Solid State Vibrators -v- Inverters ?
 

I'm just trying to make sure I know where the differences lie between these circuits / devices.

Obviously, the SSV's came in to being to replace the old electro-mechanical switching devices the likes of which I had in my Class D wavemeter. As the original vibrators had finite lives, and have long since gone out of production, installation of an SSV is very neat solution to keeping such equipment working.

From what I've read and from what I'm seeing on a current project, the SSV generates a square wave signal of an appropriate voltage to work with the old transformer to generate HT.

But what are the differences between that and a modern inverter, other than in the inverter, the transformer and the electronics are designed together with each other? Those of us who lived through a certain era on 2m will always have a sweet spot for the sound of a Pye RT inverter :D.

My reason for asking is that I have a project in-hand which uses a single valve which needs about 150V at no more than ~2mA (it's GDO), and I'd like it to run from a 12V SLA. I have two ready-made SSV's boards which came from a forum member in Australia many years ago, though he has (I think) not been on the forum for a very long time (I cannot recall his name).

I've tried these out with various transformers from my junk box and both work well and so I've just ordered a very small 2VA transformer which will keep the size and weight down. Before I commit to going down this route, is there anything else I could/should consider?

Multiple PP3's wired in series are not something I want to do!

Thanks

B

Re: Solid State Vibrators -v- Inverters ?
 

My first thought is that the frequency of oscillation is the difference.

Electro-mechanical and SSV's are designed to be used at a low frequency which will work with conventional 'mains' transformers.

Invertors often work at much higher frequencies with ferrite transformers.

Re: Solid State Vibrators -v- Inverters ?
 

If you are keen to keep the weight down, why not go with a simple push-pull inverter, but use a high frequency (say 20Kc/s or above) and use a suitable ferrite core for the transformer?

Re: Solid State Vibrators -v- Inverters ?
 

An inverter from a duff CFL might to the job, assuming that isn't why the light doesn't work. They are very small, light and efficient, the voltage probably be exactly what you want but that shouldn't be to difficult to change.

Peter

Re: Solid State Vibrators -v- Inverters ?
 

I think the key difference is that vibrators and their solid state substitues just switch a low voltage supply to make square-wave AC. From there on you just use a transformer.

There are many different types of switched mode power supplies, and they mostly do not work like that at all, because there are more clever ways to directly and efficiently make high voltage DC, often using one or more chokes, and done at high frequency so the chokes can be tiny.

Re: Solid State Vibrators -v- Inverters ?
 

As is not unusual (and can be quite irritating, but obviously useful), a while back I bought a small switch mode pcb module from eBay which supplies a voltage variable between 90 and 180V for a few pounds. I like making invertors but it's more the passion than sensibility these days.

Re: Solid State Vibrators -v- Inverters ?
 

I think of "inverter" as a broad description of devices that produce an AC output voltage from a DC input voltage, with "vibrator" being a sub-set. The picture gets complicated when synchronous vibrators get considered and for some time now, inverter has been used where "converter" (DC to different DC via AC!) used to apply. A question of definition, blurred by time and change of technology.

Vibrators necessarily ran at fairly low frequency (I think often around 100Hz or so) because of mechanical limitations. The first Royer oscillators (which could be thought of as a "similar but different" electronic replacement of the vibrator) also ran at LF but soon went to kHz or 10's of kHz. Pulling apart CFL drivers, these often look like HF Royers in principle, but I haven't examined in detail.

Re: Solid State Vibrators -v- Inverters ?
 

They are really just different names for the same thing. A solid state vibrator is by implication a substitute for a mechanical vibrator and will produce a similar oitput. Inverter is the broader, more generic term and can involve a wider range of frequencies and waveforms depending on the application.

Re: Solid State Vibrators -v- Inverters ?
 

Vibrators actually work at a bit higher frequency - 80Hz (as per vibrator specification). This allows a reduction in the size and weight of the transformer, which is useful.

The vibrator's frequency is set by the spring constant of the vibrating reed, and by the weight on the reed which tunes it, so it is fairly accurate. This is important, as the transformer often has a resonating capacitor across one winding, so that during the 'dead time' between one contact opening and the other closing, the transformer voltage has smoothly reversed and the contacts 'make' with zero sparking.

A SS replacement thus has to replicate the steady frequency, and predictable dead time, as a drop-in replacement.

Now a vibrator power supply (or a SS replacement) can not have anything in the way of regulated output voltage, or overload protection - other than fuses. But a DC/DC converter / inverter / whatever you want to call it, can!

A DC/DC converter, by suitable connection of switching transistor / diode/ transformer / inductor, can be arranged to allow control of output voltage by varying on and off time of the switching transistors, at high efficiency. This is a fundamental difference. It is true that they usually operate at a much higher frequency, but that is a detail, and made possible because of the high speed of transistor or MOSFET switches. Waveforms can be square, but often aren't (the single-transistor flyback converter has an 'interesting' waveform).

For your 150V 2mA (300mW), I'd honestly suggest look for a cheap module which already exists, eBay or wherever. You can google 'boost converter' or 'flyback converter' each of which would fit the bill (flyback also allows the possibility of isolation of input and output) but you can easily spend a few weeks of spare time developing something which is not the core activity, just getting the power rail! Of course, you'd have a lot of fun in the process :)

Re: Solid State Vibrators -v- Inverters ?
 

