dropping AC voltage on secondary

[algorithm]

Im struggling to devise a way to decrease ac output on transformer secondary.
Merely need to drop 9v. From 33v ac.
120v 60hz is going in. But I noticed this transformers label only says 110 60hz.
Its supposed to output 24, but its 33. But if my line from the wall was 110? Rather than 120. Alas without  variac. Im unsure what to do.

I could totally cobble a capacitive dropper but I dont know how to do the math. Lol
But I do have the parts. But when I try to research how to use those formulas its written in ways I dont understand. Or its a calculator where Im guessing on what value of x cap to use.

But 33v im pretty sure would fry an ic so I feel I need to do something about it.
Although the board Im sure would have drop across it, would it drop enough before the ic got fried?
Another option I thought about was get a transformer wound for 120.

Any ideas or direction is appreciated.

[Kleinstein]

Transformers have there rated voltage, when at the rated load. With no or only light load the voltage will be higher. How much depends on the size and type of transformer. With relatively small EI type transformers this can be quite a bit higher, sometimes 50%.

Most ICs don't use the AC voltage directly, but DC after a rectifier. Limiting the voltage may be easier at the DC level than at AC.
A series resistor or inductor could be an option in the AC path, but it depends on the load current. So one may still need a voltage limit at the DC side.   

[tunk]

What will you use it for, or how much current do you need?
Also note that after rectification and smoothing, you have
to multiply the voltage with the square root of two, i.e.
33*1.41=46Vdc.

[algorithm]

Not sure exactly what class transformer it is.

"Power transformer
Model: YB-E166*28F
Input: 110V/60Hz RED RED
OUTPUT: 24v.        BLU BLU"

Its a yihua 936 transformer.
The 24v (ac?) Feeds the heating element directly, or in my case 33v?. But the lm358 max voltage is 32 so Im scared to hook it up. Although I ran the station for a year before it failed initially, I didnt test the secondary then.

In eevblog episode 596 worlds cheapest soldering station you can see the board. No exact schematic that I could find. But almost. Its using a bt137 and a lm358.

If there was a way I could test the resistance of pcb to calculate voltage drop that would help. Is that a thing?


Well looking at the pcb. 33v comes in, 1 side directly to heating element, that has a diode on the trace between the heater output. And another diode at other end of trace leading to 220uf 50v cap. The other ac directly into pin 1 of bt137 and r1 (330 ohm 5%) and a 1n4007.  Then r13 is off that trace into 2n3906. R1 and 1n4007 on the other side trace directly to lm358. I dont know if it would be sound to hook it up.

I know its a cheap station but If I can make a piece of junk lying around work again its nice. Nice to have a spare.

Thanks again.

I thought about using a rheostat to drop voltage, but thats not very practical or efficient and no space for a huge ceramic glob. But I do have an old ceramite rheostat in junk boxes.

[algorithm]

Well tunk its a yihua 936. Sticker on back says 1A. But transformer doesnt mention it. I unfortunately dont have a killawatt meter.


Edit: https://www.vintage-radio.com/repair-restore-information/valve_dropper-calcs.html seems like a good resource.

[Kleinstein]

Chances are the heater element would survive the 33 V. It would be only the LM358 to have potential problems. Chances are the amplifier circuit does not need the full voltage at all. So it does not matter if the voltage drop well more than to 30 V (chances are even 12 V would be OK). So one would need the resistor only for the amplifier current, which managable woth a resistor. If paranoid add a zener / transsil diode for a clamp at some 30 or 25 V.  As a crude goess expect some 5-20 mA for the amplifier including the current to drive the power stage. So some 1-2 K ohms could work and would not create excessive heat.  It would still help to draw a schematics to get a better guess and  find the right position for the resistor.

For better tolerance to higher voltage one could replace the LM358 with a MC34172 as a rather similar OP that can tolerate up to 44 V. It may however be tricky to find the DIP version.

[tunk]

Or maybe 2-3 highish mA LEDs in series, or if you can find
a 12V LED bulb with three LEDs in series with a resistor.

[Bud]

Measure the heating element DC resistance. Find a power resistor with the nominal close to what you have measured, or compose it from several resistors. Connect the resistor across the secondary and measure the AC voltage under this load. Chances are it will be 24v or close.

