Building a DIY resistance soldering unit (RSU)

[Deleted]

Thanks for that Tim, as stated I did buy an inline spade fuse which I meant to fix on the live feed after the transformer to probe but forgot. Is this where you suggest fitting a 3amp fuse or before the transformer. I'm guessing after as 3amp doesn't sound much for the 240v feed considering the role of an RSU. I'm not sure which fuse is curreny fitted before the tranaformer but will check it.

Regards

Pete

[timb]

Yes, before the transformer, pm sent.

[Neale]

At risk of dragging this out even more, I feel that there is a common misunderstanding being propagated here that can lead to all sorts of wrong conclusions. But first, a disclaimer - I'm going to be simplifying things somewhat here in order to avoid lots of complications that aren't really relevant to the main argument, so to those who understand these things - please forgive me!

First of all, this VA/watts business. Yes, there is a difference, and sometimes that difference is important. But not so much here, so just assume that we can measure power as amps x volts and whether we call the result "VA" or "watts" doesn't matter. So, we have a 48VA transformer with two windings that we can arrange to give 6V or 12V (i.e. in parallel or in series). In either case, the maximum power that the transformer is rated to deliver is a nominal 48VA. So, we could draw up to 8A at 6V or 4A at 12V. I say "nominal" rating as there is nothing dramatic that will happen if we try to draw a bit more current. The transformer will deliver it. However, the more power we ask it to deliver, the hotter it will get, and the manufacturer has decided that a max of 48VA corresponds to the max temperature that it is safe to run that transformer. A less scrupulous manufacturer might rebadge the exact same transformer as 60VA, and the transformer will deliver it if asked. But it will get hotter, maybe to the point of being unsafe. We can carry on trying to draw more power, and eventually the transformer will burst into flames, some protective device somewhere will trip, or molten wax will run out of the bottom of it (which did once happen to me...)

Now, to get to the real point of this rambling post. Essentially, "volts" is determined by the transformer (the ratio of secondary turns to primary turns), and the mains voltage in. Those things are all set in advance. We can't change those, at least not easily. So, we connect the transformer to the mains supply, and we measure, say, 6V across the secondary. In practice for various reasons this will not be exactly 6V but that's a complication that we can ignore. But, the transformer is delivering zero current at this point. We connect a modest load to it, a 3ohm resistor, say, and from Ohm's law we know that the transformer will deliver 2A, and 12VA of power. Make that a 2ohm resistor, and the numbers become 3A and 18VA. And so on. The important point is that the current drawn is almost entirely determined by the load we connect. It is nothing, absolutely nothing, to do with the maximum potential capability of the transformer. Somewhere above, you could read one comment as saying that if we reduce the input voltage to the transformer, the output voltage decreases but the current drawn increases so that the "VA" figure stays the same. No, that's not how it works. If we reduce the input voltage to 120V (that is, by half) then the output voltage will fall to 3V. Now we have 3V across, say, a 3ohm resistor which gives us 1A flowing, and a power draw of 3VA. So, current drawn is determined by load and applied voltage. There is no mechanism that tries to keep the "VA" number constant - it is a kind of side-effect of voltage and current. So, if max current draw is 8A at full mains voltage in, then if we reduce the input voltage then the output current will reduce in proportion. We size the protection device for the maximum current (i.e. worst-case) situation in the knowledge that in practice we will be running at lower current draw when we wind the input voltage down.

[Neale]

Quick question for GG - based on your experience to date, do you see much (or any) distortion in the components due to the local heating? One of the problems with using even a small torch flame, in my experience. Avoiding this would be a good reason for going the RSU route.

[Deleted]

Hi Neale, no but then I haven't done much testing so far, still being very much under the weather which alas has worsened. You would have seen the bridge platform that I did the first test on, no distortion there and that was pretty thin material.

Pete

[Deleted]

Today I used my RSU in anger for the first time, I have to say that I'm mightily impressed. I burnt a small hole in the first part but once the voltage was set to approx 3 1/2 volts all went very smoothly. these parts are very small/thin, approx 0.3 mm

This is the underside support structure of the Admirals bridge for HMS Hood, I've placed a needle file to give an idea of size.



A close-up, the closest rib I later knocked off during clean-up, if you look closely you can see that I hadn't got good penetration here. I'll solder it back on tomorrow.



A couple of pictures to show the part fitted..



And lastly a view from above, I think that's what you call a precision fit...




To sum up, i think the the RSU works.....

Pete

[timb]

Hi GG, glad to see your fine work and that the RTU is functioning well!

I was not going to propagate this discussion further, however, sadly an accusation has been made and a defence is required!

