u67 PSU electrolytics?

[Greenie]

I am looking for 25uf 350v/ 500uf 35v electrolytics. Seems difficult to get. Any advice?

 

[Khron]

Try 22uF or 33uF / 350v or 400v, and 470uF or 560uF / 35v.

Back in those days, they hadn't yet standardized on the preferred number series in use nowadays.

https://en.wikipedia.org/wiki/E_series_of_preferred_numbers#Examples
Electrolytics you'll most often find in E3 or E6 series of values.

 

[rock soderstrom]

Try 22uF or 33uF / 350v or 400v, and 470uF or 560uF / 35v.

Back in those days, they hadn't yet standardized on the preferred number series in use nowadays.

https://en.wikipedia.org/wiki/E_series_of_preferred_numbers#Examples
Electrolytics you'll most often find in E3 or E6 series of values.

You right and one must not forget the tolerances of the time. The uF values are therefore not set in stone, best to take a little more if the space is there. But this is usually the case, because today's modern electrolytic capacitors are much smaller for the same values.

 

[Greenie]

Thanks guys.

 

[Khron]

You're welcome

 

[Tubetec]

Recapping a vintage psu is one thing,
I never use less than a 100uF for HT smoothing caps now , its a false economy ,

 

[soapfoot]

Recapping a vintage psu is one thing,
I never use less than a 100uF for HT smoothing caps now , its a false economy ,

Obviously this idea should not apply to power supplies with vacuum tube rectifiers.

There we must consult the data sheet and not exceed the maximum allowed value (I know you know this; just clarifying for any onlookers)

 

[Greenie]

Recapping a vintage psu is one thing,
I never use less than a 100uF for HT smoothing caps now , its a false economy ,

So even when 25uf is called for, then you would use 100uf?

 

[Tubetec]

A 25uf cap from 60 or 70 years ago is huge , a modern cap of at least 10 times that value fits in a smaller can nowadays .
I often use caps of 220uf or more in tube PSU 's , in conjunction with the usual series resistances of 5-10kohms it means the HT circuit has a ramp up time of several minutes .
As was previously mentioned tube rectifiers have a max value for the first filter cap , which you dont want to exceed .

 

[rock soderstrom]

So even when 25uf is called for, then you would use 100uf?

A 25uf cap from 60 or 70 years ago is huge , a modern cap of at least 10 times that value fits in a smaller can nowadays .
I often use caps of 220uf or more in tube PSU 's , in conjunction with the usual series resistances of 5-10kohms it means the HT circuit has a ramp up time of several minutes .
As was previously mentioned tube rectifiers have a max value for the first filter cap , which you dont want to exceed

Are we talking about an original vintage U67 PSU, if so what model? Reissue or DIY version?

With the old original PSUs I find it a bit questionable to simply quadruple (or more) all the electrolytic capacitors. Basically a good filtering by a lot of capacitance is a good thing but you should also keep in mind the consequences. The high capacitance stresses not only rectifier tubes but also sand diodes (especially the first generations. Selenium??) and the transformer. If you know what you are doing, this is acceptable (especially for new builds), but just throwing in capacitance carelessly in vintage gear can cause problems. (again "just clarifying for any onlookers")

What is the goal? Better than better?

 

[Greenie]

Are we talking about an original vintage U67 PSU, if so what model? Reissue or DIY version?

With the old original PSUs I find it a bit questionable to simply quadruple (or more) all the electrolytic capacitors. Basically a good filtering by a lot of capacitance is a good thing but you should also keep in mind the consequences. The high capacitance stresses not only rectifier tubes but also sand diodes (especially the first generations. Selenium??) and the transformer. If you know what you are doing, this is acceptable (especially for new builds), but just throwing in capacitance carelessly in vintage gear can cause problems. (Again "just clarifying for any onlookers")

What is the goal? Better than better?

This is a new build. The problem is that some values are hard to come by, while others I have.

 

[Khron]

Some "historical" values, yes.

 

[rock soderstrom]

This is a new build. The problem is that some values are hard to come by, while others I have.

Ok, take what's there, gladly more.

 

[soapfoot]

For what it’s worth—and this is just personal choice—when an old design is quiet and works well, I am much more likely to uprate voltage and temperature specs rather than massively uprating value.

Operating at a substantially derated voltage will tend to dramatically increase lifespan (this is usually shown on the data sheets).

If stock value is 25 @ 300, I might choose a 33 @ 500v part, 105 degree C, 10k hour rated… something like that. Or if there’s room, even going to something like a metallized film cap (which are increasingly available in large values, and will likely never fail or need replacement)

For a working design I’m typically more interested in longevity/reliability.