22:14 Those glitches in the waveform aren't a problem with DAC coefficients, that's simply a consequence of the DAC running at 115 kHz or whatever weird sample rate it's using, which your 10kHz requested frequency doesn't neatly divide. So they have to choose between having "leap samples" (analogously to leap seconds or leap years), or else they'd have the steps walking along the graph in a weird way.
14:53 I can’t figure out how they did their temperature conversions “30.2C / 62.2F” Edit: Oh! They literally just added 32 and forgot to multiply by 9/5. It became clear at 15:07
22:40, I think the imperfections of the modulation is caused by limited processing power of the MCU or the limited DAC sample rate (less likely). And the occasional glitches are interrupts of the MCU probably handling UIs or other periodic tasks.
Really thank you for this informative review. I backed this as a hunch on CrowdSupply and feel quite glad about it now. I think your comments are spot on and I hope that ERA take them on board with a firmware update. On the other hand, it's open source so any of us can have a go :) Looking forward to more videos on this unit, I am hoping to get back into RF after a long, long break.
Shahriar, as a professional electrical/computer engineer, I've appreciated the conceptual knowledge I've gained from your videos. As a hobbyist, I've enjoyed your teardown and review videos of affordable equipment, including this one. Just wanted to say thanks. Also, I've started to build a little home lab and I really like the shelving you have. If you could point me to where I could get something similar, I would appreciate it.
Great video, as ever! I was curious about the stuff made by ERAsynth. Their price seems very affordable and they come with a big promise. Your review has been very good for a decision making. Thank you very much and congratulations!
I was looking exactly for this Synth all over the World and see that it is close to me in the same country :) I will definitely use them in my projects
I am not really into RF things, but my first feeling is that if Dave would have reviewed this thing, he would probably ranting continuously about all the flaws and compromises 😉 while you accept that there are limitations but it isn't bad overall, considering how much more expensive a professional instrument is.
Judging from the fact that it is using a AtMega32U4 Processor, I'd say this is a design choice in the software to avoid floating point numbers. But it should be possible to show commas without having to process floats with some workarounds in the background. I'm doing a similar project btw, which goes all the way down to a few hundred Hz and uses LowPass-Filters to get rid of the higher harmonics. But I've been a bit low on time to work on it: www.eevblog.com/forum/oshw/open-source-hw-rf-signal-generator/
But if you add commas, you also need to add the option for dots (some countries use dots instead of commas for separators and commas where dots would go). And if you are doing that, then spaces should also be an option for the thousands separator. And that is how feature creep starts :) But since it is open source you can add that yourself (provided there is enough flash and RAM for all the features you want).
at 23:00 the AM modulation issues look like the DAC is not being updated for some periods, one might guess the CPU is outputting DAC samples and an IRQ is happening or similar to stall it.
Shielding: obvious, that was the first thing I noticed when you showed the inside. Also, is there any rf level feedback or is it just calibration data? That would totally depend on termination and could screw your measurements.
@28:40 I wonder if the analog audio could be routed via an op-amp to directly drive the attenuators to avoid the digitised noise, could still be level controlled by the processor.
Sometimes for serial links (in the same frequency range as the instrument) people use a single PLL with a reference clock @ 5MHz. Of course the phase noise performance will be much poorer (and will not be suitable for any decent RF work) but work perfectly fine for serial links. Saves a ton of on-chip area and power.
Adding to my previous comment (above), I can understand why the designers/engineers decided to implement just one set of outputs. The SMA connectors tend to take up some real estate on the board, and probably having four connectors would require a wider board. I've hit that brick wall myself, since a larger board with not that many more components (perhaps a couple more attenuators and other components to replicate the functionality of the first set of outputs). It is a hard decision to make. It is essentially choosing between designing a cheaper product that is more accessible, or a more complete product with an extra set of outputs that most users might never use. As a hobbyist designer, it is a bit discouraging having to remove functionality from a design in order to meet a target price.
That's an AWESOME little synth/generator !! And it's Open Source too !!! Too bad it doesn't go down to 100hz....... So you could do audio testing, or use it to sweep some common IF circuit's ! But...........Maybe there's a remedy, being that it's Open Source ! Guess I'm going to do some research, or should that be re-surf !! Anyways, it was a great review !!
An RF absorber, or some basic shielding would truly make this a fair bit better. And I agree that having that second output would also be a nice addition. Even if it would just be an output, with some buffering going straight to an SMA connector. (Modulation and attenuation would also be nice, but might be cost prohibitive at its price point.) Though, that it is open source doesn't really add much value as far as I am concerned. Since if a product has good specifications, performance and software for its price, then it is a good product. If the company listens to community/customer feedback and is willing to improve the product, then it offers similar advantages as an open sourced product. (Unless one is willing to go through the documentation and implement some edge case feature oneself.)
