Rigol Function Generators - experience?

I do have a analog function generator made by B+K Precision but without counter that I recall. Model 4040A, bought in 2003. That generator was unusable, and so has led an isolated, dusty existence for decades.

I always had bad experiences with B+K Precision <anything> back in the day, and stopped buying from them. Perhaps they have since improved.

Joe Gwinn

I have this thingi, too. The look&feel is a little bit strange.

But it is easy to load a waveform from your computer from USB-Stick from CSV-File without installing fat software.

And it is also nice to made a DG4162 from it. :-)

Olaf

I have a 4003A on my bench. It's been great.

I got the 25MHz Dual Channel DDS Function Signal Generator for US $61.63.

eBay item number: 111678169556

I don't know if it is battery powered, but any place you need a function gen is probably line powered. $150 for the scope and $61 for the sig gen is just over $210. I have plenty of DMMs.

The output stage on the sig gen is weak. There is a simple mod to replace it with a better one but I lost the link.

Another sweet deal is the FA-2 1Hz-6GHz Frequency Counter, US $110.96. It actually delivers 15 digits and is accurate to the last digit when locked to GPS. Keysight eat your heart out. eBay item number: 113854113701. All gone but available on Aliexpress.

Lots of other deals available. BG7TBL 1 Hz to 8 GHz sig gen, RSP1A 1kHz -

2000Mhz SDR Receiver that can also work as a spectrum analyzer, 50 KHz to 3 GHz VNA, 8 digit voltmeter, and so on. All inexpensive, should be on everyone's workbench.

Skip all the boatanchors that take up too much space, weigh too much, and are impossible to repair. Take advantage of modern ics and simple designs.

Now is a great time for anyone interested in electronics!

I have a old 15MHz HP33120A. At one point I needed a 50MHz generator, and bought the Feeltech FY6600:

It's great. Dual channel, and in some ways better than the HP

Only problem is that it is noisy, so I was testing wireless power transfer between two coils, and measurements didn't make sense. It turned out that the CM noise from the generator is quite high, so if you buy it, add a lot of clamp on cores

Other tip, it is very light, and the buttons are hard to press, so place it under some heavy gear so it stays in place :-)

The Keysight 33612A (two 80 MHz channels) is equivalent to the DG1062Z (two 60 MHz channels), but far nicer I'll grant. And it is literally ten times the price, so I'm getting the Rigol.

I looked into Keysight's TrueForm waveform generation technology. They mentioned that it was patented, so I went looking for the patent, if only because the explanations in the brochures et al were baffling. The explanation is in the "Arbitrary Waveform - A high performance AWG primer - Fundamentals of Generation" Reference Guide, 2015, Manual Part Number M8190-91050, Edition 4.0, June 2015, Keysight Germany.

This document cites "US Patent 6,812,878 B1, Jewett et al. Per-Element Resampling for a Digital- to-Analog Converter", which will expire in April 2023. To get a copy, go to Google advanced patent search, enter US6812878 in the big box, then download the pdf.

It won't be long before there is a Chinese clone of the 33612A. Probably already done, but not sold in the US until that patent expires.

Joe Gwinn

[...]

Wow! 30 MHz is CA$125.30. Double the frequency to 60 MHz and the price only increases to CA$135.57. That is a bargain.

The manual is at

I want one!

Now is a great time to be alive for electronishers.

The price trend seems to be across the board. I used to play in a dance band on the weekends while I was stationed in Europe with the RCAF. We weren't very good, but it didn't matter. Everyone was blitzed by the time we arrived and they would dance to anything.

In those days, a Fender Stratocaster would have cost a fortune if it were available. But over the years the price has dropped, and now it is a very affordable CAD $299.99 on Amazon:

It is a real strat. Made in Indonesia but excellent quality. Very good therapy after spending a day flogging electrons.

There is a streaming site that plays 1930's jazz that is nice to play along with:

http://195.176.247.102:8000/classical.mp3

I skimmed the patent, but I don't understand what problem it solves. How do those timing glitches make it through the reconstruction filter? Seems like they would settle almost instantly (and the background section says as much, near line 30 on page 6.)

It's definitely not the sort of patent that would block competitors at this point in time. There are a lot of patents like this, written to work around shortcomings in ADCs and DACs of 20 years ago that no longer exist in modern ICs.

