FeelTech FY6600 60MHz 2-Ch VCO Function Arbitrary Waveform Signal Generator

[Diabolo]

Hello,

This power supply will do the job by adding an internal transformer, or external power supply.
- https://www.banggood.com/fr/DIY-USB-Boost-Single-Turn-Dual-Power-Module-Linear-Regulators-Multiple-Output-Power-Kit-p-1022517.html?rmmds=myorder&cur_warehouse=UK
- https://fr.aliexpress.com/item/New-Arrival-DIY-Power-Supply-Module-USB-Boost-Single-Turn-Dual-Linear-Regulated-Power-Kit-Regulator/32695921692.html?spm=a2g0w.10010108.1000014.24.32b7a9aOyqOWH&traffic_analysisId=recommend_3035_null_null_null&scm=1007.13338.80878.000000000000000&pvid=506d4eea-89c5-4977-9f9d-77e306085a91&tpp=1




Diabolo

[Chris56000]

Hi!

https://www.banggood.com/fr/DIY-USB-Boost-Single-Turn-Dual-Power-Module-Linear-Regulators-Multiple-Output-Power-Kit-p-1022517.html?rmmds=myorder&cur_warehouse=UK/quote]

I'd certainly go with one of these (bought in kit form so I can document the schematic & parts list of course!!!) and have been thinking about one – I've got an RS 78xx PCB floating about to fit rectifiers & smoothing capacitors to!

Does anyone know the approximate current consumption of the FY6600 (per rail) driving 10V p–p square @ 1 MHz into 50 ohms ?

Chris Williams

[DC1MC]

Hi!

Has anybody tried modernising their FY6600 with one of these:–

https://m.ebay.co.uk/itm/New-Multi-channel-Linear-DC-Power-Supply-Module-3-3V-5V-12V-Adjustable-Output/262612931341?hash=item3d24f3530d:g:ZgYAAOSw8oFXz3Vu

Are they capable of supplying the necessary power requirements of the unit? I'm thinking of replacing the cheap and horrible SMPSU with one of these linear modules with the advantage of transformer isolation!

Chris Williams

Hi Chris, the idea of using an analog power supply it's excellent, I'll do it my self as one of the last steps of modding the FY into a better instrument, the mains isolation and lack of SMPS noise it's a great thing to have if the transformer and caps are fitting in the box. But the posted power supply looks really bad, the 12V AC secondary and the 2 anemic filtering caps will not produce such a good power.
UPS finally send my tracking number and in max two days mine will reach me, I'll do the power measurements but there is no way this PS will sustain 10Vp-p on a 50ohm load, IMHO.

[DC1MC]

In preparation to the big communication protocol reversing, before my device and LA arrives, I have a request for the people who can still open their FY and do some measurements:

 - By looking at the posted pictures and videos, it seem that the flat cable caries 2 things:
 
- 1 Serial Interface that goes to the signal board serial to USB connector.
- And the 8 (?!?!) bit parallel (?!?!) bus for configuring the the generator.

I have the following questions:
- The parallel bus has really just 8 wires, I may have miscounted looking at the pictures ?
- Has anyone put an oscilloscope on this bus, has the communication been identified or at least if there's some clock, R/-W pin and similar,  is the bus bidirectional or there are separate Is and Os ?

Also one administrative question: should we keep one of the current threads alive with this project (either this one or the one with the firmware flashing) or should I create a new thread ?

I soo much wish that the UPS guys will move a bit faster 

[rhb]

 I located a clock and events when I pressed buttons  The ribbon has 12 wires, four of which appear to be for the RS232 connection. IIRC I only observed data on 3 lines + the clock.  That suggests that every other line is ground.

I did not have a reliable connection to the bus, so I stopped until I have time to put an attachment for probing together.

[DC1MC]

I located a clock and events when I pressed buttons  The ribbon has 12 wires, four of which appear to be for the RS232 connection. IIRC I only observed data on 3 lines + the clock.  That suggests that every other line is ground.

I did not have a reliable connection to the bus, so I stopped until I have time to put an attachment for probing together.

Cool, now it's the time for Captain LA to save the day  !!!

Before I'm engaging in more (possible destructive) testing on the FP firmware I want to fully decode the protocol, especially if there are calibration constants stored in the FP firmware.
Once this is done, then there is no stress if the firmware goes puff, I was thinking anyways to move the signal board in an re-purposed scope box that has an encoder on FP and enough buttons.
Also, once the programming of the signal board it's fully reversed, a better LCD control unit can be connected, eventually a repurposed tabled with touch screen.

