Project: Re-engineered Murphy B40 Receiver

[glowinganode]

Excuse me for asking, but why are you interested in the inductance of the secondary?
Rob.

[radiozero]

Originally Posted by glowinganode

Excuse me for asking, but why are you interested in the inductance of the secondary?
Rob.

Well, the only thing I was really a bit unsure about was the inductance ratio formula. But, it is correct. The reason I mentioned that was because I'm simulating my transformer is LTspice. I'm creating a sort of manual for myself about how to model transformers use LTspice.

For a 1:1 transformer if L1 was primary and L2 secondary both should have the same inductance.

For my actual situation, with my transformer (which is 1:2.4) if we ignore the centre tap, then L2 would be 5.76 times L1.

Now lets take into consideration the centre tap and say we have a full-wave rectifier circuit. We would have L1, L2 and L3. In LTspice I read" But note that the inductance of the the two inductors forming the centre-tapped winding is one forth of the total winding inductance. For example, a 1:1 (total winding) centre tapped RF transformer might be made with inductors of 10uH, 2.5uH, 2,5uH."

Now, I'm not quite sure what is meant. So, if L1 was (say) 1H,does it mean that with my particular transformers inductance ratio, I'd set L2 and L3 to one quarter of 5.76H = 1.44H? I.e, L2=1.44H, L3=1.44H.

[glowinganode]

Ok, don't forget other factors affect the inductance, wire diam and spacing and former diam.
I would have thought winding resistance rather than inductance would be dominant.
Rob.

[Skywave]

Originally Posted by glowinganode

Excuse me for asking, but why are you interested in the inductance of the secondary?
Rob.

Originally Posted by radiozero

For a 1:1 transformer, if L1 was primary and L2 was secondary, both should have the same inductance.

No, it's not as simple as that. For a 1:1 transformation ratio, it's the ratio of the turns that needs to be 1:1, not the inductance. If the turns ratio is indeed 1:1, it does not follow that the ratio of primary inductance to secondary inductance will be 1:1 also.
Iron-cored transformers are quite complicated things to accurately analyse - and thus synthesize for your purpose. The actual inductance of each winding will be influenced by the currents flowing in each. Then there are questions of voltage & current phase relationships to consider, core and hysteresis loss, etc.

Al.

[radiozero]

Well, in my LTspice model of a transformer I do suspect that the inductance values you put in for L1, L2 and L3 (For a CT secondary) may not be so very influential to circuit performance in the secondary circuit, as simulated by LTspice.

However, if it is a matter of going by the book, (which it is at this early stage) keeping things simple, and accepting simulation to be "in the ball-park", I believe L2 and L3 (the two secondary windings) you would set in LTspice, to be 1/4 of what the inductance value is for the whole secondary winding.

So, if the whole of the secondary winding had in fact an inductance value 5.76H, you would set L2 & L3, in LTspice, to 1.44H. That is going by the book, I think. That is what I think the instruction is, for LTspice.

As to a 1:1 transformer:

The thing is,Turns Ratio (TR) is Ns/Np = Es/Ep.

I think the Inductance Ratio (LR) is Lsec/Lpri = (Ns/Np)^2, is mathematically, a correct statement, at one level. I wonder though, if in practice, it's not quite.

In basic theory, if the primary of a transformer was 1H, the secondary would be 1H. As manufactured. But, in operation, will L2, the secondary actually appear to be 1H when operating in a circuit. You note: No.

And I think you may also mean, even in manufacture, the secondary will not exactly be 1H.

[neon indicator]

There is also coupling factor

[Herald1360]

Coupling factor- yes very relevant. With no coupling, two 1H inductors in series will be 2H. However, if they have 100% coupling (roughly what exists if they're each half of the winding in a transformer), the total winding inductance will be 4H.

For a first order approximation in Spice to a transformer, you need to know all the windings' inductances and assume that they all couple 100% with each other. Start adding winding resistances, leakage inductance etc and the model will grow like Topsy!

[Herald1360]

Actually, since you're really only interested in the behaviour of the rectifier and filter circuits, wouldn't it be a lot easier to replace the transformer secondary driving them by 50Hz sine wave sources?