In AC circuits, Watts and VA can never be considered to be the same, a common misunderstanding!

Indeed to say that VA and Watts can even be considered the same is simply incorrect and shows a complete lack of understanding. Science does not generally give different names to the same values.

It would appear I am not very good at explaining but here goes.

As stated earlier VA is a product of the voltage and current, there is no resistance component. In the case of our transformer it is effectively set at manufacture and is the maximum (rated) apparent power – apparent because voltage and current are never quite in phase, the difference being the power factor. Yes I know that this will be negligible in domestic use but it is there. Now I have to include rated, this is the power that the manufacturer has determined – for the component to work correctly for its lifetime.  You wouldnt want to blow anything up by getting this wrong!

The VA is an indicator and should be used to select the appropriate device for the job in hand, and the size of the protective device, fuse, circuit breaker, whatever. The fact is it is predetermined and cannot change, there is no suggestion of any magical mechanism it is the result of mathematics, and is largely theoretical as it cannot generally exist in reality due to the phase difference as indicated above. There is no resistive component so there is nothing to change.

So what happens when you add a resistive component (or a load)? Well Ohms law comes into play and for a given voltage you will get a variable current for a varying resistance. This varying current can be multiplied with the voltage to give used power. Obviously this is different to the apparent power and is recorded as Watts. This is clearly (and done quite well in my opinion) described above but is not VA that is being calculated but Watts when a load is included. Just goes to prove how easy it is to mix the two when you don’t know.

In summary:

VA – rated aparent power determined by the manufacturer to allow selection by the designer. Used among other things for the selection of the protective device. It is the apparent maximum available power.

Watts – actual used power – variable depending on load. Used among other things to calculate your electricity bill.

[David 1/2d]

Because hobby RSUs seem hard to come by, I've just bought a Frost unit to examine, with a view to Spreading the Word - always provided the machine cuts the mustard.

I'm most grateful to Pete (greenglade) for what he's posted already.
Like him, I quickly came to the conclusion that Changes Must be Made to convert this handy automotive tool for modelling.

I may find this needs internal modification (invalidating warranty & all that jazz) but, before even considering that, I shall 'enhance' the unit as much as I can without.
There are folk who really don't want to be doing that, plus folk who really shouldn't be allowed to.

I've run a multimeter over the obvious bits, with some illuminating results already.
No doubt I shall know more once I've opened the box and found out what that little red nipple really does when squeezed. (Hint, it gives a 20% boost.)

Here's where I'm coming from, so y'all can gauge how to pitch replies:
Retired mechanical engineer who studied Electrical Machines (but not for half a century).
Owner of a London Road Models RSU, which I ain't parting with, not any time soon.
Build trains in any scale that'll run on Gauge 1 track.
Help edit a magazine heavily into G1 scratch-building.
Admin online forums on modelling and machine tools.

David Halfpenny

[Deleted]

Hi David, I for one will be most interested to see your findings on the Frost unit. So far I am very happy with the unit and the mods undertaken. I have had no side effects from its use even when holding the switch down for more than a few seconds. I'm currently only working with thin brass, from 0.3mm up to 1mm. I have the machine set at approx 3.5 volts, more than 4 volts quickly burns a hole in the thinner brass. IIRC the unit goes up to just over 6 volts with the 2000W AC controller at max. I'm looking forward to using the RSU when I make a start on my own gauge 1 loco with the much thicker material involved...

Pete

[David 1/2d]

First-off, RTFM.

Among other sensible things, the single A4 side of Instructions says:

"The voltage is very low, 6V, and is connected to mains earth so it is very safe."

Using a cheap multimeter to start with,
- Primary winding resistance is 102 Ohms.
- Secondary Open Circuit voltage is 7.2V at zero current.
- Resistance of Secondary is too small to measure.
- Resistance between mains Earth pin and either output terminal is also too small to measure.
- Resistance of the supplied carbon rod is 0.1 Ohm from end to end. Could be anything up to around 0.2 Ohms.

"The carbon tip is fragile"
- Yep. Having sharpened it with a pencil sharpener, I put the 4mm rod into the 6mm hole in the holder, and it vanished. It wouldn't come out when I slanted the holder downwards so, like an idiot, I shook it gently until it came out of the hole, skeetered across the bench, crashed to the floor and broke in two. Luckily the pointy end is long enough to work with.
(Comment: Frost do spares, I have plenty of London Road Models rods and, at a pinch, rods from dry cells might do.)

