• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Cary SLA 70

costis_n

Member
2008-08-01 1:18 am
You should place it between the 5AR4 and the 1100μ capacitor (+) terminal. Valve rectifiers cannot handle the high ripple current of such a large capacitor. Was this capacitor there originally, or is it a later modification by a "guru"? I cannot believe that Cary made such a bad design!
 

manner

Member
2010-10-07 3:58 am
Hi Costi-n,
Do you mean here?
Thanks
manner
 

Attachments

  • cary amp.jpg
    cary amp.jpg
    386.9 KB · Views: 563

6A3sUMMER

Member
2016-06-07 6:50 am
Wow!
1100 uF is far too much input capacitance for any tube rectifier that I ever saw in any Hi Fi amp.

The 5AR4 is rated for 60uF maximum, but also has a restriction of needing a series resistance of from 75 Ohms (300Vrms per plate) to 200 Ohms (550Vrms per plate), including the transformer secondary center tap to plate DCR, plus a series resistor to get the resistance listed above.
I learned the hard way when I used a 100uF cap and 5AR4 . . . Arc Away!

Just because you CAN make a circuit work that way, does not mean you should.
And I bet the power transformer runs Really Hot with 1100uF and no series resistor.
Fire anybody?

The 5R4 (2.0 Amps filament) and 5U4 (3.0 Amps filament) are directly heated filament is the "cathode". They have much more space from the filament to the plate than a 5AR4 (1.9 Amp filament that heats a real cathode).
The 5U4 3.0 Amp filament will push the 1.9A filament winding a little too hard.
And look up the 5R4 and 5U4 maximum capacitance, and the required series resistance, they will probably be similar to the 5AR4.
The GZ34 is a 5AR4 (i.e an Amperex data sheet is one and the same).
 
Last edited:
And I bet the power transformer runs Really Hot with 1100uF and no series resistor.
Fire anybody?

Methinks you are exaggerating a tad ;) This cap raises the temperature by how much? Half a degree, or a quarter?

Not arguing the cap eats rectifiers for breakfast. The only sensible way to have this amount of capacitance is not to be a cheapskate and use a choke.
 

6A3sUMMER

Member
2016-06-07 6:50 am
The circuit could do one of these:

Use a 60uF or smaller cap, driving a small choke, driving an 1100uF cap (or even smaller cap because of the additional filtering of the choke). That will give almost as much B+ voltage, and equal or less ripple. The amplifier power will be almost the same.

Use a choke input filter, and the 1100uF cap. That will reduce the B+ voltage, so the amp will have less power.

The (I)squared x R loss of a choke input filter (R = DCR of the B+ secondary, plus DCR of the choke, plus the rectifier R) will almost certainly have a lower integrated power, than the (I)squared x R loss with the 1100uF cap direct from the rectifier.

Notice that for high secondary voltages, the current rating of the 5AR4 with a cap input filter is Significantly less than that it is for a choke input filter:
Secondary 2x500V: cap filter 200mA;
choke filter 250mA Secondary 2x550V: cap filter 160mA;
choke filter 225mA

If it stresses the rectifier, I bet it is stressing the secondary wire by a similar percentage.
 

6A3sUMMER

Member
2016-06-07 6:50 am
metanastis,

You have 4uF input cap right after the rectifier, right?
That seems to be more like it, instead of following the rectifier with 1100uF.
Even compared to a 60uF cap, that is an 18 to 1 overload.

I wonder how many (authorized) factory mods there were over the production years.
Schematics and actual production do not always match.

But I wonder how many un-authorized mods there have been made by owners and by those they asked to do mods for them.
 
If it stresses the rectifier, I bet it is stressing the secondary wire by a similar percentage.

Not really. It is the initial current spike on power on that kills the rectifier. The recharging current pulses are small potatoes in comparison and neither affect the rectifier much, nor cause any significant heating of the transformer. Easy enough to simulate. The secondary winding resistance is more than sufficient in keeping the rectifier safe, that's why delayed power on using a thermistor/ shorted resistor solves the issue.

A choke and separating the cap in two is still a much better solution, only cheap manufacturers are reluctant to spend the cash.
 

emk2

Member
2009-04-16 3:56 pm
I've had Quicksilver "mini mites" with 1000uF input capacitor, PT was quite hot.

With a CLC (50uF > 1.5H > 200uF), PT was a lot cooler.

Methinks you are exaggerating a tad ;) This cap raises the temperature by how much? Half a degree, or a quarter?

Not arguing the cap eats rectifiers for breakfast. The only sensible way to have this amount of capacitance is not to be a cheapskate and use a choke.
 
metanastis,

You have 4uF input cap right after the rectifier, right?
That seems to be more like it, instead of following the rectifier with 1100uF.
Even compared to a 60uF cap, that is an 18 to 1 overload.

I wonder how many (authorized) factory mods there were over the production years.
Schematics and actual production do not always match.

But I wonder how many un-authorized mods there have been made by owners and by those they asked to do mods for them.

I am really sorry mister!!! The 4uF I wrote in the parenthesis is the standard 5R4 filter cap as it is referred in the tube manual.I operate the original circuit with two 1100 uF cap using the 5R4 or GZ34 rectifier tubes without any problems so far.
 
Hi all,


Sorry to resurrect this pretty old thread.


I own a SLA-70 signature, well working but with slight 50Hz hum and some ripple. Not really apparent in normal conditions (you have to put your ear agains the speakers to hear them) but a bit annoying when listening at very very low volumes at night, while I'm working at my desk.
I've tried to reduce the hum by doing some internal rewiring: moving DC lines as far as possible from heaters supply lines and doing tighter twisting on the latters. Unfortunately with no valuable results.

As per the ripple, I'm evaluating the possibility to add a 60uF after the rectifiers, followed by a choke before the big 1000uF capacitor, as suggested by 6A3sUMMER.
This seems the simplest way to go, wihtout too many modifications to the rest of the layout.
Since I don't have much experience on power supply filtering, could you give some hints on the way to realize the choke? A simple LC filter? Which values to use?


Thank you in advance for your suggestions.


Raf