Great video. I've been meaning to get one of these for my 53131A, but didn't know they sold them on ebay. Free_Electron on the EEVblog forum has also done his own version.
Today I received my OPT-030 for my HP53181A!. Perfect!!! I took five minutes to assemble it on the frequency counter. It works great . Really good job!
Great video! Finally you're back :) Btw, you should get on TheAmpHour sometime, it would be interesting to get to know your background in electronics and such! Keep up the great work, thanks.
Late comment, but I consider these videos timeless: The disadvantage of the diodes: they will limit the db of the SA at around 0 db and will also introduce non-linearity at the limit
I like the fact that people are coming up with solutions to these "issues". Nothing against the manufacturers but they should understand at some point that there's a huge market that nobody caters to and that not everyone can afford to splurge on an upgrade.
It is shocking how much these companies markup the prices on the OEM parts for these devices. If you look at the BOM+ manufacturing costs of more specialized devices, on sites like electronics360.globalspec.com/teardowns/archive?page=1, you will often see that the closer you get to equipment used in the IT field or by engineers, the higher the pricemarkups get, and with the exception of devices like google glass and portable bluetooth speakers (where the fad has allowed for some truly astronomical price markups), the price markups on this equipment will often put just about any consumer electronic company to shame.
Razor2048 That's how things are unfortunately. Companies make item A for dirt cheap and sell it for 5 trucks worth of gold. Then you ask yourself, where the hell does that money go if it's not in the item i pay?!?!? As i see it, it's a circle jerk of advertisement, research, product, rinse and repeat. And considering the gargantuan prices that the advertisement world requires one can surmise it all goes in to brain washing us to buy more products so we can be brainwashed some more. (brainwashed used in the dark comedy sense). Meh, it's good that people come up with these mods. It bypasses the corner these companies have stuck themselves in to by propagating the same prices from way back when. Today producing a machine whatever might it be will never be as expensive across the entire field of it's existence from incipience to final product, as it was years, decades, ago. It just can't. We know more, we've come up with a s-load of methods of manufacture, testing has become obsolete (in a way) because of mass production and programs that allow us virtual interfacing etc. Rambling.
Yes, you are right. Junction capacitance of the diode appears in parallel to the input impedance of the amplifier and thus limiting the bandwidth. For this board, expensive diodes were used to ensure the bandwidth exceeds 3GHz.
Not only stray capacitance, but I'm guessing the recovery time of the diode matters too. I'm guessing that if the recovery time is too long, a high frequency signal that overloads the input will not just be clamped to the diode's turn-on voltage, but almost completely attenuated by the diodes.
I like older instruments where this type of thing was a possibility. Newer stuff usually makes this kind of "hack" more difficult if not impossible. Actually, in many cases, the hardware is already there, it's just waiting for a license key to unlock its functionality.
Correct me if I am wrong, but as I have seen in one of your other videos, although the diodes are great for clamping, they are not so great for THD as a diode is a non linear device. You will get 2nd, 3rd, 4th etc. etc harmonics introduced on what should have been a clean signal. Also I love your videos. This type of technical content is what I am looking for. You, Dave of EEVBLOG, W2AEW, Patrick Hood Daniel and Ben Eater are my favorite electronics/computer teachers.
Hey Shahriar! I really enjoy your videos! They are very informative and fun to watch! Will you be making more electronics tutorials type videos in the future?
I'm happy to see another video from you. And, as usual, a very thorough one. I am a little perplexed about counters in one respect: since the division of the observed freq is as much as 256, how does the counter successfully interpolate values that are between those division boundaries with good accuracy, especially odd values?
The division ratio is fixed, it does not change. So all the divider does is that in translates the input frequencies of 0.1 to 3GHz to a lower range, for examples, divides everything by 64 which then the instrument can count.
Superb video. At time index 22:17 (in the video) - if there is self-oscillation from about 2 GHz to 2.4 GHz, then does this mean that the static divider module is flawed? If no input signal is presented to the static divider, then it would be necessary to suppress self-oscillation, right? Otherwise the static divider module would always develop an undesired output, right?
