aes3] [tech3250] [iec60958-4].
Locking up a S/P-DIF (128 x sample rate) is about the same job as
locking up a superclock (256 x sample rate) or wordclock (1 x sample rate).
As long as the signal is samples with low jitter and A/D converted in a
good fashion, delivery over S/P-DIF should not be too hard. An ADC is
slammed onto a AES/EBU/S/P-DIF chip which is fairly trivial extra work.
Cheers,
Magnus
-----Original Message-----
From: time-nuts-bounces@febo.com
[mailto:time-nuts-bounces@febo.com] On Behalf Of Magnus Danielson
Sent: Tuesday, June 02, 2009 10:08 AM
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] Sound Cards for locking to GPSDO 10
MHz references
Some of the "pro" sound interfaces have a "word clock" input.
There are a variety of things that take a external input
and generate a S/PDIF that's properly timed, as well. Lots of
boxes will take a S/PDIF sync input (e.g. the Edirol FA-66
which was used by lots of Flex-Radio folk), so maybe that's
something you could easily generate from your 10MHz.
A chart at Cakewalk shows that MOTU has a USB interface (828MkII)
which has a word clock sync. It's going to be a pricey
beast though,
with 8in/8out ($800?)
Even if you have a word clock input, you're going to have to
synthesize that from the 10 MHz. Maybe it's easier to just make a
S/PDIF which is a MUCH more common sync signal. ( I think S/PDIF is
something like 3 MHz)
S/P-DIF [iec60958-3] has a baudrate which is 128 x sample
rate and a bit rate which is 64 x sample rate, which is
inherited properties from AES/EBU [aes3] [tech3250] [iec60958-4].
Locking up a S/P-DIF (128 x sample rate) is about the same
job as locking up a superclock (256 x sample rate) or
wordclock (1 x sample rate).
However, if you're buying an off the shelf audio interface, you're stuck with whatever the mfr is providing for a sync input, and a (very) casual inspection of what's available these days (particularly at low cost) shows that S/PDIF seems to be the most common.
J.D. Bakker wrote:
You could always transform this from a hardware problem to a software
problem. Take the output of your GPSDO, divide it down to somewhere
inside the audio band, feed it to a spare input on your USB sound
card and have software track this reference and correct the received
signal.
JDB.
I am in complete agreement with this kind of solution. Sampling on the
second channel a reference signal of known value allows the software
to make a simple adjustment. No need to switch on the soldering iron...
Never do in hardware what can be done in software.... :-)
73 Alberto I2PHD
Hello The Net:
For portable operations with a laptop, usually only one input channel is
available
and it is at mike (not line) level.
The alternative to sum the analog reference and the analog signal of
interest may be
possible if the reference noise can be kept out of the signal of
interest bandwidth.
Maybe a external USB soundcard with at least 2 input channels is more
appropriate.
Stan, W1LE FN41sr Cape Cod
Alberto di Bene wrote:
J.D. Bakker wrote:
You could always transform this from a hardware problem to a software
problem. Take the output of your GPSDO, divide it down to somewhere
inside the audio band, feed it to a spare input on your USB sound
card and have software track this reference and correct the received
signal.
JDB.
I am in complete agreement with this kind of solution. Sampling on the
second channel a reference signal of known value allows the software
to make a simple adjustment. No need to switch on the soldering iron...
Never do in hardware what can be done in software.... :-)
73 Alberto I2PHD
Lux, James P skrev:
-----Original Message-----
From: time-nuts-bounces@febo.com
[mailto:time-nuts-bounces@febo.com] On Behalf Of Magnus Danielson
Sent: Tuesday, June 02, 2009 10:08 AM
To: Discussion of precise time and frequency measurement
Subject: Re: [time-nuts] Sound Cards for locking to GPSDO 10
MHz references
Some of the "pro" sound interfaces have a "word clock" input.
There are a variety of things that take a external input
and generate a S/PDIF that's properly timed, as well. Lots of
boxes will take a S/PDIF sync input (e.g. the Edirol FA-66
which was used by lots of Flex-Radio folk), so maybe that's
something you could easily generate from your 10MHz.
A chart at Cakewalk shows that MOTU has a USB interface (828MkII)
which has a word clock sync. It's going to be a pricey
beast though,
with 8in/8out ($800?)
Even if you have a word clock input, you're going to have to
synthesize that from the 10 MHz. Maybe it's easier to just make a
S/PDIF which is a MUCH more common sync signal. ( I think S/PDIF is
something like 3 MHz)
S/P-DIF [iec60958-3] has a baudrate which is 128 x sample
rate and a bit rate which is 64 x sample rate, which is
inherited properties from AES/EBU [aes3] [tech3250] [iec60958-4].