Thanks for the replies. I think that if my requirement was more demanding, and if efficiency was more important, looking at a "real inverter" design would be worthwhile, but in a situation where the application is so basic, and my two SSV boards have been sat collecting dust for some years, I'm going to press on with one of those, at least for the time being.

I obtained two SSV versions from the Australian guy; a small one using a SQ4942 and a larger one with a pair of BUK7608 MOSFET's. They're both built very neatly with SMD components on compact PCB's. Even the smaller one meets this particular need quite easily. I'm building a copy of the the American "Megacycle Meter" GDO, using an Acorn valve, (deemed by concensus in a previous thread on the Forum to be a particularly good (the best?) GDO), and rather than put in a mains transformer, a small SLA and SSV-psu will enable it to go outside for tuning aerials, if needed.

B

Re: Solid State Vibrators -v- Inverters ?
 

And they all generate RF interference 8-o

Re: Solid State Vibrators -v- Inverters ?
 

Hi Bazz, an electrically quieter but slightly larger solution would be to use a small 50Hz transformer to say, 30v and then a voltage multiplier (diodes and caps) to give your 150v.

Ed

Re: Solid State Vibrators -v- Inverters ?
 

Decades ago, I had a car radio whose vibrator was well past it, and which would not respond to my "mechanical adjustments". After looking thro a few US written library books, I knocked up a simple replacement consisting of a cross coupled multi-vibrator using 2 off OC28 transistors and a couple of (330 ohm from memory) emitter resistors. I optimised the resistor values by way of a "hot finger" on each transistor. It worked and I sold the radio on a couple of years later.
Les.

Re: Solid State Vibrators -v- Inverters ?
 

That's an interesting insight. If, for the moment, I just want this unit to power a GDO, I'm guessing that RFI is not a key issue, as it might be for a receiver? For sure, the leads in and out would benefit by some basic filtering, but hopefully, there should be a need to do more than that?

B

Re: Solid State Vibrators -v- Inverters ?
 

Not all. Two ways around this I have found.(Trying to shield & filter them never works 100%)

One is to make it a sinewave converter.

The other, if square waves are used, is to keep the fundamental frequency very low, around 50Hz and tune the transformer primary and control the transistor saturation so it is not too heavy. Have a look at the scope recording at the last page of this article:

http://www.worldphaco.com/uploads/WORLDFETRON.pdf

This DC-DC converter ( Called the Omega Device) generates no detectable RFI over 100Khz. It had to be like this because it was for a radio nearby.

Luckily for the OP's GDO, RFI won't be a problem most likely.

Re: Solid State Vibrators -v- Inverters ?
 

I always think of the vibratior-type PSU as a specific subclass of SMPS.

The solid-state inverters I came across in things like Pye Vanguards always operated at high-ish audio-frequencies - 400Hz to 1KHz - they whistled rather than hummed. With suitable design this means you can use a physically-smaller/lighter transformer but you also need to use suitable 'fast' diodes - an 1N4007 is _not_ a good idea at 1KHz!

A while back I experimented with an old vibrator-transformer and a pair of 2N3055 in a classic Royer-style circuit, but it didn't work as well as building a new transformer using a colour-TV LOPT as the core.

Re: Solid State Vibrators -v- Inverters ?
 

I remember those, the frequency of the inverter changed as the PA valve warmed up - which wasn't long as it was a special quick heat valve.

Peter

Re: Solid State Vibrators -v- Inverters ?
 

There are some very cheap adjustable dc to dc converters that go up to 300 volt on eBay from China. A few people have successfully used them. Interesting I have a few 8 pin dil chips from the 70s with complementary output that were designed for use in inverters. They have an internal 50hz oscillator and required no external components just a power transistor. I have replaced many mechanical inverters in the past with the electronic equivalent. Although my design is unusual in that I use a pic chip so I can tune to the transformer in software. Those step up transformer are surprisingly frequency dependant.

Re: Solid State Vibrators -v- Inverters ?
 

Yes, and the quick-heat valve's filaments were powered from the inverter!

This was actually quite a cunning scheme: at initial power-on the filaments were cold and provided a serious load on the inverter - which, because the filaments weren't hot enough to be emissive also meant that there was at the time no load on the HT-side of the inverter.

So most of the inverter's power got applied to heating-the-filaments-as-fast-as-possible rather than charging the HT-side's smoothing capacitors to a crazy voltage.....

As the filaments warmed, their resistance rose so their load on the inverter fell - but at the same time the filaments were also becoming cathode-emissive so the load on the inverter's 'HT' side increased towards full RF output.

So at all times the inverter was being worked to its fullest capacity even as the balance between LT and HT demand shifted.

See

https://frank.pocnet.net/other/Phili...k-heat_VHF.pdf

for the Mullard/Philips range of these quick-heat valves. Somewhere out there is also a series of design-notes by Mullard/Philips on suitable transistor inverters - I remember the big ADZ11-style "Top Hat" transistors -

https://www.radiomuseum.org/tubes/tube_adz11.html

being the preferred devices for such tasks.

Re: Solid State Vibrators -v- Inverters ?
 

Mmm, but my recollection is that most of the Pye RT's used standard valves rather than the quick-heat types. Perhaps the latter only came in at a late stage?

B