That's what I do unless it's likely that transformer shortcomings may have a significant effect on the results, like for a switchmode design.

[PJL]

Oh, the wonders of modern software tools.

I am not sure you are going to gain much from this exercise as you are unable to define the load which has distinctly non-linear, time dependent characteristics.

There are also other simulation complications such as the valve heater current, which I assume are powered from the same transformer and therefore share the same primary resistance, are also very non-linear and time dependent.

Your options are 1) take the risk that the surge will not cause a problem, largely mitigated by replacement 630V capacitors 2) design a complex solution to delay the application of HT 3) go for a valve rectifier as the designer intended

[neon indicator]

4) Suitable Thermistor across HT rail. A TV Degauss coil Thermistor with a suitable series resistor similar to warmed up HT load will work.

5) SMPSU type series inrush limit Thermistor between rectifier and caps. Sits on HT side of SMPSU normally otherwise the mains fuse blows. A SMPSU has no Transformer resistance and Iron losses to limit inrush. The 400uF 350V cap is connected direct to mains via RFI filter and bridge rectifier, hence the series thermistor.

6) Series or shunt regulator on HT rail..

A Simple Shunt regulator is easy with Transistor or valve. A series regulator is easy with Transistor or IC.

[murphyv310]

A TV Degauss themistor will draw too much even when warm and a series limiting resistor will not allow it to reach operating temperature anyway.

Most SMPS will soft start anyway and being honest I haven't seen an inrush thermistor in one for years

The series shunt regulator is OK but has everyone forgotten the EB91 & Relay that I posted earlier?

[Sean Williams]

I reckon stick with the valve reccy, and then nick some LT from the dial lamps.....

The B40 VFO couldnt exactly be accused of being unstable, or drifty, even in shed cases - just a case of re-inventing the wheel for the sake of doing it possibly?

[neon indicator]

Originally Posted by murphyv310

A TV Degauss themistor will draw too much even when warm and a series limiting resistor will not allow it to reach operating temperature anyway.

Most SMPS will soft start anyway and being honest I haven't seen an inrush thermistor in one for years

The series shunt regulator is OK but has everyone forgotten the EB91 & Relay that I posted earlier?

Degauss Thermistor
Yes, you have to have a small enough series resistor, not too large or indeed it doesn't work. An inrush Thermistor might be a better idea if the valves turn on fast enough, though a suitably chosen shunt thermistor is a primitive regulator if your series resistance suits.

Inrush
Soft start doesn't help. The high power ones with "big" Electrolytic across the bridge do have a series Thermistor. You can buy them at a range of 20C cold resistance for a particular operating current. Not expensive. You may want some value of shunt "bleed" resistor. Always a good idea for safety on valve gear HT capacitors.

http://www.victoryvinny.com/thermistor/thermistor.html

http://www.ametherm.com/inrush-current/

Very simple solution and needed for all high power SMPSU (as soft start doesn't limit inrush to the mains side DC smoothing capacitors at all).

[Socrates]

This is a much later posting. Murphy 310 has asked a question as to why all this bother about transformers..very sobering. I have worked on conversions of military sets on an off since the 1960's. Whereas 400Hz is an issue 60Hz really has little difference seen when used on 50Hz. I'd like to make some more sobering comment. Before getting off with the pixies about modifications and solid state conversions it is wise to know what you are doing as a competent engineer and not as a dreamer.

The engineers who designed these sets designed systems and modules which worked as a team and performed very well. They were not only "tube-competent" but in many cases genius'.

As with 're-engineering" cars...the results are mor often than not half baked tack-ons making the vehicle worse than what was started-with. Whilst one aspect might be improved for some sensible or ridiculous purpose....others suffer, including safet to drive passengers and the public. One finds cars more suited to tracks run in the kinds of conditions in daily use for which they are simply, unsuitable and dangerous. The Ford Cosworth Sierra for ezample of ruining a car through making it unsuitable to wet British roads and today the thousands spent on winding them up to 350-750Hp are a dead loss generally...originals are what people seek, not heart-stoppers.