I'm guessing if you're on the low side for the counter, if you drive it into clipping the 2nd (or 3rd, etc) harmonic might be what it ends up triggering on.
Awesome video, as always, Professor! I'm guessing that the diodes are a) unnecessary at each gain stage and b) due to their capacitance, which will be variable in voltage, would present undesirable impedance characteristics at the inputs and outputs of the amplifiers. The amplifiers also might not appreciate driving current through the diodes to ground. I hope to see more videos from you soon!
Very good video! Can you please do a video about OpAmps. I am actually ok with opamps and understand how they work, but every time a see new circuit it is still puzzling me, for example what role does the biasing voltage has in this video? Does it control the gain? And if the gain of each opamp is the same, why do you need separate Vbias here?....
How did you discover you were saturating the input at 43:45? Is there a trick or simply because of your huge experience? BTW: I suggest next time to take time and explain why you changed the reference level. I love your videos because you explain very well all your steps, in this case it was the first time you jumped too fast for common poor mortal like me with limited know how on a Spectrum Analyzer. I already researched and I know why now so don´t bother to explain me that. Just a little suggestion for your hard and unbelievable high quality well done videos. Yes it was very difficult to find an improvement (may be?) for your video :-)
Is there specific reason why there are four blocks of 7.2 dB each, can one have two block of 14.4 dB each. With OPT-030 i think frequency divider is fixed, when i look at 53132A component level specification, it is constantly set to divide by 128 by setting both SW1 and SW2 pin High(MB510) . specification also says they use MB510 , in this board we have MC12079 , i can not see very clear in video but i think both SW1 SW2 are set to high in your pcb too. but problem is with MB510 both control pin set to high will divide by 128 but , MC12079 both pin set to high will divide by 64. any help will be highly appreciated
Dear colleague, I have a frequency meter for this 53131A and I would like to know if you would have his electrical schematic or the service manual to send to me! I thank you!
Hi Shahriar, What diode does that board use for the input clamping? I am trying to select clamping diodes for a small LNA project I have been working on. I have come across the BAT54 and NXP 1PS10SB82 as possible options. Also, does a clamping diode arrangement like this also protect against ESD, not just a large signal? I have seen amplifier circuits with MOVs and GDTs even on the input and I'm wondering if I really need that level of protection for my board. Thanks!
Is it possible to add an additional amplifier or 2 to the circuit to further amplify the signal and thus allow it to work with weaker signals? What if someone were to make a design with 2 additional amplifiers, would it still function safely without harming the device? Also, can an amplifier be built for external use e.g., if you are working with a signal that is at -70 dBm, could a user build an external amplifier that will function inline with the probe and add some amplification, and thus expand the usable range that the device can function at?
The diodes at the output of the amplifier chain protect the divider from large signal swings when the last stage is saturated. So adding more stages won't harm the divider. However, at some point, the gain of the path would be too high and various noise sources will start to trigger the divider because they will be amplified. Furthermore, the isolation from the output of the amplifier to the input is limited, and you can actually cause an oscillation if not careful.
Is it bad to leave the oven on the oscillator for long periods, can the heater go bad from just being left on? What would good probing be? Calibrating the active probe first?
From what I've read, crystal aging (frequency change over time) is increased by temperature changes, which would suggest that to avoid aging, you want to leave your reference clock on all the time so that the temperature remains stable. Anyway it needs to be left on for hours (in the case of a high grade OCXO) to stabilise before usage. The heater is just a PID controlled resistor, so I wouldn't worry about it as a point of failure, especially as it's current will decrease as it reaches target temperature.
i did not understand all power supply stuff you said. If you are feeding power to RF amp with inductor (100nH-1uH), should you care about power trace characteristics in frequency ? I think not, but again, i don't understand RF stuff
You are correct in a sense that if I had an ideal inductor of very large value, I would in theory isolate the Vdd pin of the amplifier from the impedance variation of the Vdd trace. However, the inductor is not ideal and its value is not infinite. Therefore, it is possible to imagine a scenario where it could resonate with the impedance of the Vdd trace at a certain frequency and create some problems. Although, the series resistor to the inductor helps too as it reduces the quality factor of the inductor and thus dampening any resonances.
