开云体育

On ?Sunday?, ?June? ?17?, ?2018? ?02?:?46?:?49? ?AM? ?CDT, M Yachad <yachadm@...> wrote:

MachineGuy
(I’d really like to address you by your real name!)

Thanks for that info from the tin-man.
I can certainly relate to the manufacturer’s published data.
I will take this into consideration the next time I have to select caps.

To address certain points:
1.??? I am not at all biased against Panasonic caps. They are superb, and I use the EB in Linear audio applications all the time. They are my go-to for that specific application.
2.??? I also use the Panasonic TSHA and ED/EE Snap-ins as my go-to PSU filter capacitors, with Nichicon GU as my second choice.
3.??? So, the question arises, WHY do I NOT recommend the EB in THIS Tek PSU application?
4.??? It is important to understand the function of a cap in each particular place in the circuit. In any PSU application, the AC ripple BETWEEN the Bridge rectifier and the Regulators is enormous. AFTER the regulators, the AC ripple is negligible.
5.??? A few years ago, to quieten my curiosity, I did a test on an older amplifier using 4 new apparently identical 4700uF 25V and 50V capacitors mounted BETWEEN the Bridge and the bank of Regulators (some 78xx, 79xx, 317, 319). The Nichicon HE caps reduced the residual AC ripple to approx 1V p-p (with no discernible difference between the 25V and 50V cap). The other caps reduced the residual ripple to 2V p-p. See attached pictures. This is a test which any of you can easily do, any time.

/g/TekScopes/album?id=59343


6.??? That proved to me that the HE has a significant advantage in reducing AC ripple, over another respected brand-name cap. On the datasheets, all the numbers pretty much matched up, but in the real world of flowing voltage and current, my scope waveform showed a substantially significant advantage to the HE.
7.??? However, AFTER the regulators, where ripple is negligible, an EB is an excellent choice. The superior capability of the HE regarding ripple damping is of no advantage here.

On this Tek PSU, has anyone actually done any research to specify each cap’s function? From the postings here, I don’t think so. I haven’t. So, unless someone is willing to go and locate each cap on the circuit, and determine whether it is in a low-ripple part of the circuit (where an EB would be acceptable), or it is in a high-ripple part,? I think that we need to globally specify HE or PW (functionally equivalent), in order to be on the safe side for ALL applications. That’s what I did, with full confidence that my choice would give me the lowest ACTUAL ripple on all circuits.

If a user installs EB in a high-ripple area BEFORE the regulators, will the machine work? Of course it will.
But will it work as well, if an HE or PW is installed? My test states not.
After all, a SMPS with loads of AC ripple in all parts of the circuit, is NOT like a linear PSU, at all.

It’s the difference in being able to tune a high-performance engine, to get the last ounce of available HP out of it.

The posters who are asking these questions are people who do NOT have any substantial electronics background, theory or practice. I think it is incumbent on us (the techs with long experience) to point them in a direction which will give them an optimal and safe solution, so that they don’t have to tear their hair out, buying parts which don’t fit, or worrying whether what they choose will last or not.

Regarding the Lifetime numbers of the data provided by MachineGuy:

A 5000Hr lifetime is provided for a (boiling) CONSTANT temperature of 105°C. If the caps in my Tek scope are reaching that temperature, then my scope is on the way to the garbage dump, very quickly, and the 5000hr lifetime is functionally irrelevant.

I have measured the temperature of my HE caps in my A2A1/A3 PSU block, and the hottest individual cap temperature which I measured was 38°C. The average temp of the caps was room temperature.
And it IS hot here in Israel – I’m not just talking about the fires from the Gazan terrorists.



“For every 10 degree Centigrade decrease in operating temperature, the capacitor life is extended by a factor of two.”

So, if my HE caps are running at a constant 40°C, then their expected lifetime is 5000 x 2 to the power of 6 (=64, check my math) = 320,000 hrs = more than 36 years; scope on all the time, never being switched off.

Given that the data supplied above stated that cap’s casing would breakdown after about 15 years, then there is no apparent benefit HERE of choosing a 10000 hr cap over a 5000 hr, except for the status.

So, IMO what we should be concerned about, is WHAT does the waveform look like at all the HIGH-RIPPLE positions in this Tek SMPS PSU circuit, and then select the cap which dampens that ripple most effectively. To my experience, the Nichicon HE fits the bill most effectively.

Menahem