Very good review. I’m so impressed that I have ordered one. I will be using the sweep function a lot, and just wondering if there is a frequency readout whilst the sweep is taking place?
The Signal Path could you please tell how you can get so high end gear? Is there a way for the mortal of us to get a 8½-10½ DMM without having to eat dirt for 10 years in a row? :-)
From past videos what i get is that he works at bell labs in "chip design". I think he sometimes buys broken equipment and fixes it. Some of it at least.
Remember that rf prototyping kit you presented a year ago (microwave symposium 2018, 9. July 2018, on Patreon)? Still looking forward to see those in an experiment.
@@Thesignalpath This probably leftover from their crowdfunding campaign. "Order now" link leads to Crowdsupply, which now acts as a "store" for post-campaign purchases. It's quite common for manufacturers to rise the price after the initial rounds ends. Comments on the net quote even lower price, $179, what was probably the "early bird" price.
Electronic test gear which generates signals as part of its functionality, is exempt from emissions and susceptibility testing. On the other hand, this goes to show why it might actually be a good thing to test these things!
22:14 Those glitches in the waveform aren't a problem with DAC coefficients, that's simply a consequence of the DAC running at 115 kHz or whatever weird sample rate it's using, which your 10kHz requested frequency doesn't neatly divide. So they have to choose between having "leap samples" (analogously to leap seconds or leap years), or else they'd have the steps walking along the graph in a weird way.
14:53 I can’t figure out how they did their temperature conversions “30.2C / 62.2F”
Edit: Oh! They literally just added 32 and forgot to multiply by 9/5. It became clear at 15:07
Love your videos man! Learning lots every day. Thank you!
Very nice to see a review on a product that doesn’t require a mortgage or small loan to afford! Great review as always, keep em coming!
22:40, I think the imperfections of the modulation is caused by limited processing power of the MCU or the limited DAC sample rate (less likely). And the occasional glitches are interrupts of the MCU probably handling UIs or other periodic tasks.
Wait so I could actually afford something shown on the channel for once?
NO, crowdsupply says it's $249 so not really affordable after all.
@@stevec5000 200$ vs 10000k$ normally is hella affordable for the channel....
Really thank you for this informative review. I backed this as a hunch on CrowdSupply and feel quite glad about it now. I think your comments are spot on and I hope that ERA take them on board with a firmware update. On the other hand, it's open source so any of us can have a go :) Looking forward to more videos on this unit, I am hoping to get back into RF after a long, long break.
Shahriar, as a professional electrical/computer engineer, I've appreciated the conceptual knowledge I've gained from your videos. As a hobbyist, I've enjoyed your teardown and review videos of affordable equipment, including this one. Just wanted to say thanks.
Also, I've started to build a little home lab and I really like the shelving you have. If you could point me to where I could get something similar, I would appreciate it.
Nice to see a well designed product in a reasonable price range.
Great video, as ever! I was curious about the stuff made by ERAsynth. Their price seems very affordable and they come with a big promise. Your review has been very good for a decision making. Thank you very much and congratulations!
I was looking exactly for this Synth all over the World and see that it is close to me in the same country :) I will definitely use them in my projects
I enjoyed this video so much it hertz 🤓 I really appreciate this amazing level of detail and information. You are the RF Whisper. Much RESPECT ❤
I am not really into RF things, but my first feeling is that if Dave would have reviewed this thing, he would probably ranting continuously about all the flaws and compromises 😉 while you accept that there are limitations but it isn't bad overall, considering how much more expensive a professional instrument is.
It looks like it could use some commas in the rf frequency display, pretty hard trying to count the zeros.
Judging from the fact that it is using a AtMega32U4 Processor, I'd say this is a design choice in the software to avoid floating point numbers.
But it should be possible to show commas without having to process floats with some workarounds in the background.
I'm doing a similar project btw, which goes all the way down to a few hundred Hz and uses LowPass-Filters to get rid of the higher harmonics.
But I've been a bit low on time to work on it: www.eevblog.com/forum/oshw/open-source-hw-rf-signal-generator/
But if you add commas, you also need to add the option for dots (some countries use dots instead of commas for separators and commas where dots would go). And if you are doing that, then spaces should also be an option for the thousands separator. And that is how feature creep starts :)
But since it is open source you can add that yourself (provided there is enough flash and RAM for all the features you want).
Happy 4th! Thanks for another awesome video.
at 23:00 the AM modulation issues look like the DAC is not being updated for some periods, one might guess the CPU is outputting DAC samples and an IRQ is happening or similar to stall it.
So informative and interesting, thank you !
Shielding: obvious, that was the first thing I noticed when you showed the inside. Also, is there any rf level feedback or is it just calibration data? That would totally depend on termination and could screw your measurements.
@28:40 I wonder if the analog audio could be routed via an op-amp to directly drive the attenuators to avoid the digitised noise, could still be level controlled by the processor.