I'm biting my fingernails waiting for the boxes wit the FY and an LA, let's see what UPS is doing.

 

[rhb]

If you have the connector parts make yourself a 1x8 jumper cable with probe points.  I just got the connectors a couple of days ago, but I'm cleaning and reorganizing my work space (and I've got a remodeling project I need to get back to work on).  I just got my MSO-2204EA a month or so ago and am still in the process of tooling up to use it. The last time I did this everything was through hole DIPs, but those probe clips aren't much help with SMD parts.

You *might* be able to separate the factory cable wires and attach probes, but there's not a lot of slack.  I stuck the probe leads down into the back of the generator board connector but that was not a very secure connection.  I got a kit of "dupont" cable connectors from eBay and will make up a complete set of 1xN M-F probing jumpers some evening soon.

[nctnico]

I located a clock and events when I pressed buttons  The ribbon has 12 wires, four of which appear to be for the RS232 connection. IIRC I only observed data on 3 lines + the clock.  That suggests that every other line is ground.

I did not have a reliable connection to the bus, so I stopped until I have time to put an attachment for probing together.

What works well for me when probing many signals is a dual row header with enamel wires attached to it. One row are for signals, the other row is for ground. I solder the enamel wires to the board because clips let go way too easely.

[rhb]

Certainly the way to go if you need to connect to a PCB.  Sadly, to be useful with an MSO-2204EA I'll have to repackage the LA probes to use a 2x8 header.  I've planned to do that, but have other mischief I want to do first.

[DC1MC]

Well, now we have to get Cybermaus, and the other guys that suffered, on board. Time to spend some calories on something more advanced than replacing capacitors.
On connecting side, I have a crimping female FRC connector, if by chance the cable fits than that's it, if not I will pierce the isolation with some needles, if somebody know if there are some "vampire" grippers for small wires, now it will be a good chance to tell about them, and better be available in EU or I'll pester the US guys here to send me some in an envelope .

Three signals plus clock, I'm wondering what kind of contraption is there, could it be two synchronous independent channels, I 'm really wondering if the communication is bidirectional and how it's implemented.
Finally something interesting to play with, I so much wish to get on board someone with one of this advanced LAs, with a gadzillion protocol decoders.

Oh well, not much to do until the stuff arrives, I'll keep on eye here and read something more about the STM32.

I've still didn't hear the thread option, should we stay on this thread when the work starts or should I create a new thread specialized for this project ?
Here's a picture of my cat looking forward to participate on the project as a source of entropy  .


 Cheers, DC1MC

[DC1MC]

For the STM32 experiments bought myself a pair of these (one will become a controlled SWI debugger to test some of the Fraunhoffer guys experiments) :

https://www.ebay.de/itm/282786310657

Anybody has a recommendation for a non-castrated C development environment, based on GCC if possible ?

[rhb]

The STM32F429 Discovery kit includes a 320x240 touch screen for $24 US the last time I checked.  At any rate, the chips are so good, you can't have too many.

ST offers a full dev kit and there is the gcc toolchain, so everything you need several times over.  You might want to look into Matthias Koch's Mecrisp which is an approximately ANS forth of superb quality.

ST provides a large suite of examples and libraries for their devices.

[DC1MC]

I've bought the little boards not because there weren't any other more interesting or more capable, but because I needed something that has the same chip variant as the FP MCU of the FY.

[texaspyro]

Or instead of keeping with the ARM, maybe a Arduino 2560 or other Atmel CPU.   That way there is no arguing about copyright issues.

[g0mgx]

I've bought one of these units and have done some quick tests and it seems I have some wierd DC offsets in the waveforms above an amplitude of 0.5Vpp (as set on the instrument itself):

https://g0mgx.blogspot.co.uk/2017/12/feeltech-6600-another-one-really.html

Anyone else seeing this?

I've managed to "break" the software a few times and needed a power off-on to get it going again. There is certainly a firmware bug around 0.5V amplitude where a relay goes clunk inside the box; sometimes sometimes that step from below to above 0.5V can be huge even when the adjustment in amplitude made is in the decimal points of the value.

And for sure the manual states that the outputs are 50 ohm but the amplitude setting and the measurements at 50 ohm have little or no resemblance to each other.