[james_s]

You can buck the input voltage with another transformer, for example use a 120V:12V transformer to drop 120V to 108V and feed that to the primary. Are you sure the higher voltage is a problem though? It will sag under load by some amount.

[tunk]

Maybe you could post photos of the board (both sides)?
And try to see if it's anything like this:
http://www.next.gr/circuits/Hakko-936-ceramic-heater-A1321-l40041.html

[algorithm]

Cool, now we are getting somewhere. I had the idea to put the heating element on the secondary output just to see the drop. Kleinstein must have been clairvoyant there. Will report back after I try that. Gonna take some janky alligator clipping im afraid but should help put my mind at ease that Im not going to lectric chair the lm358. Main reason being I dont have more 358s. So if I fry it well... 1 of my favorite youtube comments was "bullshit! My chip got fries!!" And some wiseguy replied "did it get a coke too?"

I found an older topic where someone else noticed their transformer outputting 33, they were worried of potential stress, life decrease on the heating element, surge, etc. That poster mentioned with 2 resistor in series with iron heater it was 24v, but noted that without iron itd be 33v. The resistors he used were 1.5 and 2.2 ohm in series I think. https://www.eevblog.com/forum/beginners/28vac-transforer-to-24vac-possible/ this was a highlight from bob91343

All good comments.  The distillation is:

1. Get the correct transformer.
2. Use a dropping resistor.
3. Take a chance and use excessive voltage.
4. Modify the new transformer.

Any of these three will work with varying degrees of success.

Also found this site, I played around and got a circuit that allegedly would turn 33vac at 60hz to 26. But available nominal curreng at 1.59mA. Is that like 1 whole amp? Lol
http://www.nomad.ee/micros/transformerless/index.shtml

[BrokenYugo]

Work again? What failed? Where are these parts from? If they're made to work together I'd expect them to, at least for a while. I'd power it up and measure. Work case you smoke one of the easiest to source op amps on the market.

Note that these work by power cycling the heater quite slowly, the heater load will only pull the voltage down intermittently. The control circuit needs it's own regulation if it needs to be regulated. 

[DavidAlfa]

120-110V difference is a mere 10%, should output ~27V, but the problem is you're measuring it open-circuit.
A transformer output is sustantially higher when unloaded, ex. 12V could easily get into 18.
Try adding a small load, ex 470Ohms, 1K... will get rid of some volts.

If the electronics can stand 33V, shouldn't be a big concern, should drop a lot with the heater.
Connect the heater, just for 1-2 seconds (It won't explode!) and check the voltage.
I bet it'll be much closer to 24 than 33 is.

[floobydust]

OP this is a cheap chinese power transformer, so it has the minimum copper and steel necessary. It means the regulation will be quite poor - high output voltage at no or light loads, and hopefully the right voltage 24VAC with the heater load. Because the heater is not on 100% of the time, you can even make the transformer cheaper i.e. undersize/overload it and well, what did you expect for the price?

I'd just add a zener to lower the B+ for the op-amps. It should already have one, maybe post pics of the board. Other 936 clones use 12VDC.

[algorithm]

Well I managed to hook up the element, but testing resulted in nothing on the meter for some reason. Using a fluke 12. Double checked I was in ac mode. Heh my luck. But yeah across both pins going into heating element, got 0.xx v. Wonder why I couldnt read that. Could my element be busted?

Well I guess I will just hook it all up flip the switch and hope nothing cooks again.

So yeah this was an early model yihua 936, my first adjustable iron, so it was really special to me in some weird way. My board is green and labeled "YH-936AV22-3". Back when I first got it it being my introduction to adjustable temp I used it for a year at least. But was using it one morning and smelled smoke, it stopped heating, led wasnt blinking. Took it apart and a few resistors were burnt. CaNt recall exactly which ones originally, but I ordered bt137 and just tried to shotgun it like an idiot. Recapped, replaced resistors, triac, normal diodes, and turned it on with board exposed, got to see the resistor turn black and smoke, took a deep breath. (Lol) I think I tried that twice, but failed to even glance at thd lm358. Still not sure what the right way to test that would be? but I had found a hindi video on the boobtube about testing lm358 with dmm. Of course I couldnt understand a word, he put meter in diode mode and he test from 4th pin + and - on 5 6 7 and 1 2 3, all had about max reading of average diode junction (0.8xx). Well the lm358 I took off on my recent look had 2.000 diode drop on each pin and no beep. So im going to guess that was the cause and I tried fixing it, failed gave up, then found it and figured to try again.