Locking up a S/P-DIF (128 x sample rate) is about the same
job as locking up a superclock (256 x sample rate) or
wordclock (1 x sample rate).
However, if you're buying an off the shelf audio interface, you're
stuck with whatever the mfr is providing for a sync input, and a
(very) casual inspection of what's available these days
(particularly at low cost) shows that S/PDIF seems to be the most common.
Do they really lock up to the S/P-DIF input? I doubt it for the cheap
boards. Rather, they decode the S/P-DIF signal and ship the samples into
the DSP. The DSP tends to make very rought sample-rate conversions like
dropping samples etc.
A lockable board isn't that expensive. You can get them off ebay for
instance.
Cheers,
Magnus
Alberto di Bene skrev:
J.D. Bakker wrote:
You could always transform this from a hardware problem to a software
problem. Take the output of your GPSDO, divide it down to somewhere
inside the audio band, feed it to a spare input on your USB sound card
and have software track this reference and correct the received signal.
JDB.
I am in complete agreement with this kind of solution. Sampling on the
second channel a reference signal of known value allows the software
to make a simple adjustment.
Such a double-frequency conversion cancels fairly well the transfer
oscillators frequency and jitter, as long as it is sufficienly low.
No need to switch on the soldering iron...
Never do in hardware what can be done in software.... :-)
Respectfully I disagree. There are tasks which is better managed by
software and tasks which is better managed by hardware. In the world of
FPGAs, it is also worth mentioning that some tasks is best done there.
The big trick is to find a balance between various methods, available
resources, partitioning of the problem, doing it on time and achieving
the needed performance.
There is an overbeleif in what software is suitable for IMHO.
Cheers,
Magnus
Magnus Danielson wrote:
No need to switch on the soldering iron...
Never do in hardware what can be done in software.... :-)
Respectfully I disagree. There are tasks which is better managed by
software and tasks which is better managed by hardware. In the world of
FPGAs, it is also worth mentioning that some tasks is best done there.
The big trick is to find a balance between various methods, available
resources, partitioning of the problem, doing it on time and achieving
the needed performance.
Of course you are right, the best solution must be decided case by case.
But the biggest plus of the software is that it can be changed on the fly,
without an expensive reworking station, and the manual ability to correctly
use it. And a side effect is speed : you can test many variants of a solution
in a time frame of a few minutes. Not so easily doable with hardware changes.
73 Alberto i2PHD
Alberto di Bene skrev:
Magnus Danielson wrote:
No need to switch on the soldering iron...
Never do in hardware what can be done in software.... :-)
Respectfully I disagree. There are tasks which is better managed by
software and tasks which is better managed by hardware. In the world
of FPGAs, it is also worth mentioning that some tasks is best done there.
The big trick is to find a balance between various methods, available
resources, partitioning of the problem, doing it on time and achieving
the needed performance.
Of course you are right, the best solution must be decided case by case.
But the biggest plus of the software is that it can be changed on the fly,
without an expensive reworking station, and the manual ability to correctly
use it. And a side effect is speed : you can test many variants of a
solution
in a time frame of a few minutes. Not so easily doable with hardware
changes.
This is why we do alot of things in FPGAs today, and in the FPGAs we
often put dedicated DSPs of various complexity, often adapted to their
task. Keeping quick turn-around is on our mind, but in general, the
shorter turn-around, the poorer testing usually happends, and the
sloopier design is often found, and the longer it takes to get the job done.
In general, a CPU is suitable for doing non-common tasks. More dedicated
designs like firmware and hardware is suitable to do things which is
essentially the same but happends over and over and over and often at a
high speed. Such monotonic tasks just waste energy, space and
complexity when done in CPUs. The problem with a generic CPU is that it
is generic, so it can do all kinds of tasks, which makes timing-critical
bulk-processing tasks problematic to combine with sporadic and possibly
high-dynamic processing. Splice the bulk off to some dedicated
processing, which can be done in another CPU, and better performance is
yielded. There are loads of designs where a few well thought 8-bit
processors work together and shine over a more modern fancy design.
One such example is found in the SR-620 which has a Zilog Z-8000
processor as main CPU and a Z-80 co-processor which only does the X-Y
vector display. The Z-80 has so small program that it is loaded into
SRAM from the Z-8000 as it boots.
The HP 5334A has actually 3 different 3870 processor, one for overall
control, one for measurements and one for GPIB.
Hmm, do you get a feeling that I am actually object very much to just
toss it into the processor. I think you are right. :)
Nothing wrong with software, but use it wisely. Build the test-benches
as if you where doing a ASIC or full-custom design and thus also think
about each compile costing you milions of dollars and a pipe-line depth
of many months (6-9).
I guess I am becomming more conservative by the day. From my own and
others mistakes and succsesses.
Cheers,
Magnus