Dim Bulb vs SCR/Triac/Diac vs MOSFET for AC Current Control- Which to use? SOLVD

[t1d]

I am investigating AC Current Control for a DIY AC Power Supply made from a DIY Isolated Transformer and Autotransformer. I understand the Dim Bulb approach, but I do not like the idea of the fragile, large glass bulbs in my lab. SCR/Triac/Diac solutions look to be cheap enough and work. MOSFET solutions are accomplished by using a Bridge Rectifier. So, which to use and why? Thanks for your help!

Edt: Mines well consider Quadracs and Alternacs, too, please and thank you.

[Gregg]

A dim bulb tester output doesn’t have the spikes etc. associated with switching if a reasonable sin wave is what you are looking to obtain.  Not that the mains have really good wave forms anymore.
There still are very small halogen bulbs that should not be much concern for breakage, but they can get very hot.  Something like these:
https://www.amazon.com/JKLcom-Halogen-Cabinet-Landscape-Lighting/dp/B07QYMQV6J/ref=sr_1_5

[edpalmer42]

A dim bulb tester shouldn't be compared to an SCR/TRIAC/etc.  The dim bulb limits the current with almost no effect on the voltage under normal circumstances.  An SCR/TRIAC limits the voltage by turning the supply off at some particular point in the AC waveform.  The TRIAC can, for this application, be compared to a Variac.

I agree that a halogen bulb isn't a fragile bulb, but there are some that have a normal glass envelope around a halogen bulb so they would be similar to a regular bulb.  But even then, I don't see breakage as a big concern.

Note that heat is a complete non-issue with bulbs in a dim-bulb tester.  The whole idea is that under normal circumstances there will be very little voltage dropped across the bulb.  If the light glows brightly, i.e. hot, you've got a fault in your circuit.  Also, this is only ussed for a short period of time.  It's not like it'll be left running like this for hours.

[james_s]

How about a magnetic amplifier? That's how the modern solid state airport runway lighting regulators work.

[tautech]

A dim bulb tester shouldn't be compared to an SCR/TRIAC/etc.  The dim bulb limits the current with almost no effect on the voltage under normal circumstances.  An SCR/TRIAC limits the voltage by turning the supply off at some particular point in the AC waveform.  The TRIAC can, for this application, be compared to a Variac.

I agree that a halogen bulb isn't a fragile bulb, but there are some that have a normal glass envelope around a halogen bulb so they would be similar to a regular bulb.  But even then, I don't see breakage as a big concern.

Note that heat is a complete non-issue with bulbs in a dim-bulb tester.  The whole idea is that under normal circumstances there will be very little voltage dropped across the bulb.  If the light glows brightly, i.e. hot, you've got a fault in your circuit.  Also, this is only ussed for a short period of time.  It's not like it'll be left running like this for hours.

Yep to all this ^^ 

Like a dim bulb tester is really hard to cobble together from scrap bits lying around. 

[t1d]

Thank you for your replies!

The alternistor was of interest, but not if it is not a current limiter.

I have been looking at DBT designs. How needful/helpful is it to be able to switch in additional bulbs to be able to bracket the current limit? Let's say the design goal is to be able to supply 4 amps (Yes, I know that is probably way plenty...) Would twelve switchable 40 watt bulbs be better than a single large wattage bulb. 12 x 40 = 480/120 = 4...

I was thinking to use 40 watt oven bulbs, because they are likely to continue to be manufactured, because LEDs are not usable in hot environments...

[bdunham7]

Any phase control method using a thyristor (SCR, Triac, etc) is going to be hard to control because the thyristor always turns completely on and doesn't turn off until the end of the cycle.  So if you have a short, you will have maximum fault current until the end of the cycle.  Then your control circuit has to decide when to try and turn back on again and how does it decide that?  If you have a totally unpredictable load and you want constant current, phase control is not for you.  As far as the MOSFET + bridge, keep in mind that unless you incorporate foldback, the MOSFET will have to dissipate the entire maximum available power, which in your case appears to be 360 watts.  Toasty!

If you just can't have bulbs, consider a PPTC.  They're available with sufficient ratings these days.

https://www.mouser.com/datasheet/2/643/ds-CP-0zre-series-1266352.pdf

[t1d]

If you just can't have bulbs, consider a PPTC.  They're available with sufficient ratings these days.

PTC = Clever! Thanks! I'm warming up to the G9 bulbs and sockets...

