advertisement


Quad II work

I've tested both amps now and they behave the same...which is great :) no, I mean it, consistently is the name of the game.

I have swapped the EF86s and the GZ32 between amps and still the same behaviour. Bulb lights, bright, voltage over the PSU starts to climb to about 240VDC over 10secs, then begins to fall back.

So, I think it is the bulb...I think like this...

With just the GZ32 installed and I get a good voltage out of the PSU. I add the other valves and the voltage begins to come up, valves begin to warm, pull more current, bulb restricts too much, and the voltage collapses. The bulb stays bright all the time.

If there was a short I wouldn't get up to 240VDC out of the PSU.

I have a halogen, 60W equivalent in the tester. Says it is 42W. I guess I should go for a 60W real, coal fueled bulb.

Does my thinking feel right?
 
Check the soldering. How much soldering experience do you have? Decades? Did the amps work before you changed anything?
 
A halogen will still do the trick, it still has a filament that acts as a resistor. Just stay away from modern fluorescent or LED types. But yes, an old school TFL is the trick.
 
A Quad II pulls around a 100 watts so any 60 watt bulb is going to drop a lot of volts. The fact that you get to 240 volts before it drops back tells you that there is no short.

Your bulb tester has done it`s job, apply full mains volts* now but monitor the HT and keep your hand near the off switch.

Check the voltage across R12.

* With all the valves fitted.
 
Barrymagrec said:
A Quad II pulls around a 100 watts so any 60 watt bulb is going to drop a lot of volts. The fact that you get to 240 volts before it drops back tells you that there is no short.

Your bulb tester has done it`s job, apply full mains volts* now but monitor the HT and keep your hand near the off switch.

Check the voltage across R12.

* With all the valves fitted.
Click to expand...
I did try paralleling up a pair of the 42W lights and the HT DC comes up to about 330VDC, valves begin to glow and then the voltage falls back. Which confirms your point that there is no short and full power can be applied...

Exciting times...

Thank you all for the much needed input. As I said my first time with valves and I'm not sure what to expect. The slight thermal "tink" from the rectifier made me jump, just not expecting it.
 
So, HT before the choke is at 367VDC and across R12 27VDC.

After 40 minutes HT is at 358VDC and across R12 26VDC.

Everything seems to be working on this one. I'll soak test it for an hour or so and then put the bottom back on.

Phew!

I have no idea the last time this particular Quad was powered up, but as it is warming up it is filling the room with that old electronic, musty smell.
 
Yes, definitely keep an eye on that! I’m sure they’ll be fine, but I’d definitely go easy on them until you are absolutely certain all is as it should be, all voltages correct etc. I’d not go for any really long listening sessions until you are certain all is ok. I started my Leaks with a 15 minute listen and check, then 30 minutes, then a hour etc.

51479312579_11c273acf2_b.jpg


I have one of these IR thermometer things (eBay) so I can easily keep an eye on transformer temperature etc. Very useful for valve amps IMO.
 
This site contains affiliate links for which pink fish media may be compensated.
Zombie said:
And soon black tar will flow out from the PSU xformer and ruin the surface
Click to expand...

I've read horror stories of the dripping of tar :)

When I replaced the PSU smoothing caps with a little box from Keith Snook the bottom of the transformers looked like this:

PXL_20211217_192929708 by Garf Arf, on Flickr

PXL_20211217_181336376 by Garf Arf, on Flickr

Keith's replacements are quite neat.
https://keith-snook.info/quad-ii-block-capacitor.html
2021-12-18_11-36-18 by Garf Arf, on Flickr

Had to pop out for my Christmas hair cut, so soak testing on hold atm.

I'll break out the IR thermometer that I use on my Pass labs to check temps. R12 feels hot from a distance.

2021-12-18_10-10-34 by Garf Arf, on Flickr

I'm waiting on a couple of fork crimps to complete the earth for the mains leads, hence the flying lead. Which leads me to thinking about unhooking the chassis earth from the signal gnd, using a 10R resistor between them.

