Author Topic: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator  (Read 549194 times)

0 Members and 3 Guests are viewing this topic.

Offline cybermaus

  • Frequent Contributor
  • **
  • Posts: 674
  • Country: nl
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #150 on: November 11, 2017, 07:20:11 pm »
Ok, so I do not think we will ever reach full concensus, but because I said I would, I went and measured current as well as voltage on a range of devices.
See attached table with AC values. DC was always 0 for all, so I did not even add it to the table, my basis is a 230V 50Hz mains

My conclusion: the FY6600 is indeed not the best (and I focus mostly on current) but in fact the Rigol DS1074Z is a lot worse. If you disconnect the earth.
And if you do connect them both to earth, again the Rigol is worse.

For the rest, the official Samsung USB is the best, by far, followed by my cheapo china linear bench supply.
In fact, the Samsung is so good, I had to doublecheck my DMM had not stopped working.
I measured some more cheapo supplies, but stopped writing them down, most are similar to the FY6600.

What I have learned from this is that if at all possible, I should avoid using my Rigol in "fake differential" mode by disconnecting the earth. I have done that in the past, have a special cable for it. I will modify that cable to also have the Y-cap (actually X2 because I have those) and 1M resistor, and use it sparingly. Yes, go ahead and scold me for it.
If only the math function (A-B) on the rigol was not so annoyingly slow.

The FY6600 will get a X2 and 1M, pretty much like beanflying, but also a switch so that most of the time it is simply fully grounded and not floating at all.
 
The following users thanked this post: bitseeker, beanflying

Offline don.r

  • Frequent Contributor
  • **
  • Posts: 740
  • Country: ca
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #151 on: November 11, 2017, 07:36:22 pm »
I desoldered the mains leg of the Y cap and added a 10nf Y cap (that's what I have) from the mains side to the free leg of the original cap and then grounded the joined legs of the caps. Now getting 2.2VAC on low-z and a handful of uA (into the noise). This will be my default "floating" and then add a switch to fully ground the output.

Adding the grounded IEC was easy as the back panel is very thin and a dremel cuts through it like a knife through butter. Note that the original cable and socket are not even polarized.  ::)
 

Offline bitseeker

  • Super Contributor
  • ***
  • Posts: 9057
  • Country: us
  • Lots of engineer-tweakable parts inside!
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #152 on: November 12, 2017, 05:50:54 am »
For the rest, the official Samsung USB is the best, by far, followed by my cheapo china linear bench supply.
In fact, the Samsung is so good, I had to doublecheck my DMM had not stopped working.

That is really good. What was the arrangement in their supply (or is it inconveniently welded shut)?
TEA is the way. | TEA Time channel
 

Offline ArthurDent

  • Super Contributor
  • ***
  • Posts: 1193
  • Country: us
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #153 on: November 12, 2017, 04:03:25 pm »
Well I believe I’m pretty much finished with the modifications to my FY660.  I’ll answer a couple of questions that popped up concerning my equipment or what I did then show some photos of the finished product.

As to testing this FY6600 using my GPSDOs as standards and how ‘accurate’ they are, what I have is two 26 dB gain GPS antennas at the opposite ends of my roof that drive two GPS distribution amps to allow me to run and test multiple GPS receivers simultaneously. I had sold about 200 Trimble Thunderbolts GPSDOs in the past which is one reason for the setup I have. As to the accuracy of my GPS system, long term it is about as accurate as the GPS source and short term the 10 Mhz output stays within a couple hundred PPT worse case. A 25-year old rubidium standard I have has better short term stability but a couple of my crystal standards have less noise and very good short term stability.  For most stuff all this is far more accurate and stable than I would ever need but you gotta have neat toys, right?

I have carried the physical ground from the power cord right through on the FY6600 and almost every piece of equipment I own comes grounded that same way.  The only place I ever had to be concerned about ground loops was a place I worked where we tested (sometimes very destructively) products with high voltage/current pulses a few microseconds long and with current spikes of 20,000 amps. Even with the heavy copper bus bars for grounding we had to pay careful attention to using a common ground point so the readings we took weren’t influenced by the voltage drop across these massive copper bars.