[edpalmer42]

I have been looking at DBT designs. How needful/helpful is it to be able to switch in additional bulbs to be able to bracket the current limit? Let's say the design goal is to be able to supply 4 amps (Yes, I know that is probably way plenty...) Would twelve switchable 40 watt bulbs be better than a single large wattage bulb. 12 x 40 = 480/120 = 4...

No, you're missing the point.  Under normal operating conditions, the dim bulb tester has almost no effect on the circuit.  For this to be true, the current drawn through the bulb(s) must be much less than the bulb's normal operating current so that there's only a faint glow visible.  I can't give you any intelligent numbers off the top of my head, but in your example of a circuit that normally draws 4 amps, you would need a bank of bulbs that had a normal current drain of perhaps 10 - 20 amps, i.e. 1000 - 2000 Watts.  This would give a current limit of 10 - 20 A.  A smaller bank of bulbs would run closer to its operating point and drop too much voltage, thus starving the device under test.

Since neither the bulb nor the device under test are characterized for this situation, YMMV.  The DBT is certainly not a piece of precision equipment so you have to be careful and make sure it's appropriate for your use.  I've heard of electronic circuits that function as fuses.  Something like that might be more useful than a DBT if it had the appropriate characteristics.  Actually, that would be a nice project if it hasn't already been designed!

My most recent use of a DBT was while working on a variable switching power supply that was rated for 0 - 33 V @ 0 - 33A.  There was no load on the output but I still needed a 100W bulb before the supply would even start.  The idle current drain was about 200ma but I had to use a 100W bulb that would normally draw about 800 ma.
 

[Zero999]

The purpose of a dim bulb tester is for devices which only draw a small current, such as a power supply, or amplifier with no load connected. It's not supposed to be used for devices which draw and appreciable current.

[tautech]

The purpose of a dim bulb tester is for devices which only draw a small current, such as a power supply, or amplifier with no load connected. It's not supposed to be used for devices which draw and appreciable current.

Yet I expressly built mine for old CRO's when repairing them so to retain the magic smoke when powering them after repair. More recent use is for SMPS repair for exactly the same reasoning.
They earn their place on the repair bench.

[t1d]

Thanks, Ed, Zero and tautech...

But, I must admit that I do not understand the explanations... I thought that the bulb would pass current, until the current draw surpassed the bulbs abilities... That is, with a direct short fault in the DUT and the fault drawing more than the bulb wattage divided by the supply volts (that is, amps,) the bulb would blow. Like a fuse. Maybe a fuse is the better option? I thought the bulb more useful, because you can see the problem in the glow and you can back down the autotransformer, before cooking more components...

[bdunham7]

I thought that the bulb would pass current, until the current draw surpassed the bulbs abilities... That is, with a direct short fault in the DUT and the fault drawing more than the bulb wattage divided by the supply volts (that is, amps,) the bulb would blow. Like a fuse.

There is a flaw in your basic understanding of electricity here.  Surprisingly, people who should know better make the same mistake repeatedly when discussing fuses, breakers and fault currents.  Assuming the bulb is rated for the supply voltage (presumably 120 volts), the DUT cannot draw more current than the bulb.  At low currents, there is very little voltage drop across the bulb and the DUT gets normal voltage.  As soon as the current increases enough to warm the filament of the bulb, it gets hot and the voltage across the bulb increases while the voltage across the DUT decreases.  If the DUT is a dead short, the bulb just glows (not blows) the same as it would if it were installed in a lamp.  There's nothing the DUT can do (unless it supplies power) that would cause more current to flow in the circuit than the normal operating current of the bulb.

[tautech]

Thanks, Ed, Zero and tautech...

But, I must admit that I do not understand the explanations... I thought that the bulb would pass current, until the current draw surpassed the bulbs abilities... That is, with a direct short fault in the DUT and the fault drawing more than the bulb wattage divided by the supply volts (that is, amps,) the bulb would blow. Like a fuse. Maybe a fuse is the better option? I thought the bulb more useful, because you can see the problem in the glow and you can back down the autotransformer, before cooking more components...

A bulb is NOT a fuse, period !
It is a temperature dependent resistive element.
When used in series little current draw (read quiescent) will make it glow dimly while a fault or short circuit in the DUT will cause to to glow at full brightness.
We use the DBT as a visual indication that the DUT is operating correctly.

Cobble one together and plug various appliances into it to get a feel for its usefulness. Most loads will glow brightly at first while magnetizing currents are drawn or in the case of anything with SMPS the bulk cap charges.
Once the initial current inrush has passed if all is well and a correct wattage bulb has been used it should only glow dimly......dim bulb tester. 

