Power supply alternative for FY6900 signal generator

Question

I am purchasing a FeelTech FY6900 60MHz signal generator. Reading the reviews, its power supply seems to be its weak point. I would like to replace it with a Mean Well 65W Triple Output Switching Power Supply. It has the required +12V, -12V, +5V outputs. However since this is a switching power supply I am wondering if it will generate noise that might interfere with the quality of the waveforms. The specs for the Mean Well RT-65B state +5 out noise/ripple of 80mVp-p, +12 out 120mVp-p, and -12 out 80mVp-p. I am a novice and don't know if that much noise/ripple is significant and might interfere with the quality of the wave forms.

For $20. I am assuming this could be a good upgrade. Thoughts?

thank you,
Chuck P.Rocklin, CA

https://www.meanwell.com/productPdf.aspx?i=484

michaelkellett · Answer

I hate to say it but with the FY6900 you are rather getting what you paid for - not much Probably not worth spending much more time or money on it - but you could check the frequency of the oscillator and replace it if its way out .You are seeing about 22ppm error - so it may be in spec if your measuring has 2ppm error. The Rigol DDS signal generators are better for maybe 4x the price (although they do have cheaper ones) and the Keysight or Tek ones much, much better (for 40x) Old HP stuff is very often a good way to go if you get it for the right price - I have an HP pulse generator that cost me about £100 at an auction - I can't replicate its performance for less than £8k for a new instrument. MK

jc2048 · Answer

One of the projects I found on-line, with a quick search, suggests that the original SMPS outputs 5V/13.5V/-13.5V. That would make sense if there are further linear regulators on the analogue boards. Perhaps it would be a good idea to measure the voltages first, just to make sure what they actually are, before trying to buy a replacement [it's possible that a 5V/12V/-12V supply has been reworked]. If you do take it apart and probe around the SMPS board, be sure to take sensible precautions against electric shock. The input to a SMPS rectifies the mains coming in, so you end up with high-voltage DC on an electrolytic capacitor and that's potentially quite dangerous.

Andrew J · Answer

At the risk of pointing you away from this great site, this is something that is discussed in great detail on the eevblog forum. Just do a search - people who have bought this unit have modified it to within an inch of its life, including changing its power supply.

cdpar2000 · Answer

thank you. Ironically enough, having just rec'd a KKMOON FY6900 (FeelTech clone), the SMPS has the following outputs +/-5V, G, G, and +/-12V. However, only the +5V is used, running a +5V and ground wire to the main board. The +/-12V is not used. I'm purchasing a better quality SMPS PMC Panel Mount Power Supply 5V 15W 1 Phase / PMC-05V015W1AA from Mouser, 40mm fan, and adding an inline fuse on line from AC input. Also, replacing the thin (22gauge?) wires from AC to SMPS with 16 gauge wires. thanks for all the great feedback. I understand the FY6900 is a cheap Chinese device, but I am just getting into this "hobby" so it's a learning aid.

cdpar2000 · Answer

Yes, it appears this is a different rev board. you can see, on the PCA, the location of a 6-pin connector, no connector on board. the single wire goes to the AC ground spade. Also, it appears the earth ground wire is connected to the negative wire going to the main board as I verified with continuity. I'm purchasing a better quality SMPS PMC Panel Mount Power Supply 5V 15W 1 Phase / PMC-05V015W1AA from Mouser, 40mm fan, and adding an inline fuse on line from AC input. Also, replacing the thin (22gauge?) wires from AC to SMPS with 16 gauge wires.

baldengineer · Answer

Without knowing the instrument's schematic, it isn't easy to say. First, do you know for sure the existing power supply is linear? If it is a switching supply, the Mean Well is likely much higher quality than the built-in supply. Next to consider is how much filtering the DDS and amplifier circuits have to accommodate the supply's ripple.

In terms of frequency stability, the noise will have a minimal impact on the output. Vertically, however, I would expect some fraction of the ripple to show up on the output. Again, it depends on how it compares to the original design.

Last, if you're going to purchase a supply to replace the built-in supply, have you looked at what instruments compare at that price point? Don't forget to factor in your time and the possibility that the supply does not improve its performance.

mathijs · Answer

I only only use the USB connector for power. No main connector connected. It works fine, no noise!

cdpar2000 · Answer

Thank you. The FY6900 appears to use a very cheap SMPS. As you noted, the MW SMPS is of higher quality and for ~$20 might be an OK upgrade. As I'm learning about electronics, the process of replacing the FY6900 SMPS with the MW SMPS would be more a learning exercise. I assume expensive lab signal generators use more sophisticated (linear?) PS's.

thanks for your site. It's helping me with my hobby!

baldengineer · Answer

cdpar2000 wrote:
I assume expensive lab signal generators use more sophisticated (linear?) PS's.

I can see why you would think that. However, more expensive equipment tends to have higher quality SMPS power supplies. At least, for the bulk supply. Near sensitive sections, like amplifiers, you might see some linear low-dropout regulators. Though, even those are getting phased out as switching frequencies in point-of-load supplies get closer to 20 MHz.

The 'noise' of a switching power supply can be reduced and mitigated with good design. (Which tends to be more expensive.)

Power supply alternative for FY6900 signal generator

Top Replies