Re: 10 MHz reference inputs
> Given the counterfeiting, you can not blame manufacturers from protecting themselves.?
Perhaps not, but there are limits. For example, most people seem to regard FTDI's deliberate trashing of fake FTDI serial/usb chips as beyond the pale. While that is an extreme example, I'd put anti-repair strategies as on the route to it - and the 'right to repair' legislation now appearing in some markets indicates that it's at best irresponsible.
For me, much of the respect I have/had for HP, Tek, and their like is due to the quality and availability of their service information. I don't see Keysight engendering the same loyalty that most of us have to those brands. That will come back and bite them - it affects our commercial purchasing decisions.
?
|
Re: 10 MHz reference inputs
On Mon, 12 Sep 2022 at 09:15, Wilko Bulte < wkb@...> wrote: Well, HP put more money in R&D than the likes of Rigol etc ever did.
So it is more a sign of the times I'd say.
Wilko
Whereas HP used to provide service manuals with full schematics, the amount of service information in Keysight manuals is next to useless if you want to repair an instrument.?
In parts of Asia, counterfeiting is part of the culture. A friend bought a Rolex watch for a few pounds in a market in Malaysia. He knew it was a fake. If Rolex watches were being openly sold for a few pounds in a market in England, the police would be interested. But in some countries it is the norm.
Given the counterfeiting, you can not blame manufacturers from protecting themselves.?
Dave?
-- Dr. David Kirkby, Kirkby Microwave Ltd, drkirkby@...Telephone 01621-680100./ +44 1621 680100 Registered in England & Wales, company number 08914892. Registered office: Stokes Hall Lodge, Burnham Rd, Althorne, Chelmsford, Essex, CM3 6DT, United Kingdom
|
Re: 10 MHz reference inputs
Well, HP put more money in R&D than the likes of Rigol etc ever did.
So it is more a sign of the times I'd say.
Wilko
|
Hi Jeff, ? You can get magnetic bases for GPS units. Alternatively, if you have a bargeboard (facia) under your gutter, there’s a piece of TV hardware called a hockey stick, or J pole: ? Regards Rodger Bean ?
toggle quoted message
Show quoted text
From: [email protected] <[email protected]> On Behalf Of Jeff Green
Sent: Monday, 12 September 2022 11:11
To: [email protected]
Subject: Re: [Test Equipment Design & Construction] clock calibrator? OK don't laugh, ignorance calls for desperate measures. ? I fed the output of the GPS DO and the output of the UTC oscillator each through a 1K variable resistor with to my R2000 general coverage receiver. ? The two signals mixed and I think heterodyned. When I first applied power to the UTC the audio tone was maybe 100Hz, over an an hour the tone frequency dropped and became inaudible. ? The S-meter slowly cycled from a maximum of +20 to too low to read. ? I adjusted the trimmer resistor and the S-meter was cycling so slow I couldn't keep track of the seconds. ? I suspect the UTC will do just fine without locking it to the GPS DO. ? I don't need millihertz accuracy, yet. ? My problem is my radio room window is shaded by trees, and I have to mount the antenna and GPS DO in the living room and run a USB cable to my radio room. So I don't have easy GPS satellite visibility. ? We have a bonded metal roof and I can't drill a hole in it without voided the 30 year warranty. ? My wife is very supportive but draws the line at me running USB cables from one room, down the hall to my radio room. ? I'm looking into running the USB cable in the wall up to the ceiling and across the attic to my radio room. ? ? ? ?
|
Problem solved - That's exactly what a phase meter would have done for you.
John
toggle quoted message
Show quoted text
On 9/11/2022 6:11 PM, Jeff Green wrote: OK don't laugh, ignorance calls for desperate measures.
I fed the output of the GPS DO and the output of the UTC oscillator each through a 1K variable resistor with to my R2000 general coverage receiver.
The two signals mixed and I think heterodyned. When I first applied power to the UTC the audio tone was maybe 100Hz, over an an hour the tone frequency dropped and became inaudible.
