CATO linestage. Periodic DC offset problem.

[rebojorge]

Hi all. I have been using and enjoying a CATO line stage for a few years now. A few days ago, while listening to some music, really loud noise came out from speakers. Turned everything off and left it alone for the night. Next day tested it at low volumes and at first is all ok, but after a few minutes I get a big thump in one of the channels. I can see the cone sticking out for a second or two and then back in. A few seconds ok and then the thump again. Did not clock the time between thumps but it is periodic.

Any idea on what may be causing the problem?

Thank you all

 

[rebojorge]

And the schematic....

Cato schematic

 

[iDHT]

Hi,

Swap tubes to check if the problem changes to the other channel.

/iDHT

 

[jazbo8]

Are you trying to set a record for necro-posting?!

 

[DF96]

That circuit appears to have borrowed the input LTP from an SS amp, but used triodes instead of BJTs. Unfortunately this circuit is balanced in its BJT form but not in its triode form because BJTs have high collector impedance but triodes have lowish anode impedance. As a result, the current mirror anode load will unbalance the LTP.

We then face the issue that it superficially looks like a balanced input but it is not a balanced input because the feedback will seriously unbalance the input impedances at the two inputs.

Then a simple (and harmless) output coupling cap is replaced by a DC servo - which is possibly the source of the OP's problem.

It looks like it was designed by someone who thinks he understands circuits.

 

[rebojorge]

iDHT - Did that and the problem stays on the same channel, so we can rule out the tube.

DF96 - Failed to mention that I am using the circuit single ended so the -In is shorted to Sgnd.

So... What to try next?

Thanks for the help

 

[ticknpop]

it was originally designed by Erno Borbely and moded by Ron Welbourne - Check both power supplies and servo too especially any electrolytics, the tube seems most likely but if you switched tubes between channels then not likely.

 

[Johan Potgieter]

That circuit appears to have borrowed the input LTP from an SS amp, but used triodes instead of BJTs. Unfortunately this circuit is balanced in its BJT form but not in its triode form because BJTs have high collector impedance but triodes have lowish anode impedance. As a result, the current mirror anode load will unbalance the LTP.

We then face the issue that it superficially looks like a balanced input but it is not a balanced input because the feedback will seriously unbalance the input impedances at the two inputs.

Then a simple (and harmless) output coupling cap is replaced by a DC servo - which is possibly the source of the OP's problem.

It looks like it was designed by someone who thinks he understands circuits.

+1

Sorry to be negative so early in the morning (RSA time).

Regarding your problem, difficult to say without checking point to point. D.c. upset in a d.c. feedback circuit can be a bother to find.

 

[DF96]

Check the capacitors around the DC servo. It may be that the design is finely balanced in stability terms so too much deviation (e.g. due to a value change) could tip it over the edge.

 

[iDHT]

Next I would do..

Resolder the pcb.
Specially solder joints with large thermal changes (tube sockets and so on).
They tend to open (bad contact) by time. This might also be difficult to spot, so it's faster to just solder.

Also the P1 potmeter, could be bad.
Note , disconect your poweramp and speaker if you turn it with power on.

 

[DF96]

Simply 're-soldering' every joint is a good way to introduce new faults.

The DC balance pot P1 might be at fault, but that would be more likely to create random variations not periodic variations.

 

[rebojorge]

Thank you all for your comments. Will do some more digging and get back here.

Cheers