Great video, really enjoy the explanations of the circuit topology and design. I have a rubidium oscillator that I have built into an aluminium box. I made a video of it here: th-cam.com/video/37MtzcnvMks/w-d-xo.html You are welcome to borrow it if you have not already tracked one down. Not sure where you are but I am in Perth, Western Australia. Thanks for the really informative vids, Mike.
Great video. I've been meaning to get one of these for my 53131A, but didn't know they sold them on ebay. Free_Electron on the EEVblog forum has also done his own version.
Dave.. It would be great if you and Chris could get
Shahriar on to The Amp Hour?
Thanks Dave. Let us know once you upgrade your unit.
Today I received my OPT-030 for my HP53181A!. Perfect!!! I took five minutes to assemble it on the frequency counter. It works great . Really good job!
Another awesome video! My day always gets a lift seeing a new TheSignalPathBlog upload.
Cojotech in Romania makes prescaler options for 8GHz and 10GHz as well, and they use more modern components.
The Signal Path is the path to enlightenment. A+
Great video! Finally you're back :) Btw, you should get on TheAmpHour sometime, it would be interesting to get to know your background in electronics and such! Keep up the great work, thanks.
Late comment, but I consider these videos timeless: The disadvantage of the diodes: they will limit the db of the SA at around 0 db and will also introduce non-linearity at the limit
Finally another video!
One more coming up tomorrow! With a little surprise. :)
Thank you so much for the new content. On of the only RF based EE channels I can find on youtube
Nice, learnt a fair bit for sure and with my BEng EEE knowledge was able to keep up quite well. Thank you.
I like the fact that people are coming up with solutions to these "issues". Nothing against the manufacturers but they should understand at some point that there's a huge market that nobody caters to and that not everyone can afford to splurge on an upgrade.
It is shocking how much these companies markup the prices on the OEM parts for these devices. If you look at the BOM+ manufacturing costs of more specialized devices, on sites like electronics360.globalspec.com/teardowns/archive?page=1, you will often see that the closer you get to equipment used in the IT field or by engineers, the higher the pricemarkups get, and with the exception of devices like google glass and portable bluetooth speakers (where the fad has allowed for some truly astronomical price markups), the price markups on this equipment will often put just about any consumer electronic company to shame.
Razor2048 That's how things are unfortunately. Companies make item A for dirt cheap and sell it for 5 trucks worth of gold.
Then you ask yourself, where the hell does that money go if it's not in the item i pay?!?!?
As i see it, it's a circle jerk of advertisement, research, product, rinse and repeat. And considering the gargantuan prices that the advertisement world requires one can surmise it all goes in to brain washing us to buy more products so we can be brainwashed some more. (brainwashed used in the dark comedy sense).
Meh, it's good that people come up with these mods. It bypasses the corner these companies have stuck themselves in to by propagating the same prices from way back when.
Today producing a machine whatever might it be will never be as expensive across the entire field of it's existence from incipience to final product, as it was years, decades, ago. It just can't.
We know more, we've come up with a s-load of methods of manufacture, testing has become obsolete (in a way) because of mass production and programs that allow us virtual interfacing etc.
Rambling.
Yesssssss... Shahriar is back :) I missed you
Can't wait for tomorrow episode ... thanks for your videos.
Thumbs up. Good, clear explanation as always. Thanks.
5 years ago and we have a mortal that considers 3ghz “high” frequency! Fast forward and 3ghz is basically uninteresting DC on The Signal Path. 😁
Nice video!@11:15 parasitic capacitance would limit the bandwidth?
Yes, you are right. Junction capacitance of the diode appears in parallel to the input impedance of the amplifier and thus limiting the bandwidth. For this board, expensive diodes were used to ensure the bandwidth exceeds 3GHz.