[edpalmer42]

The purpose of a DBT is to keep the magic smoke from being released from the DUT.  It's used as a service aid when repairing a piece of equipment or on an initial evaluation of a piece of equipment that is of unknown or suspect state of health.  It works due to the unusual current/temperature characteristics of an incandescent (regular or halogen) bulb.

When an incandescent bulb is cold, the resistance is very low - in the range of a few ohms.  As current flows through it, the filament heats up and the resistance rises.  As the resistance rises, the current drops.  Eventually, a stable point is reached.  If the DUT has some sort of internal short, the bulb will glow brightly.  If the DUT is working properly, the bulb will initially glow brightly for a second or two as internal capacitors charge, but will then dim to almost nothing.  As long as the bulb is rated to operate at the supply voltage, it will last for many, many years.

By limiting the current, the DBT prevents fuses from blowing and allows circuit tracing to locate the short without burning traces, components, etc.  The cost savings from not replacing multiple blown fuses can be significant all by itself - never mind the frustration of trying to troubleshoot a problem that only allows the circuit to be active for a few seconds at a time.

I suggest that you do a Youtube search for 'dim bulb tester'.  There are many videos of various quality.  Watch a few to get a good idea of how a DBT can be used.  Note that some of them combine DBT, Variac, and isolation transformers.  This is actually a good idea, but isn't necessary if all you want is a DBT.

[t1d]

There is a flaw in your basic understanding of electricity here.  Surprisingly, people who should know better make the same mistake repeatedly when discussing fuses, breakers and fault currents.  Assuming the bulb is rated for the supply voltage (presumably 120 volts), the DUT cannot draw more current than the bulb.  At low currents, there is very little voltage drop across the bulb and the DUT gets normal voltage.  As soon as the current increases enough to warm the filament of the bulb, it gets hot and the voltage across the bulb increases while the voltage across the DUT decreases.  If the DUT is a dead short, the bulb just glows (not blows) the same as it would if it were installed in a lamp.  There's nothing the DUT can do (unless it supplies power) that would cause more current to flow in the circuit than the normal operating current of the bulb.

Ah, Mr. Dunham, you are so very correct. At some time, I learned these very details. (I did research all three assemblies; DBT, Isolation transformer and Autotransformer.) However, it is very clear that I forgot ever point, regarding DBTs. My apologies. I will be copying your explanation and posting it on my bench, until it has fully sunk in. Thank you for your patience.

My autotransformer just arrived with the mail. It appears that the unit is in great condition and that I did not get ripped off on Ebay. Terrific!

So, now I have all the needed components... The DIY Iso Trans, the Autotransformer and a Ceramic Bulb Fixture/60W Bulb... Please confirm that their proper arrangement is as follows: House Mains to DIY Iso Trans to Autotransformer to DBT to DUT. If I have that correctly, I think we can call this one done.

Thanks to all for your wonderful help!

[tautech]

Just remember to wire the DBT as a series element of your setup. 

[t1d]

Yep, in series... Thanks!

[bdunham7]

So, now I have all the needed components... The DIY Iso Trans, the Autotransformer and a Ceramic Bulb Fixture/60W Bulb... Please confirm that their proper arrangement is as follows: House Mains to DIY Iso Trans to Autotransformer to DBT to DUT. If I have that correctly, I think we can call this one done.

I would put the autotransformer first to minimize the losses in the isolation transformer and to minimize leakage.  The bulbs could go either before or after the isolation transformers (see Mr. Carlson's lab) but they would typically be afterwards. 

[kallek]

In my similar project I wired variac before isolation transformer. I would wire more powerfull unit first from mains.

With two bulb sockets and different wattage bulbs it's easy to change current limit by screwing bulbs tight or loose.

[t1d]

My thinking was that having the Iso Trans first would protect everything in the chain, afterward... Including my new Autotrans bride; we are still honeymooning.  But, having the Autotrans first, to limit losses, has good merit, too.

As for using multiple bulbs, are they placed in series, or parallel? My first guess would have been in series, but, if they can be brought on line by screwing them in, that has got to be in parallel, correct?

My Autotrans can boost 120V to 140V, so that will help with the voltage droop caused by the load.

[tautech]

As for using multiple bulbs, are they placed in series, or parallel? My first guess would have been in series, but, if they can be brought on line by screwing them in, that has got to be in parallel, correct?

Correct, multiple bulbs if used in parallel and all part of the in series mains supply.