Radford SPA.50 guru required ;-)

[GW4FRX]

I'm in the process of restoring a much-mistreated Radford SPA.50 and have found some anomalies on the schematic for the M.3297 driver board. The highest resistor number is R29 but both boards additionally carry R30 (2.2R between negative end of C3 and ground), R31 (47R between Ts1 emitter and C4) and R32. This latter is between Ts2 base and collector and in my amplifier the values are different -- 120K on one and 150K on the other.

Looking at the schematic for the later SPA.60 there is a preset in this position. In the even later HD250 service manual this corresponds to P6, adjusted to ‘minimize crossover spikes’.

Can anyone shed any light on this? In particular, was the value of R32 adjusted on test and how should it be adjusted if semiconductors are changed?

(I've also posted this to the Radford Yahoo group but there's very little activity on there.)

[GrimJosef]

I'd recommend you contact Radford Revival http://www.radfordrevival.co.uk/. Not only do they hold, as far as I know, the remaining technical literature from Radford Electronics but Steve and Will are also both enthusiasts and electronics specialists.

Cheers,

GJ

[Ted Kendall]

Mike Solomons at London Sound put in the hard yards on these...

[GW4FRX]

Originally Posted by GrimJosef

I'd recommend you contact Radford Revival http://www.radfordrevival.co.uk/. Not only do they hold, as far as I know, the remaining technical literature from Radford Electronics but Steve and Will are also both enthusiasts and electronics specialists.

I've had some dealings with Will and he's certainly very knowledgeable but they seem to specialise in the thermionic Radfords. Certainly worth a try though.

[Trevor]

i have an sp50 or should i say sp60 in my collection it was purchased as a 50 but after opening it up i found it to be fitted with 2 x sp60 modules and a derated power supply
i corrected the supply and then found a channel to be noisy be aware the correct transistors for these are hard to obtain and are very expensive
however having said this the amplifier sounds very nice
you will need low distortion oscillator distortion analyser and a scope in order to fix this correctly
regards Trev

[GW4FRX]

Originally Posted by Trevor

i have an sp50 or should i say sp60 in my collection it was purchased as a 50 but after opening it up i found it to be fitted with 2 x sp60 modules and a derated power supply...

Whatever the virtues of Mr Radford as an amplifier designer, his version control doesn't seem to have been very stringent. In my sample there are three additional resistors on the PCB that aren't on the schematic, not to mention five others that are nowhere near the published values. The resistor chain from the collector of Ts2 to ground where P2 sets the quiescent current is a case in point. The chain totals 7K4 on the schematic but 31K4 with the actual values. By the look of the PCB these were all factory-fitted components, not later replacements.

Two capacitors have different values from those stated and there's an extra that's not shown on the schematic. Ts3 is slated as a BC108 but is a BC107 on both boards. And so on.

No doubt Radford policy was one of continuous improvement. If only they'd documented the improvements as well...

[Trevor]

i would tend to agree with you Arthur Bailey was the designer look up the 30 watt amplifier world class in its time
Very much a cottage industry more so the transistor designs most of the exotic devices were of RCA manufacture
My Radfords I would describe as well home made!
Not in the same class as Leak Quad etc this comment does not apply to the valve designs

Regards Trev

[Edward Huggins]

I still have difficulty in even associating Radfords with Semiconductors....

[GW4FRX]

Originally Posted by Trevor

Not in the same class as Leak Quad etc this comment does not apply to the valve designs

That's perhaps a little unfair. Contemporary Leak transistor amplifiers had several design shortcomings and were considerably less well screwed together than an SPA.50. I don't know the Quad 33/303 very well but they don't seem to have a stellar reputation for reliability.

[Trevor]

Facts speak for themselves I used to retail all these from the late 60s through to the nineties Leak Quad Armstrong Rogers Radford Exposure Crimson Hitachi Trio etc
And while the Radford figures were very good the actual build quality can be described as woeful just look at the wiring gauge
by modern standards very poor
But to be honest non of them can be described as of military build

Trev

[Trevor]

I can remember receiving a Radford sca 30 1969 the boards were warped ie not flat like bananas in fact
On reflection i think the best build quality was Rogers

Trev

[Ted Kendall]

Originally Posted by GW4FRX

I don't know the Quad 33/303 very well but they don't seem to have a stellar reputation for reliability.

You having a laugh, sir? There are probably over half the total production of 33/303s still running sweetly today after three or four times their design life of service. The reason they turn up here so often is that there's hordes of the things about - and a jolly good thing too. Their faults are few and well documented, and the electronic and mechanical design was well ahead of most contemporary units - it doesn't look too shabby even today.

[Trevor]

Well said sir

[GW4FRX]

You having a laugh, sir?

No, but I recall from my BBC days in the 1970s that the 303s in the radio continuity studios in BH used to fail with some regularity. I also vaguely recall some problems with them in Bush House. But as I said, I'm not particularly familiar with them and it was a long time ago. We still had valves in studios back then...

[GrimJosef]

If I remember correctly there was a failure mode associated with the main electrolytics oozing liquid under thermal cycling and the liquid then dripping onto the pcb and corroding it. The solution (no pun intended) was to invert the capacitors. That generally stopped the oozing, and even if it did occur the liquid would tend to sit as a droplet on top of the capacitor where it did relatively little harm. I don't remember at what stage Quad discovered this fix though. I recall seeing 303s with the original factory-fitted capacitors already in the 'upside-down' orientation.

Cheers,

GJ

[Ted Kendall]

I think that penny dropped fairly early on. In ten years in the Beeb, during which I used and saw used Quad 303s and 405s for everything save banging in nails, I only had one failure, and that was an AM8/16, which was of course based on the 405.

This was a bit awkward, I will concede. In the middle of balancing a session for Listen To The Band, I lost the tweeter in the right-hand LS5/8. I take some pride in the fact that the remainder of the session was useable. The next morning I got on the 'phone to ED and ordered another 8/16, as I wasn't going to risk that happening again without a spare to hand. Did my manager chew me out for that! But this was in the reality-defying days of the short-lived North East Region, which thought that champagne on beer budget could be achieved by running not so much hot as actually smoking...