0048 - Improving the JDS6600 Part 1: The Power Supply

[@aerofart]

Nice mods, nice work. I am enjoying your videos. It’s nice to watch a professional do what he does. Bonus for me, we share at least three bench instruments in common, so watching your videos is almost as good as being at my own workbench. Thanks. Subscribed.

[@uni-byte]

Thanks! Which instruments do we both have?

[@jakubczajka4275]

God bless Chinese and you for making electronics accessible for poor peasants like me.

[@uni-byte]

Thanks Jacob!

[@heinzkreutziger2218]

Great video and I like the method to modify a test device rather than only to recommend to buy a better one. What I do not understand is the leakage voltage issue. What exactly is the problem? I guess you will only have an issue if your attached circuitry (device under test) has connection to earth (protective) ground. By the way you have implemented the switch bypassing the capacitor to basically get the leakage voltage down to zero. Do you really need it or was it only a provision that is nice to have but you never used it?

[@uni-byte]

@heinzkreutziger2218 That one I actually bought many years ago and it was my main signal generator until I got my SIglent. I used as it was. TO be honest for those years I was not doing much in the way of stuff I needed a signal generator for and was concentrating mostly on vintage computers, small 8-bit computers and microcontrolers. It fit my needs I guess. The leakage voltage is a common problem with switching power supplies. It is caused by parasitic capacitances in the transformers and issues with other non-ideal parts. High ESR filter caps, stray inductances, etc. Better SMPS have better components and dot no produce as much. It's not actually an issue and is of very low current in most cases but where the generator (or other device) is going to be attached to sensitive components (FETs and such) in ground referenced equipment if can cause considerable damage.

[@heinzkreutziger2218]

Ok,many thanks for explanation on leakage voltage. I have checked the market for function generator with good performance for reasonable price and I think I will give the UTG962E from Unitrend a try.

[@aquinamedia4508]

GREAT video good Sir, most appreciated. Great educational channel you got :)

[@uni-byte]

Thank you!

[@captainboing]

Lovely job

[@uni-byte]

Well, tank you!

[@EduardoCruz-ur4wq]

Hi, I have JDS6600 Lite and it has no issues like I saw in your video, it seems you have a defective one.Anyway I will follow you in mounting a power supply inside. Mine is from Joy-It but is the same.

[@uni-byte]

That's good. Wall-wart power supplies are touch and go I'd say your is unusually good. But that aside. My biggest complaint with the power supply was the leakage voltage. Have you tested for that?

[@EduardoCruz-ur4wq]

@@uni-byte Not yet, but I in my mind the power supply was for replacing, as I didn t open my set I didn t know that was a huge space inside to mount one, I saw it in your video. Thanks.

[@EduardoCruz-ur4wq]

@@uni-byte An upgrade, I have mounted a linear regulated power supply inside and changed the cheap electrolytic cap in the power input of the motherboard, and a block with an iec, fuse and mains switch in the rear.

[@uni-byte]

@@EduardoCruz-ur4wq Nicely done!

[@wv_]

So would you still recommend this function generator given it’s price and effort or would you recommend paying maybe double and getting an old fluke or Agilent function generator

[@uni-byte]

I used this for years before going ahead and doing the upgrades. However, it has gotten more expensive since I bought it. It depends on your needs. Beyond what I fixed it is a very useful instrument. Buying a used Agilent or other big brand name unit that will work out to 60MHz might do more than 2X the price damage to you savings .. a lot more. Those units don't come cheap and you may need to repair or calibrate them before making used of them. The there is the 2nd generation of these JuncTek (PSG9080) and similar machines that have addressed a lot of the problems and brought many improvements. They run about twice the price. Up from that in the 3x price range are the units from SIglent, Rigol and GW Instek. They are stellar performers and come with support and warranties. So it depends on several factors. Your needs, your budget, and if you go for the used market your repair skills.

[@wv_]

@@uni-byte thanks for taking the time to reply. I’ll check that out. I did indeed underestimate the price. eBay search was down when i made the previous comment, I was trying to search before i hit send. My needs are very basic but I prefer to buy once cry once. $250 would be my limit. Haha, just googled the PSG9080, they changed the name! Previously it was Junctek.. I think someone must have pointed this out to them as not the best name

[@choosers5177]

Grettings. Is there a way to make this generator operate in high impendance mode?

[@uni-byte]

It already is. "Hi-Z" for a signal generator is just how the output voltage is displayed. There are no hardware changes. On a generator that has 50 ohm and Hi-Z modes the outputs will remain at 50 ohms, but the displayed voltage will change depending on the choice of load. Half as much for 50 ohms and vice versa. For the JDS6600 the voltage is displayed assuming a Hi-Z load. Given this I suppose changes could be made to the firmware and add/re-program a button to change the display to half for 50 ohms. It's just as easy though to half the displayed voltage in your head when driving a 50 ohm load.

[@choosers5177]

@uni-byte sorry im mew to this What i want is to simulate those specs in around 40khz square wave, 20Vpp Output voltage impedance approx: 390 Ω Load impedance approx: 0-1 kΩ short circuit operating range 1-70 kΩ No current recognition aprox. >70 kΩ ZAPPER device, DR. HULDA CLARK

[@uni-byte]

@@choosers5177 The JDS6600 has an output impedance of 50 ohms. You can make a Hulda Clark zapper with a 555. No need for a JDS6600. The original zapper had an output impedance of 1K ohm, but that's tiny compared to the resistance of the human body. If you wanted to use the JDS6600, just take the output through a 1K resistor to the handles, then set it up for a 10Vpp square wave @ 30kHz with a +5V DC offset. Kind of overkill for something you could build for $0.50 but it might look cool to the "patients" in a quack clinic, right next the the therapeutic healing crystals and the homeopathic apothecary.

[@choosers5177]

@uni-byte @uni-byte thank you. I actually wanted to experiment with different frequencies and waveforms. That's why i want an arbitrary waveform generator. Also, there is no clinic, just a self test project at home. First i thought about buying an TS250 waveform amplifier becsuse it has 1kilo omh impedance but ill try what you suggested even its a bit complicated for me i guess.

[@gino3286]

hi very interesting video of a nice unit if low noise is a goal why not try a linear power supply to provide the +5VDC usually decent ones have noise below 10mV

[@uni-byte]

Thanks. I agree, a linear power supply would be far better. Much lower noise and zero leakage. And I did try to find a linear supply that A) fit in the case, and B) did not cost more than the unit itself. But I failed. I even thought of making one myself, but A & B above kept getting in the way. When driving a square wave at maximum amplitude into 50 ohms on both channels this requires close to 2 Amps. I do have a 3 Amp 5V supply, but with the transformer, filter capacitor and heatsink it is huge.

[@gino3286]

@@uni-byte hi thank you for the very kind and valuable reply You clearly better than me I am thinking of setting up a little lab at home to test some Audio gear This unit is very interesting Thank you very much again for the test Kind regards Gino

[@uni-byte]

@@gino3286 You are welcome. Better? No, just maybe a little more experienced. As for starting a lab, this unit would be a nice addition (drawbacks aside). I used my unit for several years.

[@gino3286]

@@uni-byte hi thank you very much again i would like to carry out some square wave tests on audio equipment I like the test a lot Very challenging but also very telling of a unit quality I have wasted money on units and lab equipment look a much smarter buy

[@uni-byte]

@@gino3286 It should be great for that. The rise time on the square wave is about 10.5 ns on the one I have (the 60MHz version).

[@t1d100]

Excellent