"The red button switch can be operated by foot and will give double current for heavier jobs. The switch needs to be held in and when released will return to normal current."
- In the current tests that I'll come to, pressing the red button switch gives a roughly 25% increase in current, rather than 100%.
- The switch is smooth and easy to operate with whatever extremity is convenient.
(Comment:I would have preferred a mechanical click, and I would have preferred a latching switch to avoid having to hold it in while working. But those are things I'm used to, and a new-to-RSU user might not miss them.
The unit lacks rubber feet and skids around on a smooth floor, but bit of rubber mat keeps it still.)

"DO NOT ALLOW PROLONGED CONTACT TO OCCUR BETWEEN THE CLIP/PART AND THE CARBON TIP/METAL BODY OF THE SOLDERING IRON TOOL OTHERWISE THE UNIT MAY OVERHEAT AND PERMANENT DAMAGE MAY OCCUR."
(Comment: Shouting, "Not Continuously Rated", which is fair enough.)

"ENSURE THAT ONLY THE CARBON TIP COMES INTO CONTACT WITH THE ITEM BEING SOLDERED, AND NOT THE METAL BODY OF THE SOLDERING TOOL."
- Must look for a reel of PVC tape.)
(Comment: Implying (no more than that) that the resistance between the soldering tool and the carbon tip might be significant.

An extra sensible thing that would be good in the Instructions is mention of the mains plug-Fuse, which is 1A, and can be replaced without dismantling the moulded-on plug. (As with the LRM RSU, there's neither fuse holder nor switch on the case.)
- Must get some 1A spares, just in case.

David

[Deleted]

Thanks for the info David...

As you would have read I took my unit apart, threw away the plastic casing and mounted the transformer in my own box with the listed mods.

I wasn't happy with the fact that there was no off position once plugged in and nor was I happy with there being no foot-switch per se. Frost call their red button a foot-switch which of course it isn't being always on and when pressed giving more power. I would be inteterested to learn if pushing the red button makes a circuit for the extra boost or breaks the circuit?.. I have no idea how it's wired up?

My Frost carbon tip is still in its packet, I used the copper sheaved rods that I bought from eBay before getting the Frost unit when I first looked at building the RSU, I have quite a collection of parts while deciding on the design including two other AC controllers.

I am very careful to only make contact with the carbon rod...

A fellow member of this forum kindly sent me a 2A MCB ( mini circuit breaker) which I'll fit ASAP.

Kind regards

Pete

[David 1/2d]

Jan 22, 2022 23:57:14 GMT @greenglade said:

I wasn't happy with the fact that there was no off position once plugged in and nor was I happy with there being no foot-switch per se. 

The designer/builder of the Frost RSU told me he uses it for jobs like attaching a wire to a crocodile clip. That is quite different from assembling an etched kit.
So here I'm aiming to identify what it is that we need to change, without in any way 'criticising' his design just because we're different.

At the same time as applauding your rebuild, I think we can accessorise-up without needed to open the box.

Key differences include (but are definitely not limited to):
1 - we can really do without any Arcing at all between the probe and the work,
2 - at times, we need to use the probe as part of the clamping of the joint,
3 - we do need crisper control of the ON time than is possible by jabbing with foot-long spear,

Leaving the Scarlet Pimple completely aside for the time being, the Biggest Deal for me is adding a foot-switch, and fortunately this is easily done.

(to be continued) 

[Deleted]

Jan 23, 2022 10:49:46 GMT David 1/2d said:

The designer/builder of the Frost RSU told me he uses it for jobs like attaching a wire to a crocodile clip. That is quite different from assembling an etched kit.
So here I'm aiming to identify what it is that we need to change, without in any way 'criticising' his design just because we're different.

To be fair, as you say it's not designed for model making and is probably fine for the job intended but I do wonder how it got a CE stamp with it always being live once plugged in?

It is however a perfect low-cost candidate for being used for model making with a few mods.

I agree you don't need to open the casing to improve things, I opened it as I wanted it to sit in my own case and was very happy when opening it to find that the original mounting tabs on the transformer were still there and just folded flat. Yes, just adding the foot-switch to remove arcing is probably all it needs although it would still be too powerful for some of our work, might be ok for gauge 1 if adding the important foot-switch. My first test using the unit as supplied on some 0.3 mm brass totally destroyed the brass, most of it disappeared, so IMHO it's not suitable for smaller scales and perhaps even small parts on gauge 1?

The reason I asked about the red button is all I have done is bridged the two wires to the button with the button removed and I'm not sure if doing so has increased the power (ie activated boost) or not? It's not important as the unit as built works better than I had hoped.

BTW the button is very fragile (mine fell to bits), I'm not sure how long it will last with foot control used?

