Newbie surge resistor question

[smeerkat]

Hello Everyone, I am renovating a Cossor 464 and have replaced the rectifier valve (6X5G) with a NOS item, as well as having the mains transformer rewound. In order to protect the power supply side of the radio, I thought it might be a good idea to fit a surge resistor in series with the rectifier cathode and C24. I assume that I will need a fairly low value resistor (around 100 ohms) but I have no idea how big a surge current it might encounter, in order to specify a power rating. Would a wire-wound resistor with a 5 W rating be OK? Or is this way over the top?

I have inlcuded the relevant part of the Trader Sheet 775 circuit diagram.

TIA, David

[GrimJosef]

The answer depends on what you're trying to protect.

If you just want to protect the 6X5G then you can start from the valve's data sheet http://www.r-type.org/pdfs/6x5g.pdf. This tells you that C24 mustn't be larger than 32uF (I take it it isn't) and that you need 150ohms in series with each anode for protection. This resistance will be made up of

i) the resistance of half the HT winding on the mains transformer

ii) the resistance of the primary winding on the mains transformer multiplied by the turns ratio between half-HT secondary and the primary squared (so if the HT is, say 350-0-350V RMS and the primary is 240V RMS you multiply by (350/240)^2 = 2.1 - actually perhaps 10-20% more as the HT voltage will be the value under load, so not strictly the mains multiplied by the turns ratio)

iii) whatever resistor you add in series with each 6X5G anode, or with the cathode.

The total resistance, which should be 150ohms, is i) + ii) + iii).

Working out the power dissipation in the added resistor isn't straightforward. It will be more than the average HT current squared multiplied by the resistance because before C24 the current will be pulsed, not DC. How much more depends on how pulsed it is (the shorter the pulses, the higher the average power).

If you can measure, or look up, all the numbers you need then you can enter them into the power supply design software psud2 available for free from the Duncan amps website. I've found this to be a very effective way of seeing what different series resistors do and how hot they might get without actually having to build the circuitry and test it. But if you have most of it built already then it might be as easy to drop a few test resistors in and see what happens.

Cheers,

GJ

[Bazz4CQJ]

I'm not familiar with this particular radio, but I suspect that you don't need a surge limiter. You have simply replaced one old valve rectifier with a new valve rectifier; these are fairly slow to warm up the voltage and current rise are similarly slow.

If you had replaced the valve with a silicon diode rectifier, then that would instantaneously conduct, so adding a surge resistor there would be a good idea.

The only exception might be that if the radio was switched off and then switched on again very quickly (while the rectifier is still conducting), that could allow a surge.

B

[smeerkat]

Thanks very much for your helpful replies. I will have a bash at calculating a suitable spec for the resistor. It is certainly not straight forward, so I will do a few cautious tests as suggested.

My main concern was to try to protect the new-old 6X5G and also the transformer. The valves appear to be quite scarce and the transformer was freshly re-wound for me by Ed Dinning and I don't want to risk any damage!

Regards,

David

[Silicon]

A resistor between each anode and the HT winding will provide some degree of protection if there is a short between anode to anode or cathode to heater.
These type of shorts could damage the transformer windings.

I think anode resistors potentially provide more protection than a single resistor connected to the cathode.

[crackle]

In normal operation the rectifier will warm up slowly and the HT current will rise slowly to maximum. (no surge created)
The only time a surge limiter would be needed in this circuit is if you intended to repeatedly keep turning the radio on and off.
The mains transformer is not so likely to be damaged by these surges, it is more likely to be damaged by sustained over current or green spot corrosion.
Mike

[Argus25]

Analog power supplies with power transformers generally do not have serious issues with turn on current surges. When you switch them on the transformer is transiently overloaded by the low resistance of the heater windings, the secondary voltages rise slowly. Also with a valve rectifier they softly charge the reservoir capacitor as the heater warms up.

Surges are generally only a problem when the internal resistance of the power supply source is very low (for example the household mains supply) where you use that directly. This is why you need surge limiters in transformer-less AC/DC sets (examples being Brimistors in the series heater chain) or in switcmode PSU's where the rectified mains power charges large filter capacitors via a silicon bridge rectifier.