Last attempt resistors that burnt were
R9-75 ohm
R7-brown black green gold
R2-150 ohm
R1-330 ohm

So yeah what Ive put into this far exceeds its value lol, but I have nothing better to do than mess with cheap junk.
Replaced bt137 twice, those resistors twice,  few diodes once, and then the lm358 a year or 2 late.

Im thankful for the attention and response from each and every one of you.

And I like the ingenuity of the idea of using leds in series to drop voltage.

[algorithm]

https://www.reddit.com/r/AskElectronics/comments/5cpral/adding_onoff_led_to_soldering_iron/

Thats what the board is like.
I just hooked it all back up yesterday and powered it. It burnt r1 and tried to do r2. So I used two resistors to increase wattage, it burnt that way too.

On the flir I saw 275 celcius on r1. It was charred black. It stunk strongly.

Zd1 and zd2 both when taken out and tested read 7.47v zener.
But the hakko (what this yihua is a flattering clone of) uses 5.1v I think?
Could that zener value have cooked r1?

Also noticed both c2 c3 47uf caps were 25v.
Could that be why r1 toasted? You know, since 33v? But the caps out of circuit both tested good on esr and capacitance.

[ledtester]

Im struggling to devise a way to decrease ac output on transformer secondary. ...

By any chance is it a toroidal transformer?

[algorithm]

I believe its toroidal





Are the zeners or caps culprit ya think

[ledtester]

I believe its toroidal

By torroidal I mean does it look like this (with a hole in the middle)?

[algorithm]

looks like that minus yellow wires.

But can anybody theorize why r1 and r2 roasted over and over?

Could 25v cap on far side of r1 and diode do it?
Or perhaps a 7v zener where 5 belonged?

I want some sort of assurance before risking burning more resistors, I cant take more of that smell.


I did think for the better part kf the year I ran this it was probably ungrounded on earth. Could that do something? But ive replaced almost every part possible except transformer and some resistors.

[ledtester]

Ok  - unfortunately that's not a torroidal transformer.

[algorithm]

When I tested the 4 zeners, I got
Z1-6.04v
Z2-2.51v
Z3-1.04v
Z4-3.27v

Then I replaced z1 z2 with 5.1
(By the way, those zeners had c7v5 printed on em, so now Im left wondering is that supposed to be 7.47v or 5.1)

But after replacing those my identical test went
Z1-5.11v
Z2-5.02v
Z3-2.11v
Z4-3.67v

Not sure why z2 was wonky in circuit on that first test. The diode tested good out of circuit. Well whats it rated to be it seemed. (7.47 aka c7v5 5T)

Maybe some swarf was stuck where the sun couldnt shine.
I really dont know. But im confused because nothing seems wrong. Ive tried cleaning and scraping between traces in case but it doesnt appear that bad...

Can anybody guess on what difference 5.1v vs 7.5v would make?

I also noticed on my original handle the element has 17ohm but another new replacement spare iron for 936 I have kicking around has 12 ohms on element.

[tunk]

Please post photos of your board, not something you found on the net.
No experience with zeners, but I guess if the voltage is measured in
circuit and powered on, then something in parallel with the zener
drags the voltage down. This could also explain your burned resistor.

[algorithm]

I wish I could post a photo of my board but no camera and no pc. This tablet im stuck using has no camera.

Well I should add z4 has c3v9 printed on it. And I can say at least z3 has a c2 designation. So they all seem to test right. Im feeling a bit more confident but still have no idea what changed. Maybe a bad joint? Hmm 

Getting close to risking burnjng more resistors. But I had an idea, instead of watchjng it burn. I thought "why not test the voltage drop across the resistor powering long enough to read and shut it off before any charring. and then determine the amp draw ?" Well that way if it will burn I can prevent that awful smell.
Wish me luck. Gonna try this again I think.

[james_s]

How do you not have access to a camera? They have been part of every mobile phone I've ever owned, including the very cheapest flip phone I was able to buy back in 2007. Cast off PCs are easy to find too, they get scrapped every day due to having such low value. Do you have a thrift store nearby? I have donated several digital cameras to those here and I see them on the shelves frequently.