Building a Workstation.

[rebelrider.mike]

I've had this idea kicking around in my head for a while to build a safer place to do electrical tinkering. I have an autotransformer, and I should be able to make an isolation transformer pretty easily. I also found this old Digi-designer breadboard that doesn't work.

I searched around a bit to make sure I wasn't destroying a piece of history. Seems these can still be purchased used around the internet. I've been looking for a safer way to make temporary circuits for both AC and DC things. Here is a picture of what I'm thinking:

This is not to scale, and I don't really know where all the components will go, or if they'll even fit. I'm pretty sure the isolation transformer would have to go before the autotransformer. As far as what comes after the receptacle. I've heard that changing the input voltage on an isolation transformer is bad for it. I'm also planning on having a 250W incandescent bulb as a current limiter. I'm not sure where that should go, but I'm thinking between the two transformers so the voltage won't drop after the autotransformer is set. Or maybe it will anyway?

I'm not sure about grounding (to the Earth) either. The general consensus seems to be that grounding the output of the isolation transformer defeats the isolation part. I've also read that grounding the output is necessary to comply with safety requirements. I get the feeling that this might be a "can of worms" thing. I'm wondering if another current limiter on the ground would be a good compromise? I haven't seen anyone else doing this though, and I'm still a noob at some of this theory, so this might be a dumb idea for reasons I'm not aware of.

I read that in some industrial settings, instead of a direct path to ground which could shut off some big machine, causing a hazardous situation, they instead use a ground fault indicator to let technicians know that there is a problem. So the idea is not entirely unprecedented. It looks like I'll have a GFCI protected circuit to plug into at my workstation, but I'm not sure how much damage could occur by the time the GFCI would react to a ground fault.

Anyway, I'm also trying to decide how to switch the thing on. Specifically, I was thinking of using an on-off-on switch so only the AC or DC would be powered at a time. I don't see needing to use the AC meter while working with DC stuff, but I do see myself wanting a DC meter while working with AC stuff. A lot of the times that I use AC, I'm testing transformers, and rectifiers to make sure they're doing what I want. Also, the AC to DC power supply is going to expect a small range of voltages, and I'm not sure if it would be damaged by very low voltage from the autotransformer or if it would just not work. I can always use multimeters if I can't make integrated meters work or fit.

Well I took the box apart and saved as many of the components as I could.

I'll definitely re-use the power cable, fuse holder, neon lamp, breadboard, switches, LEDs, connectors, and wires. I also tested out the transformer, rectifier, and capacitor. They're all good.

There are also a ton of transistors, and I have no idea as to what they do. I'll save them in case I figure out how to use them someday. The top panel will have to be replaced. All the connector pins are exposed metal, and I don't want that on the new setup. Especially the AC side. I'll have to come up with something different. Perhaps a sheet of plastic.

One thought about current meters: A permanently installed one with a CT instead of a shunt, could have the wire wrapped through the CT several times, making a much more sensitive meter. Couldn't do that with a multimeter. I'd have to be sure to label it appropriately.

 

[rebelrider.mike]

I've been shopping for components, and I don't think I'll be able to fit all of them on the panel that I wanted. The AC Volt and Amp meters are not what I thought they would be. All the digital AC Volt meters I could find use the same source for power as for testing. And they're limited to a minimum of 60V. If I want to measure lower, it seems I've got to use analogue meters. They're larger, and so take up a lot of the limited panel space.

There also seems to be few options for AC current meters. I think I will be using an analogue meter here too. I found a couple that measure 0-3 Amps. I've decided to limit the input current to 2A. That's around 250W. I can't think of anything I'll be doing that would require more power than that.

The DC meters were much easier to shop for. I picked ones that have CT sensors instead of shunts for the current so I can wrap the wires around the CT a few times to multiply the current reading. I also found an oscilloscope that should be able to be modified for panel mounting. Here's what I've got so far:


I've also been reading more about isolation transformers and grounding. It seems that how you ground an isolation transformer depends on whether the priority is human safety or equipment safety. In fact, there is a 4 page discussion on this forum about this topic. I didn't quite understand all of it, but I think I will go ahead and ground the secondary side and have a small bulb as a current limiter. I drew up a diagram for this too:


The general feeling I get is that we want a GFCI outlet protecting this device. I've read that the secondary of an isolation transformer doesn't have a hot and neutral since there is no ground reference, and so the wires act as line to line with 60V each. I don't know if this is true, so I'm curious to try it out. I guess one could test that by seeing how many Volts there are between each line and ground.

