View Single Post
Old 17th Jan 2008, 8:47 pm   #16
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Quad 2 Valve power amp questions

Quote:
Originally Posted by mastermanx2001 View Post
I see the HT centre tap to earth, but cannot see it on the transformer tag strip (see photo). I presume that it must be connected internally
Right again Paul, the E terminal on the mains side of the transformer tagboard is where the centre tap for both the HT and the 6.3V heaters is brought out. I'm trying to think if it would still be safe to put the additional fuse in this line. In principle the heaters should only induce a very small current in it since they are balanced. But what happens if one half of the GZ32 goes short ? Does this blow the fuse ? If it does then are the heaters now connected, through the short, to whatever charge is still left in the smoothing capacitor ? Maybe this scenario is way too far-fetched to worry about and maybe various other things save you even if it does happen

The idea of fusing the HT is a good one though. So good that Quad actually did it in versions of the Quad II they supplied to Rediffusion. I've never seen one in the flesh but there's a photo on this page http://www.geocities.com/ResearchTri...diiwipmar.html Apparently it was a 500mA fuse in the HT, presumably between the rectifier cathode and the smoothing cap or between the cap and the output transformer/choke.

Some people are so concerned about meltdown that they add a solid-state circuit to monitor the cathode voltage and, if it rises, switch the power off via a mains relay. Details are on this very long page http://www.turneraudio.com.au/quad2powerampmods.html This seems a bit OTT to me, but can't hurt I guess.

The significance of the cathode resistor voltage is a) that all the KT66 current (anode plus screen for both valves) passes through it and b) that it tells you, more or less, what the grid bias is. In your case the total current is 25V divided by 193 ohms i.e. 130mA. According to the circuit it should be 144mA total so the KT66s are running a little light. Assuming the grids are close to 0V (it's important that they are) the valves' bias voltage will be about 27V because there will be 2V or so dropped across the cathode windings of the output transformer which you have to add to the 25V. So it sounds like your KT66s have 305V screen-to-cathode and 310V anode-to-cathode and they're drawing 65mA total current each. The graph at the top of p7 of the KT66 data sheet http://www.dc-daylight.ltd.uk/Valve-...a/KT66-GEC.pdf lets you check these numbers against what a new valve should do. If you run your finger along the bottom to about 310V and then run it up to cross the different grid voltage curves you'll see that you cross the 30V bias curve at about 38mA and the 20V curve just off the top of the graph, let's say 104mA. So with 27V bias you'd expect the current to be 38mA +(104-38)mA x 3V/10V which comes to 58mA. So your KT66s measure up rather well against the GEC data sheet and it seems the Quad numbers are optimistic. The reason your total current is a bit low is the low HT voltage (can you try another rectifier ?). It's not anything wrong with the KT66s themselves.

I have to say in my experience the Quad circuit currents are about right but the HT voltages are usually higher than your 340V/330V by at least 20V. This makes the KT66s currents and bias voltages consistent with the GEC data sheet and it's just the Quad circuit HT voltages which are awry. Has anyone got any more info about this ?

If you want you could repeat these sums for the 7581s and compare with their data sheet. I'd do it but my wife's just got home with the week's shopping and if I leave her to unload the car on her own in the rain the consequences will be too awful to contemplate.

One last thing. There may not be any significance in the different HT voltages you measure with the KT66s and the 7581s. In my house mains voltage fluctuations routinely cause bigger changes than this.

Cheers,

GJ

Last edited by GrimJosef; 17th Jan 2008 at 8:55 pm.
GrimJosef is online now