WAD 300B PP Rebuild Advice Needed.

[ZakB]

Hi All,

After many happy years with a WAD K5881, I've just acquired a WAD 300B PP from a nice chap on this forum .

The amp wasn't working when I bought it and has been extensively modified. I've got it functioning, but I've decided to give it a complete rebuild over Christmas. However, I'd really appreciate the thoughts of those who know these amplifiers on the effect of some of the mods that have been carried out and whether I should incorporate them into the rebuild.

Having made no significant performance measurements yet, the two mods I'm mainly concerned about are shown in red on the attached circuit diagram:

1) The addition of a 15H 30mA Hammond choke to the input stage HT of both channels.

2) The addition of an 820R 50W wirewound resistor to each line of the choke input supply.

Do any of you know what the thinking was behind these mods and if they offer any real improvement? Should I keep them as part of the rebuild?

Your thoughts are much appreciated,

Rgds,

Zak

[IMG]300b by Zak Barrett, on Flickr[/IMG]

[Richard]

Hi Zak, Rather than us trying to second-guess it might be better to ask the chap who sold you the amp

[ZakB]

Morning Richard,

I didn't ask the chap who sold it, as I believe (although I may be wrong) that the mods we are discussing were done prior to his ownership so he may not have a view on their effectiveness.

I was also hoping to draw on the experience on this forum to discuss the engineering thinking behind the mods. I detailed the changes on the circuit diagram attached to the post so hopefully no second guessing is needed.

Rgds,

Zak

[Ali Tait]

As I recall, one good thing to do was to reduce dissipation on the 300b's if not already done.

[ZakB]

Hi Ali,

Thanks for your reply. That's already been done and I'll keep that modification in the rebuild.

Rgds,

Zak

[Richard]

Hi Zak,

Well, the 820R resistors will drop the HT.

The HT is 565V nominal and as standard each pair of 300Bs draws 160mA so they'll drop approx 130V (= 21W) each. There'll be some sea-sawing of the voltages if you're using 1K5 instead of 1K2 at the cathodes and the net plate voltage (HTV less cathode V) will probably end up a little over 300V. Original spec 300B were rated 400V plate and typically 300V or 350V in use. Modern 300B such as Svet were rated 450V. Mesh plates such as TJ Sophia were 300V.

The chokes will have an HT smoothing effect working with C2 of 160/(160+9425) = 0.0167 at 100Hz. Voltage drop will depend on the choke's DCR but certainly lower than using a resistor.

Mine was reliable with modern 300B with just the change of cathode value to 1K5 by snipping out the 6K8. That put plate dissipation of my otherwise standard amp nearer 30W, stopped the red spots, and a set lasted going on a year of regular use. If you have mesh plates or more expensive types you might want to keep the mods as shown and have a listen and check voltages and plate dissipation

[ZakB]

Hi Richard,

Thanks very much for your thoughts.

At the moment the amplifier uses a set of Electro Harmonix 300b, so a maximum specified plate voltage of 450V. The R at the cathodes is currently set to 1K5.

Given your findings, I think I'll remove the 820R resistors for the rebuild. They seem like a fairly agricultural way of reducing the 300b plate voltages as they will also have (i assume) the unintended effect of changing the input stage operating parameters....or am I missing something?

The choke's DCR is low at just over 1K. Without knowing if there is remaining ripple on the input stage HT, i'm not sure if your calculated smoothing effect is needed. Thinking about it, I'll remove this too, make some measurements and listening tests to see if the amp actually benefits from the addition.

A couple of other questions and then I'll leave you alone ;-)!

- The heaters on the input stage have been converted to DC. Do you think this is worthwhile? I ask as I'm assuming that most of the hum in these amplifiers is developed in the 300b's directly heated cathode and that any input stage hum will be relatively negligible? My K5881 is very quiet with AC heated input triodes.

-On the subject of hum, in its current state, I'm measuring 20mV RMS of hum on both channels. This seems very high. I guess directly heated triodes are not quiet, but in your experience, what can I expect to achieve with proper wiring and earthing practice? It would be great to know what to aim for....

Thanks once again for all your advice. It is really much appreciated.