Valves wise I bought a pair of Harma GZ32 rectifiers, four Mullard NOS EF86s (solid plate) and a matched quad of GEC replica KT66s from Watford valves (https://www.watfordvalves.com/product_detail.asp?id=5942)

I have some posh KT66s on my birthday list.
 
Last edited:
Just spent a happy hour or so in the company of these Quads and they do sound kinda lovely; certainly not the tone I was expecting, much more rhythmic and up front, really bring the 57s to life.

A couple of glow in the dark pictures...
PXL_20211219_185356911 by Garf Arf, on Flickr

PXL_20211219_185336774 by Garf Arf, on Flickr

I'll pop them back on the bench tomorrow to run another soak test to check that everything is still okay.
 
Looks fine in the daylight shots, the dark shot does admittedly look slightly worrying (but don't panic, it might just be near IR pickup by the camera), my rule of thumb for a superficial check is if I can't see the anode glowing by eye in a dark room then it's fine or at least not an emergency.

If you can actually see it by eye in the dark then it may in fact be an issue. Might be worth checking some voltages and currents. Given the fact that you can't see it on the daylight shots it's clearly not massively over dissipating but it might be on the high side.
 
After an hour's soak test this afternoon, these are the voltages I measured:

20211220_Voltages by Garf Arf, on Flickr

Looks like my mains is a little high.

I've been reading around on B+ being a little high and it does appear to be quite common, to the point that the voltages on the original cct diagram, cira 1953, may be a little out due to the mains voltage at the time and/or the use of an analogue AVO meter pulling the HT line down.
 
Radford Revival said:
Looks fine in the daylight shots, the dark shot does admittedly look slightly worrying (but don't panic, it might just be near IR pickup by the camera), my rule of thumb for a superficial check is if I can't see the anode glowing by eye in a dark room then it's fine or at least not an emergency.

If you can actually see it by eye in the dark then it may in fact be an issue. Might be worth checking some voltages and currents. Given the fact that you can't see it on the daylight shots it's clearly not massively over dissipating but it might be on the high side.
Click to expand...

I shall double check tomorrow, thank you for the re-assurance.
 
Based on my quick calculations assuming a 180 ohm shared cathode resistor, zero screen current and zero resistance from the cathode windings (which of course will have some) the total dissipation on each valve is around 25W, which is the limit, but will be less once the winding resistances and screen currents are taken into account, so more than likely absolutely fine.

The actual voltage across the 180R cathode resistor will tell most of the story.

Workings assuming zero cathode winding resistance: 180R / 28V = 155mA
Assuming perfect balance, 78mA each
Voltage Anode - Cathode = 360 - 28 = 332V
Power = 332 * 0.078A = 25.9W

In theory that's the absolute worst case, in reality the screen current will be some of that (approx 15mA) and the actual anode current will be somewhat less (around 130mA)
 
I just saw you have in fact noted the cathode resistor voltage which is 26.7V which corresponds to 148mA, assuming around 15mA of that is screen current, the rest being 130mA of anode current is in fact bang on for the Quad schematic. The only thing that seems to differ is your HT seems a little high. Even so it's still within the limits so should be OK.

Another thing I noticed is that your screen voltage is higher than your anode voltage - this is normal in a ultra-linear amp but unless I'm being thick (possible) that shouldn't be the case on a Quad II?
 
Radford Revival said:
I just saw you have in fact noted the cathode resistor voltage which is 26.7V which corresponds to 148mA, assuming around 15mA of that is screen current, the rest being 130mA of anode current is in fact bang on for the Quad schematic. The only thing that seems to differ is your HT seems a little high. Even so it's still within the limits so should be OK.

Another thing I noticed is that your screen voltage is higher than your anode voltage - this is normal in a ultra-linear amp but unless I'm being thick (possible) that shouldn't be the case on a Quad II?
Click to expand...

That's brilliant thank you, I'm in learning by doing and asking people for help.

I was considering replacing R12 with 220R to bring the bias down a little more, but maybe not. Though the more I read the more think that splitting the single cathode resistor/cap network into two networks, one for each KT66. It will still be reversible, just not wholly pretty.
 
You must log in or register to reply here.


advertisement


Back
Top