Regards

Pete

[Deleted]

Hi David, you need to use a host, there are free versions out there such as Imgur and Flickr, take a look at the thread near the top of the page 'Posting photos' for details.

Cheers

Pete

[David 1/2d]

Sticking with Foot-switches for the time being:

The only pre-wired Momentary foot-switch I can find is this Proxxon FS, which has two snags:

ProxxonUN-EarthedFootswitch by David Halfpenny, on Flickr

The more visible one - EU plug and socket - is easily fixed, either by UK adapters as here, or by cutting them off and fitting UK equivalents.

However the less visible one (which some will have recognised from the fittings) is that this is a two-wire set-up, viz no Earth.
While an RSU doesn't inherently need an Earth, if whatever's in the Frost box relies on it for safety, Avoid this solution.

So, unless somebody knows of a pre-wired Earthed foot-switch, we'll need to do a bit of mains electrics at the 'Wiring a Plug' level of competence.
As ever, don't try this unless you know you are competent; however it won't be hard for most people to find someone nearby who is.

This is an unwired Kit of Parts for an indoor foot-switch extension:

Footswitch Extension Lead 1 by David Halfpenny, on Flickr 

Because both the donor extension and foot-switch are pre-wired, all that's needed is to cut-in a Junction Box with four sets of terminals, like this three-port 16A one with internal cable clamps.

This particular foot-switch is 10A, and all-plastic, so it doesn't need earthing itself but, if it were a metal one, it could readily be earthed using the spare core in its cable.

A slight variation for somebody wanting more weatherproofing (maybe you spill a lot of coffee, or are bonding a garden railway) is this similar set-up using an outdoor JB:

Footswitch Extension 2 by David Halfpenny, on Flickr

And while details of which wire goes where can come later if needed, this is what these JBs look like inside:

Junction Box Innards by David Halfpenny, on Flickr

FOUR terminals, please!

(I'm going to pause again now, to make sure this photo-alchemy really works.)

[David 1/2d]

So now I've struggled along behind you all to get as far as this aspiration from Pete:

• "Yes, just adding the foot-switch to remove arcing is probably all it needs although it would still be too powerful for some of our work, might be ok for gauge 1 if adding the important foot-switch."

let's put some meters on it, namely digital voltmeter, clamp-round ammeter, and a domestic plug-in mains power meter.
The last above shows my mains varying between 243V and 246V at the socket, so don't take too much heed of last decimal places.

I'm putting the croc onto a piece of 30thou brass (scrap from a G1 live steam locomotive), and placing the probe between 5mm and 100mm away from it, just until the digital displays stabilise. Each is a representative figure from several sparky red-hot interactions:

• • • Open Circuit      7.1V and 0A at Probe, 6.6W at Mains socket (Watts not VA)
      Red Button pressed:      7.2V and 0A at Probe, 5.0W at Mains socket (unexpected!)
    
    
    • Normal Operation      ~0V and 30A at Probe, 215W at Mains socket
      Red Button pressed      ~0V and 40A at Probe, 290W at Mains socket

I think it's the probe that pongs a bit, but it could be the transformer.
Anyroad, the transformer's plastic box isn't getting hot, though the metal shank of the carbon holder is getting warm - no problem for those used to conventional soldering irons.

I thought the designer told me he'd got 13A out of one, but maybe there was a crackle on the line and he said 30A.
Anyway, I'm now confident that the Frost RSU can deliver as much current as a London Road Models one.
But since Red Hot is over-the-top, can it be persuaded to deliver as little?

Pete's shown how to control the current by cutting the output voltage.
Sorry Pete, having designed SCR drives for motors, I wouldn't have put a domestic solid-state device onto an inductive load without seeing the waveform on a 'scope - and maybe you did? But if nothing's smoking, dripping, interfering with broadcasts or trashing your internet, you're probably getting away with it.
For a man with my last-century prejudices, a small Variac autotransformer is the way to go - as fitted to some American RSUs - and one this size can be had for £27 at the moment. Heavier though.

Finally for this post, I do have a Big Worry about the figures above:
How much of the 200W to 300W is being dissipated by the defenceless little transformer, rather than the target joint?
Looking at your photos of the wiring inside the Frost Box, will any of the soldered joints inside the box let-go?
When I've sent this, I'll go back to find the transformer rating you quoted.

And I'm still intrigued by the Scarlet Pimple - what does it do?
If you've s/c'd those wires on you rebuild, Pete, you're probably getting Full Whack all the time.

[David 1/2d]

I see Pete's unit had a 115V + 115V / 6V + 6V transformer rated 50VA.