As noted if you are worried about high peak current pulses from the 6X5's cathode, don't exceed the recommended max filter cap value or install some additional plate resistance, if the transformer's secondary resistance (plus the primary resistance transformed into the secondary by the impedance ratio) is not adequate.

In cases where you are dealing with some peak current via a resistor, dissipating heat, and you are not sure of the exact form of the current pulse for integration and rms calculation, to know the exact power (and you don't have a correctly reading RMS voltage or current meter) there is a way to find out what the resistor's dissipation is.

Simply measure the resistor's temperature rise above ambient with a temp probe, then put the same resistor on a variable DC power supply and get it to the same temperature and you can then easily calculate the power from the DC current and resistor value (I^2R) or the voltage drop an the resistor value (V^2/R). This technique is called thermal matching.

[crackle]

There are lots of 6X5GT valves on ebay, they are a later version and don't seem to be rare.

Mike

[snowman_al]

My 5 pence worth is if you have fixed the cause of the original failure (most likely leaking coupling cap(s) and electrolytics) and have checked the HT current is now in spec., then do not bother.

If you 'must' have a safety measure, why not fit a 6.3v (or 12 volt) 0.3A bulb in the centre tap to ground line? If you do get on over current situation again the bulb either lights brightly or pops...

[paulsherwin]

I agree with everybody else. It's in the nature of valve rectifiers to soft start, even the ones with directly heated cathodes. You only have to worry about limiting resistors when using silicon rectifiers, as they hard start and have very low internal resistance.

[ms660]

So far as I understand it, with valve rectifiers the series resistance (whether by transformer and or resistor) is there to ensure that the valves maximum hot switching current isn't exceeded, if that's exceeded things can go bad such as arc back which can result in excessive AC across the reservoir capacitor etc).....If unchecked that could result in free Xmas decorations complete with sound effects.

The magic formula for the series resistance presented by the transformer is:

R=Rs+Nsquared*Rp

Where R is the resistance presented, Rs is the transformers secondary resistance, Nsquared is the turns ratio (Vsec./Vpri.) and Rp is the transformers primary resistance.

For bi-phase full wave, Rs is one half of the transformers total secondary resistance.

Lawrence.

[Bazz4CQJ]

I've been asking Mr Google for his thoughts on this, which brought me back to some earlier threads on this forum.

I'm increasingly getting the impression that the reason why some people add resistors in the anode lines of valve rectifiers is more related to protecting the transformer in the event of a worst case failure of the rectifier. If the valve starts flashing over between the two anodes, or there's a short between anode and cathode, both might cause damage to the transformer.

Adding the resistors can help to stop that, and there are also references to "suicide resistors" which have power ratings just adequate for normal no-fault use, but will quickly fry, perhaps going open circuit if overloaded. They are acting akin to a thermal fuse.

So, in the absence of a need to be able to do "hot switching" of the HT, adding those resistor may be worthwhile, but their primary benefit is to protect the transformer from rectifier failure rather than provide surge suppression, though that too may be a secondary benefit?

The radio worked well for some decades without those resistors, but adding them may be regarded as helping to conserve it in its old age.


B

[Graham G3ZVT]

Also I feel that it's wise not to use the maximum value of reservoir capacitor as stated in the data sheet, on an old rectifier with likely a delicate cathode.

[smeerkat]

Thanks once again for all these words of wisdom. On balance it sounds as though I am worrying too much here - having replaced all the electrolytic smoothing caps with new 10 microFarad ones and also having replaced the HT smoothing resistor R17, I think from what I am hearing things should be OK.

I will take some time to reflect on all your comments and decide whether to do further tests, once I have got the set fully working.

Regards,

David

[Craig Sawyers]

Originally Posted by Bazz4CQJ

Adding the resistors can help to stop that, and there are also references to "suicide resistors" which have power ratings just adequate for normal no-fault use, but will quickly fry, perhaps going open circuit if overloaded. They are acting akin to a thermal fuse.
B

There is a class of component called a fusible resistor. They are often used (by at least Fluke) as part of the input circuit protection of DVM's.

It is not really that relevant to the question, the soft start nature of valve rectifiers being the important observation - a point already made.

[paulsherwin]

Fitting the resistor as you suggest won't do any harm apart from reducing the HT voltage a bit, and it might have a small protective effect. Ultimately it's your call.