Zak

[John Caswell]

Hi Zak,
20mV rms is somewhat high, but can depend upon the 300Bs. I suspect a rewire will help a lot and careful routing of the wiring.
I was not very impressed with the PCB layout, too much common earthing between channels and think that a rebuild on tagboard/tagstrips will be better.
I have also had problems with this amp and preamps where when both are connected there is a hum, I am sure that this is due to a higher gain than is necessary in the front end.
I have alleviated this by raising the input phono earths with 10R paralleled with a 0.1mF.
Watching this with interest.

John

[Richard]

Hi Zak and John,

Yes I used EH valves too and think you should have no problem reverting the amp to standard with just the cathode R change. Mine was second-hand and came with a few mods which I ended up taking out except the DC heaters on the first valve which I kept.

I didn't like the sound with fb so the wiring and switch were quick to go and a re-wire of the earthing and signal grounds sorted the hum.

20mV hum is far too much and you can hope to get it down to 2mV or less without much work. Check your amp earthing carefully. As received, mine was lifted 10R from mains earth but done the wrong way around; by connecting signal ground to chassis then chassis to mains earth with a 10R resistor. Very bad as that gives too much resistance for the safety mains earth and no protection at all if the 10R resistor fails.

Instead; isolate signal ground inside the chassis (fit insulated phono sockets if it doesn't already have them). Connect mains earth IEC socket to the chassis solidly. Check the connections with a meter then connect signal ground to the chassis via 10R//0.1uF and re-check with the meter.

Star earth the speaker posts, 300B pairs, and the input pcb back to the earthiest point (neg of C10) with separate wires.

Check and connect the 300B heaters in the same phase AC so any hum may cancel in the OPTX. The wires are the same colour so check phasing (valves out and mind the HT) in a pair of valves by connecting a crock clip lead between one heater wire from each valve and measuring the ACV across the other 2 ends. From memory the voltages just sum or cancel, 0 or 10V, so fit the wires the same way to the valves sockets; same phase wires to the same valve pin number in each pair.

Quite a lot about this amp on the forum, here are a couple of threads covering some of this,

6 Aug 2006
http://www.world-designs.co.uk/forum...ead.php?t=1490

15 Mar 2008
http://www.world-designs.co.uk/forum...ead.php?t=4074

[Greg.]

Zak, you will see from the links Richard has posted that together with others, I built a clone of this amp. I was very happy with it for many years although I have now sold it on and use something different. I believe you have bought Alistair’s (Buddam’s) amp, which previously had upgrade (?) attention from David Coe.

With reference to your original questions, your amp is the only one I have seen (and I’ve seen quite a few) with 15H 30ma chokes fitted to the input stage.

I also changed 300B cathode load resistance to 1K5 as part of a scheme to reduce dissipation on the 300B’s. Andy Grove has typically always designed with the output tubes working at maximum dissipation. At 1K2 and with increased HT owing to high mains voltage, I had hot spot plate glow on my EH gold grids. I Also needed to reduce HT as part of this reduced dissipation process because of the high mains (248vac) here so fitted dropper resistors on the HT, although in my case I introduced them after the choke rather than the way you have it. I don’t recall the value I used but they were proportional to accommodate my mains voltage.

I agree with Richard, your hum target should be around 2mv. In addition to what has already been said, consider the age of your GZ37. I got quite a bit of hum when my original was a few years old and swapping it out for a new one was a cure.

I used DC heating for the input and driver valves. A good idea imho. I also preferred battery bias on the 6072, but only with Ni-Cad rechargeable batteries now outlawed. Other and current alternatives (NimH, Lithium etc) brought a sharp glassy edge to the sound.

Here’s a pic of my chassis insides. Batteries are out of view.
https://www.flickr.com/gp/139667940@N02/Q98318

As John has suggested a rebuild on tagstrip, here’s a pic of Tony Moore’s tagstrip built clone (now owned by Nigel (vinylspinner)).
https://www.flickr.com/gp/139667940@N02/FZ1hBA

As an aside, you might be interested in a brand new set of unobtainium Blackgate caps for C4 (2 x 470uF 16v FK type) and C6/7 (4 x 100uF 160v standard type) which I have for sale. They are completely unused and never soldered. PM me if interested.