overvoltage to a circuit board lamp

[eccerr0r]

Re: overvoltage to a circuit board lamp

It's really getting confusing here, you have some statements that are either inconclusive, hard to trace, or don't make sense.

From what you're saying, the output to the lamp is PWM at 91% duty cycle and 400Hz. This is way too high duty cycle to safely power the 12V21W lamp, but perhaps this is this way because you're testing without the lamp installed. If the duty cycle goes down with the lamp installed, then theoretically the device is "working as expected" though a bit high in voltage. The ideal change would be a firmware change which is likely not possible.

With the duty cycle at 91%, the voltage of 1.6V makes absolutely no sense. The multimeter should be doing an average of the highs and lows of some sort (not known unless you have a true RMS meter) but in any case the MOSFET needs usually above 5-6 volts before it will actually turn on and light the bulb.

The "540r" resistor I have no idea what you're talking about. Draw a circle around it on one of the photographs or better yet, the schematic. 540Ω is not a 5% standard resistance value and there's no R540s that I can see. There is a R54 which is 4K7 or 4.7KΩ but that does not seem to be part of the power circuitry despite it being rated ½W.

[Cornboats]

Re: overvoltage to a circuit board lamp

I'm really sorry for any confusing info- its my in-experience in trying to read the schematic and understand how this board works-.
I was reluctant to put the lamp in but I'll give it a go.
Regarding the 540r resistor. Its a typo which I missed.
Its a 470r, 3 watt. This is located adjacent to cap 35v 2200. Yellow,purple, brown, gold. I can't see it on the schematic. Its in the purple feed from the transformer. From the trans it goes through a diode then onto the resistor, It links into 35v 2200 cap,then onto the motor. Before the resistor voltage is 33v after to the motor is 13.5V.
This is live all the time and gets really hot-Due to its discolouring I changed it before Xmas to a 3watt 470r ceramic.

[Cornboats]

Re: overvoltage to a circuit board lamp

I've taken some scope traces of the source and gate for IRF540N fet connected to the courtesey lamp circuit.(not the flashing lamp), Both traces were taken @2ms/div.
I make the gate on 1.5 divs(ms) & off .25 divs(ms) x 2.2divs x 0.2v. I can't see the bottom of the wave so I'm just guessing- even tried swopping over probes and scope inputs but to no avail.
& the source on .40 divs(ms) & off 2.90 divs(ms) x 5.25 x 0.5v
I've used 35v 2200 negative as a reference.

The voltage taken on a multimeter across the illuminated lamp was 4.95v dropping to 3.75v when the motor was off. The light goes off about a minute after the motor does.

For the time I was testing R3 0.47r appears to be cool but as mentioned in above posts R8 to me is uncomfortably hot after a only 5 mins.

Any comments would be welcome

[R_J]

Re: overvoltage to a circuit board lamp

R8 Is used to drop the voltage from 33 volts to 13 volts, That causes heat. this is normal but if your are not familiar with electronics this may not seem right but it is completely normal. It is droping about 20 volts, so the current is about 43ma which is just shy of 1 watt. the resistor is a 2 watt which is double what is needed, the type of resistor used has changed color over the years but thats all.
You need to accept the fact that the lamp just burnt out and all that is needed is for it to be replaced.

[eccerr0r]

Re: overvoltage to a circuit board lamp

Well, since the light is turning on and off as expected, the microcontroller is working and there's nothing we can do about the duty cycle, as that's what it is.

R8, towards the left of the schematic that feeds the 13V1W Zener D1, is specced at 470Ω 2W. This sounds like it should get quite toasty as it's a zener regulator, depending on the voltage drop across the resistor. If the voltage across the zener DZ1 is 13V, then we can assume that circuit is working (though it would seem that C13 2200uF 35V would be very close to its voltage limit...)

There's still some confusing statements and pictures here. I would expect the "nicesource.JPG" with the small spikes to be "correct" but this doesn't match "nicegate.JPG" - both of these should be in step and similar shape with each other except the voltage swing should be much higher on the gate. BTW, could you mark which transistor (MOS1 or MOS2?) and what the gate waveforms are for each of these? The source for both MOS1 and MOS2 should look identical as they are connected together.

R3 0.47Ω3W should remain cool unlike R8, if LAMP SPLA P.A. E.BLOCCO and LP1 are both off.