It is not so commonly known that the mods. sometimes designed and issued out through the poular Ham radio magazines in USA on using services gear as communications receivers for the public proved to be adverse to their performance. ..not all but some and they were in some cases serious mods.....which sounded brilliant. I own a highly 'improved' Drake Rx which I bought not knowing the 'inventiveness' installed ....what's Sherwood's view?...get rid of the junk and let them do the job properly.

To take one section of a set and rebuild it ignores what is seen looking in and looking out of the 'new black-box'. Apart from ruining any intrinsic value of the set gains from solid state conversions of an oscillator are a waste of time especially in a set built to be stable even under battle conditions. The feeling of achievement can be realised from restoration rather than 'improvements' and the forum shows how sidetracked rather than focussed people can become when enthralled by a concept with limited engineering ability where you don't realise what you don't know.... I know, I was there when I was 18!! and decided to modify 400Hz ..sets for customers of a military "disposals" store.

By the end of the forum even the matter of voltage surge was still unresolved. Making a solid -state power supply is reasonably straight forward unless you want to then supply solid state circuits in a manner which exceeds their stability over the tubed supply.

Matters of surge are dealt with in some sets with tubed p/s by ensuring filaments are well warmed prior tp applying ht. Matters of resonance, arising on this site, are handled in design by engineers who were usually not only more competent than those wanting to alter their designs but much closer to the whole array of science at the time.

I suggest if you will not be deterred from playing Brer rabbit and the tar-baby with your sets/parts that (manually or with circuit design) once the set is warmed that you mechanically or electrically wind up the volyage to solid state additions rather than using thermistors or single value resistors whilst maintaining the traditional crow bar or other regulators used in solid state. Use the "KISS" concept.

Yes we can chnge to more modern tubes for example as Murhphy's did but there are such variations in the electronics of any one tube and its 'equivalent' that a specialist team, as at Murphy when doing a complete change will take into account factors which amateur (by comparison) will not...and will be trying to re-invent the wheel.

If you are a dedicated fiddler and a truly a competent electronics engineer but more likely to be somewhat of a dreamer, better to work on one set and leave the other as original so you have one which wll work in the years you take either inproving or ruining the other.

Military equipment has an 'aura' a spirit, think about removing it from purgatory towards a rebirth through really well done restoration rather than creating hell on earth for it and yourself.

Remember...thse were extremely competent enginers overseen by DoD who created these better sets like the B40, to a spec. which suited the times and which even today will serve you pretty well and be a real buzz to use.

My opinion and advice is this...leave the set alone other than testing and replacing all wayward components, one at a time and marked off on a circuit copy. If you are determined to use some moden electronics stick with the power supply alone and do it as a plug-in.

I realise the initiator said he has components....rather than a complete set...well why not see what you can do in building say a G2DAF receiver using the Turret band changer and the IF strip mixer and so forth. rather than trying to reinvent the Murphy. Better to buy one and restore it whereas as with say "geloso" front ends, the Murphy gear might be a bonus if introduced into a modern design in which great pleasure can be taken in the entire layout.

Life's too short to be grafted to ideas of no real value. Unless you have an electricity bill issue with tube rectifiers, leave them be...or build plug in solid state equivalents of the tumes (and such have been invented and come up for sale. Years ago when thousands of BC348's and AR88's and 11, 19,22, 62 URR 391's and 390's. National HRO's R1155's and so on and on and on were readily available and hams did what hams do...experiment ..often to the ultimate detriment of the set.

Yes circumstances have changed and some fabulous conversions and brilliant engineering done by the occasional Ham...and appalling work and butchery by others...to get narrower bandwidths, more stable osillators and so forth because in war we didn't need ssb nor narrow cw.

That all said...your contribution to history by maintaining sets in restored but virtually original condition as opposed to modified nd butchered (including by installing the totally unecessary "S" meter on sets without them also maintains their pricing. If you want to do mods please consider making them external...for example 100kHz or 85kHz if strips or added audio (inevitably amplifying any defect) or audio filters....or "S meters"