Deep Breaths while I nip back to the workshop and unplug mine :-)

[Deleted]

Jan 23, 2022 17:00:32 GMT David 1/2d said:

Pete's shown how to control the current by cutting the output voltage.
Sorry Pete, having designed SCR drives for motors, I wouldn't have put a domestic solid-state device onto an inductive load without seeing the waveform on a 'scope - and maybe you did? But if nothing's smoking, dripping, interfering with broadcasts or trashing your internet, you're probably getting away with it.
For a man with my last-century prejudices, a small Variac autotransformer is the way to go - as fitted to some American RSUs - and one this size can be had for £27 at the moment. Heavier though.


Scope? I'm a complete amateur when it comes to electronics, no such equipment here but I do have a good head and can usually work my way around things, I read up on any info that I could find online with some interesting solutions, dimmer switches to variac's, etc. I went with an AC controller, if that's bad practice then so be it, I actually bought 3 in case there were any problems but the one fitted worked perfectly straight off, so no need to try the others. The one I'm using now is rated for 2000w, another that I have goes up to 4000w, they are all very cheap things to buy, less than a tenner.

'And I'm still intrigued by the Scarlet Pimple - what does it do?'

I think you may be referring to the switch that I fitted in the back of the machine, you'll note that this is not in the actual wiring diagram given. The reason is that I bought the wrong type of switch, DC rather than AC. The plan was to use a rocker switch in place of the red button but since the unit is working fine with more than enough power I won't proceed with replacing the switch. Especially as your last comment 'If you've s/c'd those wires on you rebuild, Pete, you're probably getting Full Whack all the time.' suggests that I'm already using full power and thus have no need for the switch, the Ac Controller does a much better job of controlling the power down to low settings.

Regards

Pete

[David 1/2d]

That makes sense, Pete, if Frost's own Red Button is Push-to-Make (which it might not be).

Next, I set the mains Power Meter to read Amps instead of Watts, which is when more pennies began to drop.

Mains voltage: 248V
Open circuit mains current: 0.150A
Normal Operation mains current: 0.94A
Red Button Pressed mains current: 1.25A i.e. over the 1.00A plug-fuse rating
Red Button Pressed total dissipation: 297 W = six times the 50VA transformer rating

These high figures were now bugging me, so I thought back to what I'd said earlier about probe resistance, and checked the probe.
I found I'd substituted a piece of 5mm diameter carbon rod for the broken 4mm rod.
Oops! - yet easily done when the hole is 6mm diameter.


And swopping back to the 4mm stump actually did calm things down a bit:

Mains voltage: 249V
Open circuit mains current: 0.157A
Normal Operation mains current: 0.71A
Red Button Pressed mains current: 0.95A compared with 1.00A plug-fuse rating
Red Button Pressed total dissipation: 185W but obviously with a corresponding reduction in current delivered.

So I'm staying bugged - but at least there's a good chance the plug fuse might do its job.
Unless somebody replaces it with whatever fuse the corner-shop has in stock, namely 3A or 13A.


I'm signing-off tonight with this uncomfortable thought going round in my head:
Is the main 'control' on this RSU a resistance that isn't obviously critical, even to an RSU user, and is far too small for the average person to measure?

If the answer is Yes, my mission to tame it without modification is more involved than I'd thought this afternoon.
Horlicks Time.

[David 1/2d]

I've been collecting kit to 'drive' the Frost RSU.

First, 4mm banana adapters, so I can use accessories from other brands:

4MM Sockets for Frost RSU by David Halfpenny, on Flickr

simply-made from inline banana sockets, just by cutting the insulation with a sharp knife:

4mm Sockets by David Halfpenny, on Flickr

Here's a cheap 1kW autotransformer (tradenames: Variac, Powerstat, and others) with a small bench kiln it regulates:

1kW Kiln with 1kW Variac by David Halfpenny, on Flickr

and by itself:

1kW AutoTransformer by David Halfpenny, on Flickr

(I just love this machine! Partly because it's red and shiny - obviously - but because it channels Faraday's bench in the Royal Institution in the 1830s. The great big knob gives very fine voltage control from 0-330V with no interference, and minimal losses. As this one is for the little kiln, it's twice the bulk we'd need just for the Frost, but it has the useful refinements of digital voltmeter and UK plug and socket built-in.)

and here's an alternative SCR (silicon controlled-rectifier) controller, that I made myself 53 years ago, so that an elderly University could evaluate a TRIAC two-way rectifier:

TRIAC Mains Controller by David Halfpenny, on Flickr

(More on how they work later)