Grampian ENSA UA

[Doris The Diode]

Hi everyone,

I am repairing a Grampian ENSA UA which belongs to a friend of mine and his father, who is 92, used to use during his time in the forces.

It is in very good condition and he has asked me to repair it for his father's 93rd birthday so that he can show it to him working once again.

So far one of the rectifier valves arcs over so I have replaced that and I will be replacing all the capacitors. Two resistors in the PA cathode were out of spec so those will be replaced too.

His main concern and mine is how do I make this unit safe so that no one gets a belt from it when handling the microphone?
The shield of the mic goes to chasis and then via a 0.005uF cap to the nuetral rail.

My question is, can I ground this shield point with mains earth?

I've been reading up and I see that an isolating transformer is the real answer to this.
If I were to use, let's say two transformers back to back, would now connecting the mic shield to mains earth be the right way?
What else would I need to do or connect to earth etc?



All the best,
DD

[PJL]

The chassis and mic are isolated by the .005uF cap and the matching transformer in this design but the PU input isn't.

Use an isolating transformer, connect earth to chassis and short out the 0.005uF capacitor. You need to make sure it is never used without the isolating transformer!

[Doris The Diode]

PJL,

Thank you for your reply, I've redrawn the diagram and have included the mains isolation transformer showing the earth connections as you described above.

I've had another thought, since the drop across the heaters is 113V I intend to get a 240V-110V transformer which will power the heaters separately, eliminating the dropper resistor and will not waste the 26W dropped by it.

Now to find that isolation transformer!

Cheers,
DD

[Nickthedentist]

Or did they mean mains isolation transfomer. These aren't hard to find secondhand.

[PJL]

The drawing shows a mains isolation transformer. Using a separate step-down transformer for the heaters sounds like a good idea.

[Doris The Diode]

Yes, it's my drawing and I haven't yet included to step down transformer for the heaters.

I've started the repair work on it doing it section by section, rectifier, main amp, pre amp.
One of the rectifier valves was arcing as I increased the voltage on the variac so just got a new one and all is ok, 250V DC is being delivered to a 3K load for test.

The 75R resistors were replaced by 100R as the capacitor was not the original 16uF but instead I put in a 22uF as closest value so with a bit of interpelation the value came 93R, 100R being the closest.

DD

Will report on further progress.

[Nickthedentist]

Originally Posted by PJL

The drawing shows a mains isolation transformer. Using a separate step-down transformer for the heaters sounds like a good idea.

Sorry, I missed it, tucked away down at the bottom.

[Doris The Diode]

Update,

I've replaced the cathode resistors and capacitors.
I have isolated the pre-amp and injected a 400Hz signal and all is well, the PA works and the voltages on the cathodes are identical so the vlaves are drawing equal current, 48mA each to be precise.

I've run into a problem though, the set up did not come with a mic lead and perhaps someone on this forum could suggest where I could source the three pin plugs or even two pin plugs which are found on the speakers as they fit into the mic socket too.
Any alternative will suffice.

Cheers,
DD

[Doris The Diode]

Amp is all working now

My next project is to add an isolation transformer, which I have done and run into a problem, it trips the circuit breaker in the house.

Attached is the diagram of the amp and my isolation transformers.

The first one drops the voltage for the heaters to 120V which works well and also means that I don't have to waste power in the dropper resistor.
The another transformer is added to bring the voltage up for HT. In this case, due to wastage, 225V comes out, which is good enough to power the unit.

The heaters are totally disconnected from the circuit and only go to the 120V on the isolation transformer.

With the earth connected as suggested it trips out the circuit breaker in the house.

Any thoughts anyone?

Cheers,
DD

[dave cox]

What sort of trip device ?
For example RCD or 'earth leakage' detector.

dc

[PJL]

Are you saying it does not trip if the earth is disconnected?

I would not let the heater circuit float like that and would leave the connection to chassis and just earth the chassis. Having said that it should not trip an RCD so it may be the first transformer is faulty or unsuitable (inter-winding capacitance too high).

The other possibility is that the surge current from the cold heater chain is the issue.

[Doris The Diode]

Dave,

I just checked and it is RCD.

I'll do some investigating like disconnect the first transformer and just put a load resistor across it and see what happens.

The transformers are from that large country in the East so it wouldn't surprise me if the transforfmer is of poorer quality.

If it is the surge current then I'll have to make a delay which after 30 sec or so shorts a dropper resistor in the 120 line.

DD

[Doris The Diode]

Oh well, it looks like the transformers from a far off country are leaky/badly made as I put a 3K resistor on the end of the second transformer and grounded one end as per circuit diagram and it tripped

Back to the drawing board!

DD

[Doris The Diode]

PJL

It does not trip when the earth is disconnected.

DD

[TrevorG3VLF]

The second transformer has a large DC current through it. This could put the core into saturation but this should not cause an earth leakage

I would put a bridge rectifier after the second transformer and feed the set HT with DC.

[Doris The Diode]

Trevor,

Could you explain the DC through the second transformer and why it would go into saturation?