Not only stray capacitance, but I'm guessing the recovery time of the diode matters too. I'm guessing that if the recovery time is too long, a high frequency signal that overloads the input will not just be clamped to the diode's turn-on voltage, but almost completely attenuated by the diodes.
you beat me to it.! ;)
I like older instruments where this type of thing was a possibility. Newer stuff usually makes this kind of "hack" more difficult if not impossible. Actually, in many cases, the hardware is already there, it's just waiting for a license key to unlock its functionality.
Correct me if I am wrong, but as I have seen in one of your other videos, although the diodes are great for clamping, they are not so great for THD as a diode is a non linear device. You will get 2nd, 3rd, 4th etc. etc harmonics introduced on what should have been a clean signal.
Also I love your videos. This type of technical content is what I am looking for.
You, Dave of EEVBLOG, W2AEW, Patrick Hood Daniel and Ben Eater are my favorite electronics/computer teachers.
Yeah, more videos, please. Great video. Thanks
Great video! The probem can be no linearity of the the input impedance?
Hey Shahriar! I really enjoy your videos! They are very informative and fun to watch! Will you be making more electronics tutorials type videos in the future?
Thank you. Yes, there will be more videos of course.
I'm happy to see another video from you. And, as usual, a very thorough one. I am a little perplexed about counters in one respect: since the division of the observed freq is as much as 256, how does the counter successfully interpolate values that are between those division boundaries with good accuracy, especially odd values?
The division ratio is fixed, it does not change. So all the divider does is that in translates the input frequencies of 0.1 to 3GHz to a lower range, for examples, divides everything by 64 which then the instrument can count.
Hey man thanks for the video been looking into PCB development this was a very informative video. thank you.
Superb video. At time index 22:17 (in the video) - if there is self-oscillation from about 2 GHz to 2.4 GHz, then does this mean that the static divider module is flawed? If no input signal is presented to the static divider, then it would be necessary to suppress self-oscillation, right? Otherwise the static divider module would always develop an undesired output, right?
Thanks I have like this frequency counter can I test 10mhz signal or not.
With such good quality, why do you not give us a link to the Polish seller?
I'm guessing if you're on the low side for the counter, if you drive it into clipping the 2nd (or 3rd, etc) harmonic might be what it ends up triggering on.
Old video, but still interesting! 👍
He is alive.
Awesome video, as always, Professor!
I'm guessing that the diodes are a) unnecessary at each gain stage and b) due to their capacitance, which will be variable in voltage, would present undesirable impedance characteristics at the inputs and outputs of the amplifiers. The amplifiers also might not appreciate driving current through the diodes to ground.
I hope to see more videos from you soon!
thank you for the excellent video!
Very good video! Can you please do a video about OpAmps. I am actually ok with opamps and understand how they work, but every time a see new circuit it is still puzzling me, for example what role does the biasing voltage has in this video?
Does it control the gain? And if the gain of each opamp is the same, why do you need separate Vbias here?....
Thanks for a great video
Would it be as good or even better to use a 10M DISCIPLINED GPSDO GPS clock OSCILLATOR instead?
How did you discover you were saturating the input at 43:45? Is there a trick or simply because of your huge experience? BTW: I suggest next time to take time and explain why you changed the reference level. I love your videos because you explain very well all your steps, in this case it was the first time you jumped too fast for common poor mortal like me with limited know how on a Spectrum Analyzer. I already researched and I know why now so don´t bother to explain me that. Just a little suggestion for your hard and unbelievable high quality well done videos. Yes it was very difficult to find an improvement (may be?) for your video :-)
How does the amplifier get its operating voltage? the top and bottom pins are ground, so does it use the DC bias as a power supply?
