GEC BRT400 power cct. Just curious.

[Gi4CZW Cliff]

Fred, on 80m's tells me of a receiver. with no electrolytic condensers and that power unit smoothing is achieved using an extra valve. So I was interested to see how this was done but internet not finding this circuit. Perhaps I need a correct name to find this circuit diagram. Just me being curious. Thanks in advance. Cliff.

[ms660]

The BAMA site lists a BRT400 manual under GE:

http://bama.edebris.com/manuals/ge/brt400/

Also one here:

http://www.rigpix.com/miscrx/gec_brt400-402_manual.pdf

Lawrence.

[Radio Wrangler]

The AR88 also has no electrolytic capacitors. The main reservoirs are paper as are all the LF ones.

The BRT400 has no electrolytic capacitors and employs a shunt power valve as a hum canceller, whose capacitor coupled grid drives it to try to cancel ripple on the HT rail.

It works to reduce ripple, but it increases the power dissipation in what is already a rather hot running radio.

The BRT400 has a reputation for heat-induced failures AND fro being a right devil to get in to replace components due to the way components are wired on top of others.

A good radio of its era in many respects, with very good sound, but you should understand its downsides if you think of buying one.

David

[Gi4CZW Cliff]

Many thanks Lawrence, I took a good look at the rigpix link and surprised how simple it seems. Not sure of the reason to not have electorlytic's. David, thank you for the information and no, can't buy any more stuff, already too much here.
So I will tell Fred of my findings. thanks again. Cliff.

[Radio Wrangler]

One electrode in an electrolytic capacitor is a conductive , water-based jelly. The actual dielectric is a very thin layer of aluminium oxide on the other electrode. The jelly is needed to conformally cover the dielectric.

Eventually water loss spoils the conductive jelly and air voids appear. Capacitance falls, series resistance grows and the capacitor degrades.

The mechanism has always been known and early electrolytics were known for a short life expectancy which plummeted at higher temperature.

Avoidance of electrolytics was seen as an essential step in trying to make long-life equipment. Paper/oil capacitors were then considered to be immortal!

David

[turretslug]

It's an interesting approach- but perhaps the fact that the BRT400 became noteworthy for employing it was itself an illustration that it was unusual if not quite a dead end. Shunt ripple cancellation makes an appearance every now and then in low-current thermionic circuitry, e.g. the VHF oscillator HT feed in the Gee ground test set but the BRT400 is an unusually high-power application. It's a radio that already takes a fair amount of HT power and the active smoothing takes HT current to around double that of the comparable architecture AR88.

Perhaps GEC were taking things a step too far here- as mentioned, the AR88 manages to be perform perfectly adequately with paper block PSU capacitors and the decade between the design of these sets meant that paper block capacitors had become more compact, so values of, say, 8uF could have been accommodated in the extra space not used by smoothing valve and its huge load resistor- besides running much cooler! Similarly, post-war electrolytics were more reliable and smaller than pre-war types, reducing the argument in favour of avoiding them.

[Radio Wrangler]

It's not just the power dissipated in the anode and it's resistor, there is also the added loading of the heater of that quite powerful valve.

I think owners of BRT400 might do well to improve their capacitors and dump the ripple canceller. It will save appreciable heat in the cabinet, and take some load off the mains transformer.

David

[G6Tanuki]

The BRT400 ripple-canceller always struck me as being the product of designers-with-obsessive-compulsive-disorder and a marketing-department with very narrow sight.

Pre-WWII electrolytics had - it must be accepted - lifespan-issues. But WWII and the early-1950s saw some massive improvements in chemistry. By the time the BRT400 was being designed it was possible to produce electrolytics with in-service reliable lifespans measured in decades, and t relatively low cost.

The BRT400 didn't use electrolytics - but it ran spectacularly hot - which apart from the way it cremates the big dropping-resistors/tagstrip mounted above-chassis, must also have caused thermally-induced frequency-drift issues. Part of me thinks that the increased initial-cost of the 'hum eliminator' valve coupled with the extra electricity-consumption would easily have exceeded the downtime- and repair-cost of using electrolytics and accepting that you may have a few failures after a decade.

GEC advertised the BRT400 as 'being used by the BBC and Goverment agencies'. They could always rely on the taxpayer's pockets to pay.

[Biggles]

I have had a BRT402 in the to-do pile for a number of years now. I gave up on it because it was so difficult to work on. The fault was in the final IF with bandwidth switching into the bargain so not an easy fix. Weighs a ton too.
Alan.

[turretslug]

^^ Wot Tanuki say ^^

If nothing else, it provides an interesting off-shoot that illustrates some of the quandaries of PSU designers when only valves were available- shunt smoothing or regulation could get away with relatively low voltage head-room but was wasteful of HT current and, whilst still needing appreciable heater power, at least didn't need highly insulated heater supplies or cathode/heater. The slightly later Collins R390 took the series regulator approach- presumably the designers realised that they effectively had a blank cheque book and could indulge themselves. Even the anode feeds to audio and line driver valves were included, involving 2x 6082 double power triode series pass valves with no less than 30W of combined heater power plus ISTR 300V in to provide 180V HT out- loadsa heat in a very busy chassis! The later R390A was a comprehensively value-engineered and more maintenance-friendly device (apparently even the '50s US military jibbed at the price of the original version) and adopted conventional choke-assisted smoothing with the by-then time-honoured gas-filled shunt stabiliser for a few selected circuits such as oscillators.

Perhaps both BRT400 and R390 PSUs are both best regarded as examples of losing perspective of what was actually really needed, or of the law of diminishing returns.