DD

[Radio Wrangler]

Magnetic cores like the iron laminations of a transformer have a 'hysteresis curve'. If you plot the amount of magnetic force applied (MMF in ampere-turns) an the magnetic flux that it creates on a graph, the operation of the core at low levels goes backwards and forwards on the same line when handling AC. Swinging as far in one direction as in the opposite direction.

With larger peak drive levels the operating point goes up the curve far enough to 'permanently' magnetise the core a bit, and it will then return down a slightly offset curve, and not passing through the zero, zero point. There is now residual magnetism. The same happens in the opposite direction on the other half of the cycle and the previous residual magnetisation is undone and over-written in the opposite direction. So the operating point no longer reciprocates along a single line, it goes around a sort of double-S shape.

With larger still drive levels, the split between the two directions of the curve gets bigger and towards the high ends get flatter. The reduced slope of the curves are the onset of saturation where the core doesn't support much more magnetisation, and so the inductance of the winding falls. The fall in inductance means that the current taken by a transformer increases. The onset of saturation can be fairly abrupt and can blow fuses, trip trips etc with a simple current surge.

That's all for an AC transformer driving a nice AC load.

Now, a simple single-phase rectifier like that ENSA amplifier has passes current in only one direction, and blocks everything in the reverse direction. So the operating point of the transformer core moves of to one side and then returns to not quite zero magnetisation for the unused half-cycle. On the next conduction cycle, it starts with a bit of magnetisation, so it climbs further up the curve and gets higher. On the way back, it finishes with an extra bit of residual magnetisation, and so it builds up and up. So transformers powering half-wave rectifiers never get the swing in the reverse direction which keeps resetting their residual magnetisation, and so residual magnetisation pushes the peak operating point further towards saturation, and heavy saturation pops fuses.

With a half-wave rectifier, you need a much heftier transformer than you would with a full-wave rectifier, to keep it out of saturation.

However, saturation would give you excess line to neutral current. It might blow a fuse or trip a current sensing circuit breaker, but there should be no current to earth that would trip an RCD (used to be called an Earth Leakage Circuit Breaker, but they got re-named because the old name mase sense!)

David

[Hartley118]

I have a 300W toroidal transformer feeding the low-voltage lighting in my workshop which persistently used to trip its 15A RCD. This suggests:

1. A transformer can be driven into saturation on switch-on even though it isn't feeding a rectifier. I think it depends on the remanent core magnetisation from when it was last switched off.
2. An RCD can be tripped not only by earth leakage but also by a load current surge.

In my case, the problem was cured by fitting a suitably rated NTC thermistor permanently in series with the transformer primary. It of course gets warm in use, but if suitably mounted away from cables etc, that's not a problem.

Martin

[Lucien Nunes]

RCDs do not normally trip on transformer inrush. The only thing that should be able to cause an RCD to trip here would be faulty insulation of the first step-down transformer primary, so I would do a high-voltage insulation test if you have a suitable tester available. Even though the heaters are floating there would be enough emission current from heater to cathode, despite good cathode insulation, to trip an RCD if the secondary of the stepdown transformer were not properly insulated from the primary. Of course if there is a serious insulation defect it will show on a multimeter. Hmm, you have checked that they haven't sent you an autotransformer haven't you? Many commercial 230-110 stepdowns are auto, except for those specially designed for safety isolation to BSEN 61558 etc.

Is the RCD that tripped an overall one for the house (or at least a large group of circuits?) There can be situations where the aggregate leakage of all the appliances connected to those circuits is just below the trip threshold. Of course this setup shouldn't cause significant leakage, I don't think inter-winding capacitance is a significant factor. I have run literally hundreds of mains transformers of various shapes and sizes from a single 30mA RCD. Be sure to distinguish between RCDs and RCBOs which combine the functions of RCD and MCB in one unit. RCBOs will only be protecting a single circuit in the distribution board, but can be tripped by inrush current just like an MCB.

[Doris The Diode]

Thank you all for your replies.

The AC theory on transformers put a smile on my face and reminded me of when I was back in class doing my City & Guilds i Radio and Television servicing, an HNC of not much use in today's society of throw away radios and tv sets but keeps me well with restoring valve radios

I was asking actually where DC came into this and perhaps Trevor meant AC all along in his answer.

Thanks to Lucien, I have discovered that the large country from the East has sent me autotransformers, boooo!
This is certainly the cause of the problem, I didn't think of this and well done Lucien.
The secondary winding has 8.5 Ohms but one lead has 0.4 Ohms (what I get when I short my leads) to nuetral so there it is.

I've sourced a 'real' 150VA (the ENSA draw 75W) isolation transformer manufactured by Carroll & Meynell, class B insulation and class 1 isolation and the title does read 'Mains isolation transofmer'.

I'm going to put the circuit back as is was, connect up the heater chain to its origal setup and just stick the isolation transformer across where the old mains wiring was.
The earth will be connected to where the nuetral rail once was.

DD