The S-meter slowly cycled from a maximum of +20 to too low to read.
I adjusted the trimmer resistor and the S-meter was cycling so slow I couldn't keep track of the seconds.
I suspect the UTC will do just fine without locking it to the GPS DO.
I don't need millihertz accuracy, yet.
My problem is my radio room window is shaded by trees, and I have to mount the antenna and GPS DO in the living room and run a USB cable to my radio room.
So I don't have easy GPS satellite visibility.
We have a bonded metal roof and I can't drill a hole in it without voided the 30 year warranty.
My wife is very supportive but draws the line at me running USB cables from one room, down the hall to my radio room.
I'm looking into running the USB cable in the wall up to the ceiling and across the attic to my radio room.
_._,_._,_
------------------------------------------------------------------------
|
Looking at an eBay photo, I see the fastest sweep speed is 0.01 mS/Div. If 10 MHz were visible, it would show 100 cycles of waveform across the screen, not useful for seeing the waveform drift across the screen when compared to the trigger freq.
It should work to divide both signals by 50 or 100 to see 1 or 2 cycles on the screen.
Might be easier to just feed both signals into a mixer and low pass filter and look at the very low frequency difference signal.
John
toggle quoted message
Show quoted text
On 9/11/2022 3:45 PM, Jeff Green wrote: I was given 2 10:1 probes and I used my Leo
Bordnar "Mini Precision GPS Reference Clock."
Output signal is a square wave at 3.3V CMOS levels with 50 Ohms characteristic impedance. Output drive level can be adjusted.
Output power level (measured at 10MHz, fundamental power channel):
+10.3dBm, drive setting 32mA
+9.9dBm, drive setting 24mA
+9.0dBm, drive setting 16mA
+6.4dBm, drive setting 8mA
10.3dBm at 50 ohms is a bit more then 0.71V. I might not be providing enough signal to get any display.
I'm trying to locate a schematic for the Hitachi to see if there are any obvious filters on the V and H input.
Considering this unit was designed primarily for a portable analog TV waveform monitor, so it's possible it is bandwidth limited.
The fact it is battery powered is nice.
|
Unless that scope has a filter designed specifically to limit the
bandwidth, you should see a gradual rolloff past the specified
bandwidth.
If by chance you are using a 1x probe, that's your problem. Many
typical (~1m-long) 1x probes hit a wall around 5-7MHz. It's a
transmission-line limitation. Dual-mode probes might advertise
100MHz bandwidth, but that is only in 10:1 mode.
-- Cheers,
Tom
--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
On 9/11/2022 15:04, Jeff Green wrote:
toggle quoted message
Show quoted text
Re: "While
the scope response will start falling off at 6 MHz, most
scopes will have a usable response at 10 MHz, just not
accurate for voltage measurements If the trigger circuit
doesn't work at 10 MHz, X-Y mode, if available, should."
----------------------------------------------------------------------------------?
The
oscilloscope is a Hitachi Oscilloscope V-059B
?
?
Ebay says
7MHz but it's deaf at 7MHz. [Is it right to call a visual
display deaf?]
10MHz on the
V and/or H input don't produce any deflection.
?
This scope
was designed to be used as a portable ntsc waveform monitor
and is optimized for that function.
?
It's great
for audio and I'm learning about Lissajous Patterns with
computer generated audio tones.
?
I'm looking
into:
?
I'm
interested in ELF through perhaps 20MHz.
?
This unit
won't do Lissajous patterns but it is dual channel and the ham
who has one says you can trigger one channel with your
reference signal, display that channel and the other channel
will show frequency difference as the pattern driting.
|
On Sun, Sep 11, 2022 at 02:04 PM, Mark Jordan wrote:
As you already have a MCU in your circuit, it could be used to produce a voltage proportional to the frequency difference. An external DAC, R2R or even PWM can generate the voltage you need.
