Resistors getting so so...Hot :o

Thread Starter

RIKRIK

Joined Oct 11, 2019
25
Basically, I have a circuit, square wave inverter, I have 4 1/2w 270R resistors , which are getting too hot too touch. But i can leave the circuit on for hours and the resistors havnt burned out. I am wondering if im missing something, and if so do i need some higher wattage resistors . Iv tested with one bulb and they get as hot as all three bulbs.
details below.

kind regards,
Rick

B1.pngB2.pngB4.jpgB3.jpg
 

SamR

Joined Mar 19, 2019
1,304
First off those are not incandescent bulbs so forget about their 300W load effect since LEDs are used to reduce wattage lost as heat. Secondly, what is the power calculation wattage for the circuit and does the resistor watt rating support it?
 

Dodgydave

Joined Jun 22, 2012
8,476
The resistors in question, are they R1, R4, if so the Wattage is 530mW each at 12V, i would use 1W,, ideally measure the current in the Collector load of the transistors.

What is happening to R7, R8 resistors?
 
Last edited:

Wolframore

Joined Jan 21, 2019
1,322
Have you considered rebiasing for less current? The circuit is only providing switching voltage for the mosfets which doesn’t need any current to switch. You’re burning up a lot of power... theoretically up to 600mA or more while the transistors are on.
 
Last edited:

bertus

Joined Apr 5, 2008
20,109
Hello,

@Tonyr1084 , In the simulation he uses a lamp as load.
I think that in real life he uses the transformer on that spot.
It is a crude inverter that he is making.

Bertus
 

Thread Starter

RIKRIK

Joined Oct 11, 2019
25
Hi thanks for your replies back, yep the LEDs are just there to show the on and off of the switching circuit. The single 3 terminal block at the bottom goes to the 12 0 12 of the transformer, the other two 3pin terminal blocks at the top goes to two MOSFETs. The lights are 9w led bulbs.
 

Janis59

Joined Aug 21, 2017
819
Generally, some changes was been happened to resistor materials, if 80~ies soviet resistors of 2 W had maximum T around 60C then nowadays some (but not all) 20W reistors are having even a smaller physical sizes but their normal temperature may happen 150-180C and all is OK. So, the only complication is must to be distance between such high-T resistor and pcb, at least 1 or 2 cm.
 

Thread Starter

RIKRIK

Joined Oct 11, 2019
25
Tbh a lot of the circuit has been trial and error, this is my 3rd version of the prototype, it's to go with a low battery indicator of been working on, still working out my last issue the resistors, which fingers crossed after all your good advice hopefully I should resolve. compared to modified sine wave and pure sign wave, it isn't going to compete , but as square wave inverters go , it's cheap, with good response over a unregulated input. Better than a 4047.

This my friend is a crude inverter
https://www.electronicshub.org/12v-dc-220v-ac-converter-circuit/?amp
 

Thread Starter

RIKRIK

Joined Oct 11, 2019
25
Okay had to change the fuse in my volt meter, and my circuit doesn't seem to like being probed at, magic smoke, 12.5v battery, .3 across the diode, 11.70 across 3 of the the resistors, at 280mah, however I think I damaged one resistor while probing with reading 12.2v across it.
 

Tonyr1084

Joined Sep 24, 2015
3,676
11.7V ÷ 270Ω = 43mA (0.043A)
43mA x 11.7V = 507mW (0.507W)

That's a fair amount of heat. You need to control that.

No matter what - you have that much heat to dissipate. Whether you use a 1 watt resistor or a 5 watt resistor, the heat is the same. The DIFFERENCE is in how much area that heat is dissipated over. Obviously a 5 watt resistor is going to feel cooler because the heat energy is spread over a larger surface area. 5 watts is overkill; but I mention it just to make a point.
 

MisterBill2

Joined Jan 23, 2018
4,065
The big question that I have is why are the collector resistors 270 ohms? The gates do not need that much drive current. 2700 ohms should work as well and the resistors will not get so hot. That value might be intended for an inverter using bipolar transistors that need a lot of base drive, where it may be a reasonable value.
 

Dodgydave

Joined Jun 22, 2012
8,476
If you're going to make an inverter,I would use a CD4047 and set it to 50/60Hz with mosfet outputs, the problem is setting the output voltage feedback.

Better still but a readymade PCB...
inverter_100w_12v-220v_by_ic-4047_irf540.png
 

Thread Starter

RIKRIK

Joined Oct 11, 2019
25
Thank-you for the calculation, so at 13v, full battery voltage, 12.7 / 270 = 47 . 12.7×47= 597ma , so maybe even a 2w waiting on some 1w and 2w resistors. 2.4a for all four resistors, geeze, may have to bump up my battery capacity, IV found a good deal on a 35ah battery, a guesstimate, say if my circuit uses 35w, and my 3 9w bulbs use 27w, at 20% efficiency that's 5.4w,so 32.4w+35w=67.4w. ÷12=5.6a, 35ah ÷ 5.6= 6.25hours all three bulbs, one bulb would be 9.1hours.

What would be the best way to measure the efficiency , as I googled square wave inverter efficiency and it said 20%, however this circuit works differently.
 

Thread Starter

RIKRIK

Joined Oct 11, 2019
25
The 270R resistors set the hz's of the ossilator, changing them would mean I would need a higher ceramic cap value, and there hard to find, I did see some where about using Darlington transistors, to resolve the issue, let me just find it
 

MrSoftware

Joined Oct 29, 2013
1,576
Efficiency is power-out / point-in. So measure your RMS output voltage and current and divide by your RMS (DC) input voltage and current. It should be a strait forward measurement if you have a true RMS volt and current meter.
 

Thread Starter

RIKRIK

Joined Oct 11, 2019
25
Okay this is what it does. Il do a simulation later.

The maximum usable R1-R2 values are, in fact, limited to hfe x R3 (or R4), and one obvious way of improving the waveforms is to replace Q1 and Q2 with Darlington connected pairs of transistors and then use very large R1 and R2 values, as in the Figure 4 circuit, in which R1 and R2 can have values up to 12M, and the circuit can use any supply from 3 V to 18 VNV_1203_Marston_Figure004.jpg
 
Top