The Fluke/Philips 6681 or Pendulum CNT-81 counters I use are excellent counters that let me display the 10 Mhz frequency I am generally interested in directly to 10 digits (2 decimal places) in one second or by using the built in math function to 5 decimal places in 10 seconds which is perhaps giving me more resolution than accuracy, but not by much.  I have a saved function on the counters that subtracts 9,999,999hz from the 10,000,000 Hz signal. If the test 10 Mhz input is exactly equal to the 10 Mhz external reference from my standards, in 10 seconds the counter will display 1.00000 or 5 decimal places (ignore the 1, it just helps the display format).  I generally also feed the same signal into my scope which is externally triggered by my same external 10 Mhz standard and check for drift at 2 ns/div giving me a visual indication as well. 

Why the 50 Mhz timebase in the FY6600 bugged me so much is that the frequency output can be set to 14 digits but the timebase was only accurate to about 5 digits. Also it wasn’t stable or trimable. With the top cover off the FY6600 and the output set to 10 Mhz and displayed on a scope triggered from an external 10 Mhz standard, if you just blew across the 50 Mhz oscillator you would see the frequency change rapidly and drastically. The OXCO I built into the FY6600 looks very good after warm-up and I can always use an external 10 Mhz reference if I want more accuracy and stability. Here is what the frequency output set to 10 Mhz (and the oscillator) looks like on warm-up after having been trimmed previously.  Start-up  -200hz ; 3min +.05 ; 20min +.01 ; 30min onward +/-.005hz or +/-5 at the 11th digit. Because the rear switch on the FY6600 turns the power supply on/off and the front panel power switch just puts the main board in standby mode, if you leave the rear switch on the OXCO is always powered and you eliminate the warm-up period. Generally the longer a good crystal oscillator is left on the more stable it becomes and some of the better standards I have take over 2 weeks to really stabilize. Using a TXCO with low power supply current requirements instead of an OXCO like I used could allow you to use the existing SMPS with some loss of accuracy and stability if you want to keep the modifications simpler.

The photos show the back panel with the added BNC connector for 10 Mhz ext ref in, the int/ext switch, and the recessed trimpot. Unfortunately installing these parts covered some of the text on the rear panel but I can live with that. Perhaps a small SMA connector and a smaller switch would be a better choice. The inside photo of that area shows how the OXCO is mounted on a phenolic board that was hand wired and that board mounted on the aluminum bracket held to the rear panel by the BNC connector and the switch mounting hardware.  There are a few resistors near the switch to adjust the levels the PLL chip sees from the OXCO and the ext ref and I may add a buffer to further protect the PLL chip from possible input spikes from any ext ref. The 3rd photo shows the sub-miniature Teflon coax I securely soldered to the ground plane near the PLL chip and the center wire soldered to the input pin. It was important that there was little strain on that cable so it wouldn’t rip off and break the input pin off the PLL chip.
 
Sorry for being longwinded but I hope this explanation helps others decide what they want to do and what is important to their needs.
« Last Edit: November 12, 2017, 07:13:08 pm by ArthurDent »
 
The following users thanked this post: battlecoder, Johnny B Good

Offline cybermaus

  • Frequent Contributor
  • **
  • Posts: 674
  • Country: nl
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #154 on: November 12, 2017, 06:28:13 pm »
So your stated credentials are such that I have to assume you know your way around time standards. So please consider these questions as education for me, and not as questioning your setup.

As to the accuracy of my GPS system, long term it is about as accurate as the GPS source and short term the 10 Mhz output stays within a couple hundred nanoseconds worse case.

So a couple of "a couple hundred nanoseconds" does not actually sound very precise. When I read the NEO-M8N specs, it mentions 30ns, but graphs in the application note show this to be -17/+13ns on a 1pps pulse, mostly due to quantification of the internal 48MHz TCXO.

So am I interpreting this wrong, or is your GPS older and not very precise when compared to recent GPS modules?

The application note also lists other precisions, like 0.062ppb frequency stability (maybe very long term?) and a 4E-9 (4ppb) Allen deviation at 1 sec, and I cannot make heads or tails of the sub second precision expressed in spectrum graph, though I do see it is based on a 24MHz signal with only 3.046ns jitter.
Which I am hopefully interpreting as possible to get 3ns precision, if you stay with an integer division of 48MHz timepulse, which is also advised in the summery of the application note.