Hi, I think you're mixing up my project (OP) with someone else's question about using a difference freq to drive a meter. I already have a tiny OLED display which shows the difference in both numbers and graphically. Pete
|
As you already have a MCU in your circuit, it could be used to produce a voltage proportional to the frequency difference. An external DAC, R2R or even PWM can generate the voltage you need.
Regards,
Mark Jordan
|
As the original poster, perhaps now is the time to mention/remind that I built the "difference frequency" device because it was fun/interesting, NOT because I didn't have a good way to fine-tune an oscillator. I have an HP5345A counter, and 9 different o-scopes, with bandwidths from 200 KHz to 350 MHz.? :-)
Pete
|
While the scope response will start falling off at 6 MHz, most scopes will have a usable response at 10 MHz, just not accurate for voltage measurements If the trigger circuit doesn't work at 10 MHz, X-Y mode, if available, should.
John
toggle quoted message
Show quoted text
On 9/11/2022 8:39 AM, Jeff Green wrote: That of course suggests you have an oscilloscope with a bandwidth of 10MHz.
I've been given a Hitachi oscilloscope designed for ntsc analog that has a maximum frequency response of 6ish MHz.
I've got a line on a Tracor 527E that should do the job.
|
I keep my live GPSDOs feeding a two channel scope.? The sines tell me a lot more about the health of both systems than some other instrument, and we had a spare 2ch scope that nobody wanted to use.
That of course suggests you have an oscilloscope with a bandwidth of 10MHz.
I've been given a Hitachi oscilloscope designed for ntsc analog that has a maximum frequency response of 6ish MHz.
I've got a line on a Tracor 527E that should do the job.
?
-- K1FZY (WA4TPW) SK? 9/29/37-4/13/15
|
Dear: If there are not many pretensions, let's not forget that to calibrate two oscillators you can use any oscilloscope
in XY mode (I prefer the tube ones). Has it already been discussed? apologies if so, if not a simple solution and
available to everyone.
Saludos, Marcelo - LU2DR
|
Re: 10 MHz reference inputs
On Sun, 11 Sep 2022 at 14:45, Wilko Bulte < wkb@...> wrote: Well, fly-by-nights called Rigol etc sand/laser the partnumbers from the generic chips they use.
So it is not uncommon these days. I also hate this practice with a passion.
Wilko
If you put money into R&D, it doesn’t seem unreasonable to try to prevent others copying your designs. As much as I not like to try to fault-find equipment, I can understand why companies do it.? -- Dr. David Kirkby, Kirkby Microwave Ltd, drkirkby@...Telephone 01621-680100./ +44 1621 680100 Registered in England & Wales, company number 08914892. Registered office: Stokes Hall Lodge, Burnham Rd, Althorne, Chelmsford, Essex, CM3 6DT, United Kingdom
|
Re: 10 MHz reference inputs
Removing the markings from ICs is what the makers of CB noise gadgets
did, back in the day.
Why? Because (at the time) their $20 gadgets cost about 75 cents to
make, and were so dirt-simple that any CBer on the planet could
duplicate them and sell them for a dollar less.
Nowadays we have chip reader gadgets that will tell you what nearly
any TTL or CMOS chip is, so unless a manufacturer embeds microcode
that can't be extracted, almost anybody can reverse-engineer almost
anything.
Back in the day before back in the day, Don Lancaster (of "Cookbook"
fame) had a solution to this. He published everything, and released
it all into the public domain. He monetized his publications, because
once he published, his designs were unpatentable.
Plus, that was easier than sanding off the chip markings. FWIW,
nowadays Hans Summers sells a cute little digital transceiver, and I'm
going to sand the marking side of my finals flat for better heat
transfer. Please don't diss me for removing the markings. ;)
73
Jim N6OTQ
toggle quoted message
Show quoted text
On Sun, Sep 11, 2022 at 8:45 AM Wilko Bulte <wkb@...> wrote:
Well, fly-by-nights called Rigol etc sand/laser the partnumbers from the generic chips they use.
So it is not uncommon these days. I also hate this practice with a passion.