Sometimes for serial links (in the same frequency range as the instrument) people use a single PLL with a reference clock @ 5MHz. Of course the phase noise performance will be much poorer (and will not be suitable for any decent RF work) but work perfectly fine for serial links. Saves a ton of on-chip area and power.
Nice video, as always! This seems to be a nice piece of equipment, and it is good to see a crowdfunded project such as this being released.
Adding to my previous comment (above), I can understand why the designers/engineers decided to implement just one set of outputs. The SMA connectors tend to take up some real estate on the board, and probably having four connectors would require a wider board. I've hit that brick wall myself, since a larger board with not that many more components (perhaps a couple more attenuators and other components to replicate the functionality of the first set of outputs).
It is a hard decision to make. It is essentially choosing between designing a cheaper product that is more accessible, or a more complete product with an extra set of outputs that most users might never use. As a hobbyist designer, it is a bit discouraging having to remove functionality from a design in order to meet a target price.
That's an AWESOME little synth/generator !! And it's Open Source too !!! Too bad it doesn't go down to 100hz....... So you could do audio testing, or use it to sweep some common IF circuit's ! But...........Maybe there's a remedy, being that it's Open Source ! Guess I'm going to do some research, or should that be re-surf !! Anyways, it was a great review !!
Please please please review the N1201SA VNA that is less than 120USD with a screen and built-in battery!
I have one of those, what do you want to know about it?
When I need a cheap modulated generator for the frequencies Shahriar - Bell usually work at, I use an incandescent bulb - with a dimmer.
:-)
There are definitely applications that benefit from using the arduino plattform, but RF synthesisers aren't one of them.
An RF absorber, or some basic shielding would truly make this a fair bit better.
And I agree that having that second output would also be a nice addition. Even if it would just be an output, with some buffering going straight to an SMA connector. (Modulation and attenuation would also be nice, but might be cost prohibitive at its price point.)
Though, that it is open source doesn't really add much value as far as I am concerned.
Since if a product has good specifications, performance and software for its price, then it is a good product.
If the company listens to community/customer feedback and is willing to improve the product, then it offers similar advantages as an open sourced product. (Unless one is willing to go through the documentation and implement some edge case feature oneself.)
Is this still available for purchase? I can't find it online on any shop
You should do the HackRF one! It’s really popular among the ham radio community, but I haven’t seen any good review of the circuitry.
Very good review. I’m so impressed that I have ordered one. I will be using the sweep function a lot, and just wondering if there is a frequency readout whilst the sweep is taking place?
The Signal Path could you please tell how you can get so high end gear? Is there a way for the mortal of us to get a 8½-10½ DMM without having to eat dirt for 10 years in a row? :-)
From past videos what i get is that he works at bell labs in "chip design".
I think he sometimes buys broken equipment and fixes it. Some of it at least.
Remember that rf prototyping kit you presented a year ago (microwave symposium 2018, 9. July 2018, on Patreon)? Still looking forward to see those in an experiment.
TSP125, 21:20
@@adrianschneider4441 I'll do it. It is on my list.
Excellent as usual!
Please review the ERA++
Can i use it as local oscillator for dvb-t receiver please
nextion displays really are awesome
On the 1KHz FM modulation, I wonder if it is the ADC sample rate is limited to such a low rate because the frequency shifting cannot be done quickly.
at 28:46 One of the electronics god spoke and said the modulation rate shall be 100Hz, so the measuring device instantly corrected its mistake :o
6.4GHz no shielding.
Looks like you might want to design your own low cost product... may I suggest a cheap scope...
$199?! On croudsupply it's listed as $249 + shipping (free for US / $25 anywhere else).
Interesting, on their site it shows up as $199. (erainstruments.com/)
@@Thesignalpath This probably leftover from their crowdfunding campaign. "Order now" link leads to Crowdsupply, which now acts as a "store" for post-campaign purchases. It's quite common for manufacturers to rise the price after the initial rounds ends. Comments on the net quote even lower price, $179, what was probably the "early bird" price.
Great vido!
But where du you geh your equipment from?
You must be super rich :D
i think the number of your's subscribers is the maximum number of people who can afford what you have on your bench :))
So no way to modulate phase I assume? Would be cool to transmit QPSK or something, as a test signal. But I guess it's just a synthesizer not an SDR.
☹️. Still waiting for an equivalent quadrature source.
VSG from Signal Hound?
👍👍
eating ramen with an egg on top...*some which cost a million dollars*...thisisfine.jpg
I guess they didn't bother with any of that CE "nonsense".
Electronic test gear which generates signals as part of its functionality, is exempt from emissions and susceptibility testing. On the other hand, this goes to show why it might actually be a good thing to test these things!
türk malı MADE IN TURKİYE
speak slowly please