Could I tempt you to read the application note, and give me your though on it.


I could not find a local source (only potted modules for drones), so am waiting for china, but alas, ebay kicked out my seller, so I had to re-order, and wait even more.
 

Offline ArthurDent

  • Super Contributor
  • ***
  • Posts: 1193
  • Country: us
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #155 on: November 12, 2017, 07:28:50 pm »
Cybermaus, when you say: "So a couple of "a couple hundred nanoseconds" does not actually sound very precise." you are correct. I have corrected that reference in my previous post after reading your comment. The line should have read "PPT" for parts per trillion, not nanoseconds. Generally the time instability is 2 ns or less. Sorry for the screw up and thanks for picking it up. Here is a 10 hour graph that shows the 10 Mhz output stability of one of the Nortel/Lucent GPSTMs that I modify.
 

Offline Ebel0410

  • Contributor
  • Posts: 41
  • Country: fr
Just ordered on eBay a TCXO 25MHz and 5x ICS511M for a total amount of 10$.
This TCXO is announced better than 1ppm (it means 0,0001%) and it can be trimmed +/-5ppm around 25.000MHz.
I hope that the ICS511 (used as frequency doubler) will not introduce too much jitter.

The frequency tested on the stock CXO 50.000 XTAL in my FY6600 is 50.000.492Hz,

I'll keep you posted when I'll receive them.
« Last Edit: November 12, 2017, 07:54:09 pm by Ebel0410 »
 

Offline don.r

  • Frequent Contributor
  • **
  • Posts: 740
  • Country: ca
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #157 on: November 13, 2017, 12:08:01 am »
Following the lead of others, I am also looking at building a TCXO and PLL with maybe an optional 10MHz input. Might be a nice add-in board kit for someone to design. Apart from the PSU, the timebase is the other major weak part of this device.
 

Online beanflying

  • Super Contributor
  • ***
  • Posts: 7358
  • Country: au
  • Toys so very many Toys.
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #158 on: November 13, 2017, 02:20:56 am »
And to think all I wanted was a 'cheap' signal generator that worked  :palm:

Nice work on the mods all  :clap:

off to look at 'cheap' spectrum analyzers next .........
Coffee, Food, R/C and electronics nerd in no particular order. Also CNC wannabe, 3D printer and Laser Cutter Junkie and just don't mention my TEA addiction....
 

Offline don.r

  • Frequent Contributor
  • **
  • Posts: 740
  • Country: ca
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #159 on: November 13, 2017, 02:46:10 am »
And to think all I wanted was a 'cheap' signal generator that worked  :palm:

Nice work on the mods all  :clap:

off to look at 'cheap' spectrum analyzers next .........

Yeah... to be fair, I knew about the PSU going in and its not an expensive nor time consuming mod. The poor time base is a disappointment but the added cost is not huge, although it may require more time to complete. I hope Feeltech is paying attention here. They could have a real killer AWG for the money with a few small changes.
 

Offline Vytautas

  • Regular Contributor
  • *
  • Posts: 67
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #160 on: November 13, 2017, 06:18:34 am »
Hello, everyone,
I just finished my modifications of the fy6600:

1. Replaced the power supply with a linear one. The voltage was increased to +-15V because +-12V does not give enough headroom for the output amplifier to be able to go all the way up to the maximum amplitude of 20Vpp (or, rather 10Vpp into 50 Ohms).

2. Then, I replaced the dodgy output amplifier op-amp. The original one is an unknown 30021. At first I thought this is a dual variant of the single THS3001 but some guy has inquired about it from Texas Instruments and they replied that they do not manufacture such. So, who knows what FeelTech uses here. Therefore, I chose to replace it with two single THS3095 op-amps. Thankfully, the board has places ready for them, with all bypassing and anything needed. Beauty! This change (and, of course, new +-15V power supply) now lets the generator output full 10Vpp into 50 Ohms with no distortion (the original 30021 would go only to somewhere above 7Vpp and heavily clipped, and increasing power supply did not do much).