Wilko
|
Re: 10 MHz reference inputs
Well, fly-by-nights called Rigol etc sand/laser the partnumbers from the generic chips they use.
So it is not uncommon these days. I also hate this practice with a passion.
Wilko
|
Jeff, ? A good Frequency Difference Comparator is a Tracor 527E. ? Ross ?
toggle quoted message
Show quoted text
From: [email protected] < [email protected]> On Behalf Of Jeff Green Sent: Saturday, September 10, 2022 4:47 PM To: [email protected]Subject: Re: [Test Equipment Design & Construction] clock calibrator ? Lissajous figures were demonstrated to me today with a quick lesson on what they show and how you can measure phase and harmonic relationship. It is similar to the L/R phase display Adobe Audition gives you. ? I suspected the phase meter was too simple. I searched for frequency comparison using a 4046PLL but couldn't find anything. I'll look again Monday.
|
Re: 10 MHz reference inputs
Hi, ? Sanding Part Numbers off of ICs is a regular practice by many Car Alarm and some Car Audio vendors. ? Throw it away if traced to that part. ? ? McIntosh never did that but used House Part Numbers for Semiconductors and many other parts. Lots information available on McIntosh Parts, though. Worth fixing. ? ? Ross ?
toggle quoted message
Show quoted text
From: [email protected] < [email protected]> On Behalf Of Adrian Godwin Sent: Saturday, September 10, 2022 6:23 PM To: [email protected]Subject: Re: [Test Equipment Design & Construction] 10 MHz reference inputs ? I've never seen ics sanded on equipment from reputable manufacturers. It's always the marginal fly-by-nights. ? The IC numbers have been visible in every Extron DA that I've looked at. No reputable OEM removes common part numbers. Even if they did, there aren't that many ICs designed for this application. I remember the first IC designed for Video. It was introduced around 1988 and barely handled NTSC video. It was disappointing that the frequency response was so poor. I could have used a couple dozen at the TV station? where I was working in Engineering. On top of poor specs, I couldn't even buy a sample to test in our shop. The first video DAs that I worked on were early, discrete Grass Valley which were fully documented. All of the Extron DAs that I have came with a large clear label on the bottom with basic information on how to use them. ? Link to Extron manual:
There are plenty of these on Ebay.
For Michael A. Terrel, be aware that many US manufacturers sand the numbers off ICs and transistorls.
For Leon Robinson, given your opinion of politicians, which I agree with, you might like:
? [aka Lone star planet]
"Apparently, on New Texas, killing a politician wasn't regarded as?mallum in se, and was?mallum prohibitorum?only to the extent that what happened to the politician was in excess of what he deserved.?"
|
Re: 10 MHz reference inputs
The last that I saw was in the mid '80s when ON TV dumped all of their pay TV decoders on the surplus market. I was laughing as guys were buying them by the hundred at the Dayton hamfest. About two weeks later , they shut off all of their pay TV systems. This left all those greedy characters stuck with thousands of dollars worth of junk built with unmarked ICs. The only thing useful were the Ch 64 Vagi antennas and mounting kits
On Sat, Sep 10, 2022 at 8:23 PM Adrian Godwin < artgodwin@...> wrote: I've never seen ics sanded on equipment from reputable manufacturers. It's always the marginal fly-by-nights.
?
|
Re: 10 MHz reference inputs
I have about 30 different Extron DAs on hand, along with some Grass Valley and a half dozen Extron video routers.
On Sat, Sep 10, 2022 at 5:59 PM Vince Vielhaber < vev@...> wrote: I used an Extron dist amp for mine.? I had seen a youtube video of
someone making the conversion for his rubidium source, I just adapted it
to my GPSDO (the Russian one).
It's so far been working fine 24/7 for at least 2 years now.
Vince - K8ZW.
?
|
Dr. David Kirkby, Kirkby Microwave Ltd
John Kolb
saipan59 (Pete)
Marcelo - LU2DR
Michael A. Terrell