At this point I would have to agree with R_J that this is what it is, can't really do much other than change the light bulb every 24 hours it's on, or whenever it burns out. I don't think there's anything easy to change to make it last longer other than opting for a higher voltage lamp. Then the other question is if the brightness of the lamp is any different than what it was when it was new? At 15V it would be quite bright and noticeable when new too...

---

BTW, your oscilloscope should have a switch that changes input coupling from AC/DC/GND. You can use GND coupling - which shorts out the input to the vertical amplifiers - and get a GND point. With it at GND, move the trace to one of the graticules and use that as the 0 reference. (If you're not sure about how to say the settings of the 'scope, it would be great if you can take the picture with the knob settings as well.)

[R_J]

Re: overvoltage to a circuit board lamp

I suspect that once the new lamp is replaced it will last at least as long as the original, likely for years.

[Cornboats]

Re: overvoltage to a circuit board lamp

The scope traces are for mos 2 only. I didn't bother with mos 1 as this is for a flashing light that is an external option. I need to programe the board if this is to be connected.
If you want this done I can do it but may take a bit of time. Here's pics of my scope and the settings used for the traces.
If you can talk me thro the setting s required to give better results I'd appreciate it.

There's no point me accepting that things are ok with a new lamp in. The damage to the capacitors on the board in my opening thread surely means that the heat from the 21w lamp and or resistor r8 was probably the cause and will be repeated if not rectified. atm my action would be to remove the lamp fitting and use the operator lamp free.
Lamp or not r_j has made the point that r8 should cope with the heat. Will a 470r 3watt ceramic absorb the heat better without causing any problems?


Thks for your continued involvement its much appreciated.

[Cornboats]

Re: overvoltage to a circuit board lamp

I've followed your instructions regarding ground coupling in order to line up the gratiscule to allow easier reading.
attached are the latest trace results for both Mos1 and Mos. Mos 1 has no load as it for the external flashing light. Mos2 has the 12v 21w bulb as load.

settings for all traces are .5v, 2m/s

If I press chop on my scope both traces gate and source traces are overlaid, the base looks exactly the same which is what you were expecting. I can't tell about the height as for some reason both fets gate traces only show as per pic.

I've put a temperature detector on r8. After 30 mins its 80 deg c. Thats with the board open on the bench with air movement based on a room temp of 20 degrees.

[eccerr0r]

Re: overvoltage to a circuit board lamp

Ok yeah that looks better, they should look like pulses. So it's working as expected as far as I can tell. You should be able to see the tops of the gate pulses, but since the pulses should go all the way to 13V, at 0.5V/div it would be 6.5 divisions that may go off the top of the screen (again, if 'scope is calibrated).

As for R8 there's not much you can do. Sticking in a higher wattage resistor will spread out the heat but it will still generate that heat. You may be able to experiment with a slightly higher value resistor at the risk of breaking something (say 560Ω) but I suspect they chose 470Ω for a reason. What's the voltage across that resistor?

[Cornboats]

Re: overvoltage to a circuit board lamp

When R8 was at 80 degrees c I measured the voltage across it as 21v, before 36v, after 15v.
Looking at the drawing this line ends at TP4- can't see TP4 anywhere else- but my multimeter shows its in common with brown going to the motor. The motor needing forward and reverse, as one direction is fed off the bridge rectifier I'm assuming that this is required for the other direction. Can you see the motor connections anywhere? - just interested in knowing.

[R_J]

Re: overvoltage to a circuit board lamp

TP4 is ground. The motor connection is (tp43 & tp44) Positive voltage is applied to one side of the motor or the other, depending on which relay (rl2 or rl3) is energized, the ground side is (pwm) controlled by mos3 through R7 (current sense resistor)to ground.
In the off condition, the relays short the motor winding to provide dynamic breaking

[Cornboats]

Re: overvoltage to a circuit board lamp

Thnks for your explanation. It certainly helps me understand the circuit a lot more.

[Cornboats]

Re: overvoltage to a circuit board lamp

Took another scope trace of mos 2 gate. Changed divs to 5V , now the height is visible.




I changed R8 470r from 2watt to 3watt ceramic. Its a tight fit but its operating temp has come down from 80 to 55 deg c.
I'm not going to try and run the led lamp. Consensus of opinion says that messing with the line by altering the resistor is going to mess up the the operation somewhere else. I think my best cause of action is to remove the light fitting so it can't be used as the 21w lamp and the R8 resistor caused heat damage to the board. I don't want to keep being called back to this one.
Many thanks for everyones contributions. Whilst not completely resolved its given me more confidence in trying to sort out problems rather than just saying -"oh throw the board out and put a new 1 in."