3. Some other mods - changed the output amplifier heatsink to a larger one, added a small heatsink to the FPGA, added a 12V fan (I run it from 8V, so it is quiet; a separate power supply for the fan, because connecting a fan to the 5V or 15V rail will affect the performance of the generator).

Here is the list of pictures:
1. Original board with the dodgy 30021 op amp.
2. Modified board with two THS3095 op amps.
3. 1M sinewave at full amplitude of 10Vpp into 50 Ohms termination.
4, 5. 1M sinewave at 5Vpp and 20Vpp. In both, harmonics are -60dB down. Beauty!
6. Modified internals.
7. 1M sinewave AM modulated with the 1K signal from the second channel.
Cheers!
« Last Edit: November 13, 2017, 06:40:13 am by Vytautas »
 
The following users thanked this post: Mozee, Ebel0410, beanflying, Bob Sava, Johnny B Good, jebem

Offline don.r

  • Frequent Contributor
  • **
  • Posts: 740
  • Country: ca
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #161 on: November 13, 2017, 04:17:48 pm »
Nice work but won't the presence of a fan affect the 50Mhz time base?
 

Offline cybermaus

  • Frequent Contributor
  • **
  • Posts: 674
  • Country: nl
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #162 on: November 13, 2017, 04:56:51 pm »
Just picked this up from the manual:

"Use the process, long press the parameter adjustment knob (OK button), you
can quickly save the instrument output parameters, the next boot,
automatically load the saved parameters."

And it actually works, except (Feeltech, if you read, please take note!)
- There is absolutely no feedback that the save has happened. No beep, no message.
- It saves the waveform, but it does not save if the channels are active or not. I feel if for example Ch2 is inactive during save, it should also be inactive after reboot.

Still, a useful bit of info I thought.
 

Offline cybermaus

  • Frequent Contributor
  • **
  • Posts: 674
  • Country: nl
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #163 on: November 13, 2017, 05:08:49 pm »
Checking @Vytautas: Indeed, 20Vpp at high-z is OK, but at 50R is at 7.x V only, and looks to have very rounded domes.

But is that due to the 12V supply (actually about 14.8), or due to the "donky" OpAmp? It must step up the 12V to 22V anyway, because at high-z I can even add 1.5V offset before clipping starts, getting a nice shape up to 21.5V and -21.5V

Also, 20Vpp at high-z, it looks perfect sine (eyeballing) up until 3MHz or so, then it starts to visibly distort. Hence I was reading the manual to check if the 20Vpp was supposed to be up to what frequency. But it automatically steps down to 5Vpp at 20MHz. I wonder if @Vytautas can check that with his new OpAmps.

 

Offline Vytautas

  • Regular Contributor
  • *
  • Posts: 67
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #164 on: November 13, 2017, 06:17:40 pm »
Hi, cybermaus,
The problem is mostly due to the "30021" op amp. As you say, it is only just above 7Vpp into 50 ohms. And I had tried at first powering the generator from my lab supply with +-15V. The improvement was miserable if any. So, it must be the op amp.

With the THS3095 the generator easily outputs 20Vpp (10Vpp into 50R), no problem. Still, if to be picky, the behavior of the THS3095 is in line with what the manufacturer shows in their datasheet for the component. The pictures bellow are taken from the datasheet. At high frequencies, as those graphs also show, I see some distortion (not clipping though but exactly the same as manufacturer graphs show). The manufacturer suggests properly paralleling two op-amps for each channel (second picture) to remove most of the distortion. That is what major manufacturers, like Siglent, and others do in their designs. For the sake of interest, I might have a look at that later (accidentally, I do have two more THS3095s in my drawer). By the way, Texas Instruments have recently introduced a new power opamp - THS3491. If what they say is true, that one is a real beast with unmatched performance. But they do not have them yet.

Sure, the generator steps down to 5Vpp at 20MHz. But that is to be expected. Normal behavior. Difficult to achieve frequent swings at high voltages.
 
The following users thanked this post: Johnny B Good

Offline Vytautas

  • Regular Contributor
  • *
  • Posts: 67
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #165 on: November 13, 2017, 06:23:56 pm »
Nice work but won't the presence of a fan affect the 50Mhz time base?

Why would you think it would?
 