Anyone else seeing the same thing?

Mark
G0MGX

[DC1MC]

Hi Mark, what is the firmware version of your unit ?
There were other reports of strange DC offset, but FeelTech claims that this is fixed in fw. version 3.2

The calibration constants and all the necessary math are (allegedly) stored and done in the Front Panel little MCU, and the firmware there has lots of bugs, it could be that the MCU they've chosen doesn't have enough resources.
 
I hope you can participate in the effort to reverse and understand the communication protocol between the front panel and the signal board.
The signal board is reported reasonable good as-is, and can be made better with inexpensive modifications.
My unit it's finally "in delivery" and will reach me today, so I will contribute also reports.
Also when my LA will reach me, I will start working on the communication protocol, but I need other owners to look on the bus so we can compare results.

  73 de DC1MC

[Yansi]

Hi!

I have just gone through the whole thread. Interesting device this is.

Can somebody please enlighten me how the output amplitude control is done in this device?

Thx, Y.

[DC1MC]

My FY6600 arrived, the piece of excrement Chinese distributor send one with firmware 3.1, solidifying my opinion that they are scum of the Earth. 
Of course after confirming in their crappy Chinglish that their devices have 3.2, you've seen the exchange somewhere else in the thread where that was a question about a version 4.2.

I'm asking for your opinion, what should I do, should I send it back as is or should I start working on it.

I don't care about their firmware shit, but I needed one with a (semi)decent fw. to not get their stupid mistakes in configuring the generator, decoding and extracting the firmware it's a lot of work and I don't want to do it twice. and this seem to be the firmware that does not fuck the display but has some numerical errors in setting the DC offset and signal level, exactly the stuff I wanted to have.


Is there somebody with a version 3.2 and a Logic Analyzer willing to reproduce my steps here to decode the protocol (nothing will be done on the firmware extraction until the protocol it's fully understood, so it should be relatively safe) and compare the results ?
Please let me know about your opinion.

[DC1MC]

Hi!

I have just gone through the whole thread. Interesting device this is.

Can somebody please enlighten me how the output amplitude control is done in this device?

Thx, Y.

I was planning to do a schematic capture today, while replacing the crapacitors, not sure if I'll do it or I'll be waiting for the one with the 3.2 firmware.

Anyway, a schematic will be done.

[Yansi]

OK thanks.  I have noticed the DAC904 has an external VREF input and someone have noticed there are some Microchip DACs like MCP4821 or what. I think they obviously use these to set the VREF of the DAC904, which makes sense.

(In the neighbouring thread about the JDS6600 generator, they use a 4quadrant multiplier to do it and I was curious abotu the solution in the FY6600.)

[kahe40]

a firmware bug around 0.5V amplitude where a relay goes clunk inside the box ...

no problem at all with v3.1     (push SYS, maybe you have v3.0)

< 1mV it is not really perfect, e.g. 200uV positive and 400uV negative sinus,
but from 1mV upwards everything is perfect

[g0mgx]

My firmware is V3.2 (thats what is reported on boot and in the SYS screen).

There are quite a few issues with the software all of which seem to be fixed by a mains cycle.

Another issue I have found with my unit is that **sometimes** if you power it off with a 0.5V amplitude using the soft power switch on the front, when you switch it back on again there is no output until you increase or decrease the amplitude; it then works fine at the 0.5V setting.

I wonder if I have a sticky relay or two?

I need to read through the whole thread - but it's very long now - happy to get involved in the decoding of comms.

Mark

[rhb]

I had proposed an instrument wiki so that we could organize such things more easily.  I'd love to be able to read a summary of the $20 LCR thread, but at 170+ pages it's just too long.

[g0mgx]

I've now read the entire thread and have taken the PSU board out of mine to take a look at the caps.

The 10uF 450V cap in mine has only one leg soldered to the PCB, so I don't think that will have been helping much 

I've already changed the mains socket for 3 pin type and connected the mains earth to the 0V line; saves doing this with my PC and a USB cable instead. I'll change all the caps in the output for 25V types and see what happens.

Mark

[rhb]

You may want to reinforce the panel.  Your mount is like mine and when I pushed a cord on, it deflected too much to be comfortable. So I took it apart and added a metal brace scured by hot melt glue to the bottom of the case.  The socket screws thread into the metal which was cut from on old PS case.