-- john, KE5FX

The secret source is in the Encoder block (number 102), which is completely unconstrained in US 6,812,878. The reference guide gives some idea of what's being done.

Often the various details are scattered about in multiple patents, which are usually most easily found by searching what else the named inventors have done. Jewett seems to be the main name, and may have published an article in the International Solid-State Circuits Conference.

And in any case, the above may be our best source to understand the principles of operation.

Yes, but the point here is to get around the limitations of cheap DACs, on the trailing edge of technology, not the current leading edge. Texas Instruments has some amazing DACs and ADCs, which cost thousands of dollars per chip.

Joe Gwinn

Interesting hack that one.

All our 350 MHz Rigol scopes are actually 500 MHz. We just asked.

I have a BK triple power supply on a tertiary bench- it's just a rebadged supply made by one of the major Chinese (Shenzhen) power supply firms, also sold under their own name, Velleman etc.

More of a consumer product construction than an instrument, but solid enough.

Yes, there's that.

But my answer to American firms selling the lower end of Chinese stuff is: If I'm getting Chinese quality, I want Chinese price too.

So, I do buy Harbor Freight stuff, but with low expectations.

I've had better luck buying Chinese brands that are selling in their own name, who are clearly are trying to be world class. Rigol being one example.

Joe Gwinn

Quality is only a small part of the price equation. Chinese sellers do not have to provide any statutory (mandatory) warranties, are not tested to any performance standards, and cannot easily be sued if the device fails to meet specifications, or even if it blows up and kills someone.

Most "American" products are just Chinese-made devices with proper testing, warranty and liability protection. Adding those things (without any change in the design or manufacture) approximately triples the cost of doing business.

Harbour Fright provides just the minimum of these protections, which is why you pay a little less.

CH

Huh? How is triple "a little less"?

More to the point, quality has nothing whatsoever to do with "proper testing, warranty and liability protection" - all these do is to ensure that some government mandated requirements are met, which may or may not be a good idea, but is unrelated to how well a product works for its intended purpose, and how long it continues to do so.

One cannot test quality into something, quality must have been designed and built into the product from the beginning.

Nor should "proper testing, warranty and liability protection" triple the cost.

Joe Gwinn

That's a three year old model! Feeltech / FeelElec have since done an FY6800 and, currently an FY6900 model which looks a better bet. The latest 6900 model revision now uses a single 5v rail smpsu board with the

+/- 13.7v rails generated on the main board making the option to replace the source of the high levels of CM switching noise with a small analogue PSU board a whole lot easier.

Unless you want to upgrade its cheap 10MHz smd XO chip to an OCXO, you only need 5v at 2A to power the main board.

A comprehensive source of info on these dirt cheap AWGs can be found in the EEVBlog forum.

I didn't join this FY6600 topic thread until reply #1702 (page 69)

I started contributing to this FY6900 topic with reply #33 (page 2)

Anyone interested enough to view these topics (you can read any EEVBlog topic as a guest - no need to sign up for membership unless you want to post) will see that the FY6600 was mentioned in this very news group way back in October 2018!

You'll also see the story about how reading the "amazing ARB pricing" thread in the SED NG had led me to my very first DSO purchase, swiftly followed by the infamous FY6600-60M a week later after comparing it against the cheapest of Siglent's 2CH AWG models the SDG1032 which then weighed in at an extra 280 quid.

The choice, for a first time punt on a signal generator was a no-brainer and, despite some obvious but fixable shortcomings, proved in the event to be one of the wisest choices I ever made. You can see why I'd come to that conclusion by reading my contributions here:-

BTW, you're not wrong about how lightweight it is! :) However, I'm using my shiny new SDG2042X simply as a plinth to raise my much modded FY6600-M to a more convenient level above the bench top.

Until Siglent finally fix its UI issues to fully realise its potential, that expensive SDG2042 simply isn't up to the task of syntonizing a Rubidium reference to a GPSDO reference in 10μHz (or even 1μHz) increments as I am able to do by feeding it into the external reference socket on the back of the FY6600 so as to dial in calibrated frequency offsets without putting needless wear and tear on the LPRO-101's trimpot whilst I work on my project to create a temperature stabilised RFS for my workshop.

In essence, to quote from the SGD2042X thread,

"For now, I'm experiencing the rather unexpected situation where the expensive AWG is reserved for the "Less Demanding" work whilst that cheap 'toy' FY6600 is my go to AWG for the more demanding tasks."