Is there specific reason why there are four blocks of 7.2 dB each, can one have two block of 14.4 dB each. With OPT-030 i think frequency divider is fixed, when i look at 53132A component level specification, it is constantly set to divide by 128 by setting both SW1 and SW2 pin High(MB510) . specification also says they use MB510 , in this board we have MC12079 , i can not see very clear in video but i think both SW1 SW2 are set to high in your pcb too. but problem is with MB510 both control pin set to high will divide by 128 but , MC12079 both pin set to high will divide by 64. any help will be highly appreciated
i have done a little search around i think chip on this board is not MC12079.
Dear colleague, I have a frequency meter for this 53131A and I would like to know if you would have his electrical schematic or the service manual to send to me! I thank you!
What are you using for shelf supports on your bench?
Referring to metal tubular supports from bench to first shelf.
IKEA Capita # 602.635.74 - find in IKEA’s Kitchen section of store.
4-5” (10-13cm) adjustable. Choice of chrome or white
Hi Shahriar,
What diode does that board use for the input clamping? I am trying to select clamping diodes for a small LNA project I have been working on. I have come across the BAT54 and NXP 1PS10SB82 as possible options. Also, does a clamping diode arrangement like this also protect against ESD, not just a large signal? I have seen amplifier circuits with MOVs and GDTs even on the input and I'm wondering if I really need that level of protection for my board. Thanks!
Is it possible to add an additional amplifier or 2 to the circuit to further amplify the signal and thus allow it to work with weaker signals? What if someone were to make a design with 2 additional amplifiers, would it still function safely without harming the device? Also, can an amplifier be built for external use e.g., if you are working with a signal that is at -70 dBm, could a user build an external amplifier that will function inline with the probe and add some amplification, and thus expand the usable range that the device can function at?
The diodes at the output of the amplifier chain protect the divider from large signal swings when the last stage is saturated. So adding more stages won't harm the divider. However, at some point, the gain of the path would be too high and various noise sources will start to trigger the divider because they will be amplified. Furthermore, the isolation from the output of the amplifier to the input is limited, and you can actually cause an oscillation if not careful.
I got 3 of these for free, would love to find a cheap 3ghz upgrade
Is it bad to leave the oven on the oscillator for long periods, can the heater go bad from just being left on? What would good probing be? Calibrating the active probe first?
From what I've read, crystal aging (frequency change over time) is increased by temperature changes, which would suggest that to avoid aging, you want to leave your reference clock on all the time so that the temperature remains stable. Anyway it needs to be left on for hours (in the case of a high grade OCXO) to stabilise before usage.
The heater is just a PID controlled resistor, so I wouldn't worry about it as a point of failure, especially as it's current will decrease as it reaches target temperature.
will it cause clipping?
i did not understand all power supply stuff you said. If you are feeding power to RF amp with inductor (100nH-1uH), should you care about power trace characteristics in frequency ? I think not, but again, i don't understand RF stuff
You are correct in a sense that if I had an ideal inductor of very large value, I would in theory isolate the Vdd pin of the amplifier from the impedance variation of the Vdd trace. However, the inductor is not ideal and its value is not infinite. Therefore, it is possible to imagine a scenario where it could resonate with the impedance of the Vdd trace at a certain frequency and create some problems. Although, the series resistor to the inductor helps too as it reduces the quality factor of the inductor and thus dampening any resonances.
@@Thesignalpath Thank you so much, I really appreciate your videos, comments and help!
Great video, thanks :-)
Someone made a 8GHz version. Search on eBay for:
HP AGILENT 53131 53132 53181 8G Simple Option Frequency Counter Input 100M-10GHz
What do you think?
Nice hacking, keep it up :)
Maybe you have some videos planned about precision DC gear too? :D
👍👍
Как много букв...
I have a 10mhz rubidium standard from ebay. I pay it u$s 30. it was from and old cell phone tower.
Great video, really enjoy the explanations of the circuit topology and design. I have a rubidium oscillator that I have built into an aluminium box. I made a video of it here:
th-cam.com/video/37MtzcnvMks/w-d-xo.html
You are welcome to borrow it if you have not already tracked one down. Not sure where you are but I am in Perth, Western Australia. Thanks for the really informative vids, Mike.