I've also read that isolation transformers sometimes have Ferriday shields between the primary and secondary windings. They are supposed to absorb line noise from the primary and shunt it to ground. If I understand it correctly. I'll see about adding one to the transformer when I make it.

I haven't tried calculating light bulbs as current limiters before. For now, I've chosen a 250W lamp for the line limiter. When cold, it has 4.5Ω and should only drop the voltage a little bit. I figure if I have the input voltage set to 130V max., and don't use more than 2A of current, the voltage to the breadboard will still be 121V, and the lamp will dissipate around 18W of heat. If there's a short, the lamp will quickly heat up, and the resistance will reach 58.8Ω, limiting the current to 2.2A. I think if I had a 2A fuse it would probably not blow at that current if it was shut off fairly quickly.

The ground current limiter was easier since there shouldn't be any current on the ground wire. I chose a 20W lamp with a much higher cold resistance of 63.8Ω. That should be low enough to get rid of any static electricity fairly quickly I hope. And if there is a short to ground, the bulb will quickly get to 775.8Ω, limiting the current to 0.17A. I'm thinking though, that a short to ground should also trip the GFCI pretty quickly.

My autotransformer blew its fuse (user error) ages ago and I never bothered to replace it. I'll have to go find one, and also see what the maximum output voltage is. I think it's 130V but I'll double check that before I start buying stuff.

 

[rebelrider.mike]

I brought out my autotransformer and decided to clean it up a bit. It works, but sounds very crunchy when adjusted.

There was a lot of dirt and debris inside. I mostly just used a brush to clean it all up. Put some new grease on the turny part of the shaft. Inspected the brush. Took out a couple dents in the housing. Double checked that the voltmeter on top doesn't work and I don't think it can be replaced. It's a flat surface mount style and I've never seen any thing like it for sale. No matter; the Probulator will have its own AC input voltmeter.

Got it all back together, but decided to remove the voltmeter leads. They've been squashed at some point anyway, and there is actual wire exposed at the squash sight. Maybe the blown fuse wasn't user error after all...

Interestingly, there are two unused taps on the internal panel. Almost looks like a potentiometer where one side increases while the other side decreases. I'm not sure that's the case here. Anyway, I'm not changing that. The first tap is the common neutral for both sides. The second tap is the hot input. The third tap is the variable output hot.

While I had it apart, I added some rubber feet, and some washers to the screws that hold the front panel together since one plastic corner is broken off. Got it all reassembled and it works! It actually goes up to 140V, not 130V like I thought. Shouldn't affect the other parts though. I still need a proper fuse before I can use it. I've temporarily bypassed it in order to test it. For now I'll put it away so it doesn't get crusty again.

I noticed some arcing when I first plugged it in. I got out my clamp meter and tried it again. It has about 3A of inrush current for a fraction of a second. From what I've read, some inrush current is to be expected with transformers of any kind. Makes me wonder what two in series will do. Maybe I should put the current limiter bulbs between the transformers and the receptacles. Maybe it doesn't matter. I'll look into it further.

The isolation transformer will take a while. I went out to the shed to look at what I had. I thought I'd pulled all the transformers out to be easily accessible last time I had the shed open. Turns out I actually hadn't. I don't want to pull everything out again until the outside dries out a little.

 

[rebelrider.mike]

Finally had some good weather (its gone already) to clean out the shed. My son helped, and we found all sorts of useful junk. Amongst the various treasures were several transformers:

Two big suckers that are attached to some kind of vacuum tube power supplies. One of them had a label:

Searching yielded very little information. I think they may be from the 1940's. Another is really small. Reminds me of an old doorbell transformer. The last one though looks promising:

I think it may be from the late 70's or early 80's from what little info I've found based on the company and part number. It's an old-school ungrounded transformer. I suppose I'll have to modify it a little. It does have a Faraday shield and wire though for noise filtering. So that's nice. Its also limited to 1.3A. A little low for what I was thinking, but I suppose it will be fine.

I found it in its original box. And its in almost pristine condition. It may never have been used before.