Offline don.r

  • Frequent Contributor
  • **
  • Posts: 740
  • Country: ca
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #166 on: November 13, 2017, 06:29:30 pm »
Nice work but won't the presence of a fan affect the 50Mhz time base?

Why would you think it would?

A previous poster has noted that just blowing on the oscillator changes the frequency. It appears to have poor temp stability.
 

Offline cybermaus

  • Frequent Contributor
  • **
  • Posts: 674
  • Country: nl
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #167 on: November 13, 2017, 06:33:19 pm »
Sorry, my question was not properly constructed:

"Also, 20Vpp at high-z, it looks perfect sine (eyeballing) up until 3MHz or so, then it starts to visibly distort. Hence I was reading the manual to check if the 20Vpp was supposed to be up to what frequency. But it automatically steps down to 5Vpp at 20MHz. I wonder if @Vytautas can check that with his new OpAmps."

I did not mean to ask if it also stepped down to 5Vpp, because that is kinda obvious.
Instead I meant: Can you check with your new Op-Amps if the range from 3MHz to 20MHz in high-z is still so horribly distorted, or did that improve with those op-amps.

And now I have a new question, based on your graphs: the old op-amps had a pretty good high frequency signal at lower voltages. @Ebel0410 showed -44dB
Are you saying that is now worse with your new opamp, or are those errors only at full voltage swing?
 

Offline Vytautas

  • Regular Contributor
  • *
  • Posts: 67
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #168 on: November 13, 2017, 07:26:26 pm »
Rest assured, the performance with the new op amps is way better.
Here are some of the photos - 3MHz, 20MHz, 30MHz, 60MHz - at maximum possible amplitude. Left is High-Z, right - 50 ohms terminated.
 

Offline don.r

  • Frequent Contributor
  • **
  • Posts: 740
  • Country: ca
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #169 on: November 13, 2017, 07:41:43 pm »
Do the upcoming 3491's share the same footprint? Might make a good future upgrade.
 

Offline Vytautas

  • Regular Contributor
  • *
  • Posts: 67
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #170 on: November 13, 2017, 07:48:42 pm »
Yes, they are a direct pin compatible replacement for the 3095 with performance figures hard to believe. Lets see. They certainly should become a worth replacement.
 

Offline Vytautas

  • Regular Contributor
  • *
  • Posts: 67
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #171 on: November 13, 2017, 07:51:32 pm »
Here Texas Instruments have a comparison table:
http://www.ti.com/product/THS3491
Better BW, better slew rate,  better noise performance.
 
The following users thanked this post: don.r

Offline Vytautas

  • Regular Contributor
  • *
  • Posts: 67
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #172 on: November 14, 2017, 10:41:16 am »
Hi, cybermouse,
Just to be sure, I tried powering the modified board with new op amps from the original switching power supply. Sure enough, the waveform is clipped and heavily distorted in high amplitude settings. So, the voltage from that supply is not enough. +-15V is a must (no wonder, this is to be expected).
 

Offline cybermaus

  • Frequent Contributor
  • **
  • Posts: 674
  • Country: nl
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #173 on: November 14, 2017, 10:56:17 am »
Ok, noted. Not sure how mine does the high impedance 20Vpp then, but noted anyway.

So just to make clear: You state to have a +-15V linear transformer, but are feeding linear regulated (79/7815, so +-15V) to the device? Or are you regulating higher, using the  sqrt(2)*15V=21V peaks?
 

Offline Vytautas

  • Regular Contributor
  • *
  • Posts: 67
Re: FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator
« Reply #174 on: November 14, 2017, 11:36:06 am »
Ok, noted. Not sure how mine does the high impedance 20Vpp then, but noted anyway.

So just to make clear: You state to have a +-15V linear transformer, but are feeding linear regulated (79/7815, so +-15V) to the device? Or are you regulating higher, using the  sqrt(2)*15V=21V peaks?

I am using a 20V linear tranformers (more precisely, two 9,8V windings of the transformer in series) and regulating with LM317 and LM337 regulators both for +-15V and +5V (these, actually, have a better noise performance than 7915/7815s). The voltage stays +-15V under load, naturally.
« Last Edit: November 14, 2017, 11:39:15 am by Vytautas »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf