Extremely hot resistor

[syn]

Hi

   I'm using this power supply, (Gyraf G9 style) to power a tube circuit that I'm playing with right now.
All is well apart from the fact that R33 gets extremely hot (can not be touched for more than a couple of seconds) and I'm using 15W part now (r33,470r,15W and extremely hot). 2W part was "untouchable". I've calculated from the V and A (given on the schematic) that even 1W part could do the job. All V and A on the schemo, are correct values, measured in the actual circuit. Also ripple is HUGE. When not regulating (no load to the TL783 output) I have 334V after the bridge,before the R33.
Any help greatly appreciated.
UPDATE
I will post relevant values and updates here, in the first post, so you don't have to go through the whole thread:
R37 V drop is~12V and it passes ~22mA (~250V before R37 ~238V after R37). It is getting barely warm (not hot at all).Loaded.
I changed all the HT caps- to no change- R 33 still burns, RIPPLE is still HUGE.I used 4x bigger caps- no change.
I changed 15W R33 to another one(same value)- again - nothing changed.
I'm using "true rms" (UNI-T/ UT60H) meter.
Anyone? The question is still the same...

 

[Walrus]

C14 leaky or backwards maybe.

 

[Mark Burnley]

What's the volt drop across R33 with and without the load applied?

Mark

 

[syn]

Caps are brand new Nichicon, no backwards, checked ~50 times
Voltage drop without the load was very small if I recall correctly like 1-2V if so i might be wrong...
With load applied it is ~22V.

 

[syn]

Yes, without, the load it drops ~2V and current through it, is ~3.5mA.

 

[Mark Burnley]

Have you included the R37 470R/2W resistor in your PSU?

If so, is this getting hot? What's the volt drop?

Mark

 

[syn]

Yes i have,R37 V drop is~2V (~250V before R37 ~238V after R37). It is getting barely warm (not hot at all).Loaded.

 

[Walrus]

That doesn't make any sense as 22 volts across 470 Ohms is about 1W. What type of meter are you using? The only thing I can think of is the ripple is enormous and actual peak to peak volts is much more than 22 volts hence higher dissipation. Have you got a scope to check the waveform on C14?

 

[Mark Burnley]

R37 V drop is~2V (~250V before R37 ~238V after R37)

Check your readings and/or arithmetic here!

Mark

 

[Mark Burnley]

...I'd be tempted to measure all your R values around the regulator,

and also make sure the diodes and regulator are all working/correctly installed!

Mark

 

[syn]

Thank you so much on your kind help. I rechecked and all the above is correct (regarding r37-(unsoldered=470r), before it ~250V after it ~238~240VDC). All the components are installed correctly, and are working, right value, type, orientation as well. I have large 100Hz ripple (RMAA -40dB ref. to 1.227VAC), and  I think it has something to do with this situation, but I don't know where is coming from. Caps are all new .All grounds connected, checked, to the star. My scope is packed as I'm about to change flat. I'm puling my hair off, on this one for the last couple of days, and everything is checked to the point that I unsoldered and measured the values of the each and every component at least twice, their orientation as well many times. I know it sounds crazy, but this is the way it is. I have a number of smaller caps in parallel, so it is not easy to pull out just one and to see if it is problem. But if everything else looks good to you, I'll pull them out and replace all of them. Once again, RIPPLE IS HUGE.

 

[syn]

Mark
        I just noted that I missed 1 in front of 2 for r37 reading.It is ~12V drop not 2 volts as I wrote above.Sorry.

 

[rotation]

Hi Miloš,

i also had problems with this psu, so i'm using this one now:
http://www.electronicsinfoline.com/Projects/Electronics/Audio_Project/Audio_Amplifier/regulated_vacuum_tube_power_supply.shtml
Caps can be much much smaller because cap from TIP50 base works as a capacitance multiplier. It will also work well with different voltages and currents. If you don't drop too much volts and need a lot of current you don't need any heatsink (i can touch any part of it at 300V/20mA and about 30V drop). I really like this psu, it's quite similar to what D.W.Fearn use in VT1/2..

I'm not saying Gyraf psu is bad, many people used it for G9 without problems. Maybe it just don't like abuse and p2p i like so much.

Miha

 

[syn]

Thanks Miha I'll check that one in the future. I want to sort this one out, as it really pisses me off. Hopefully I'll learn something along the way.

 

[warpie]

sorry to hijack the threat but can please someone explain me how can I calculate the maximum power for the R33 (in this case)?

Aplying the formula W = (VxV) / R doesn't really work in this case, does it? 

thanks

 

[jdbakker]

Aplying the formula W = (VxV) / R doesn't really work in this case, does it?  

...because?

JD 'Ohmmmmmmmmmmmm' B.
[even if you sometimes have to get help from Mr. RMS]

 

[warpie]

jdbakker, I'm not sure I follow you...

 

[jdbakker]

jdbakker, I'm not sure I follow you...

You said:

Aplying the formula W = (VxV) / R doesn't really work in this case, does it?

Why would you say it doesn't work in this case? Is there any reason for Ohm's law to not apply here?

JDB.

 

[warpie]

ah ok I see 

because (294V x 294V) / 470R = 183.9Watts

this can't be right  ???

 

[jdbakker]

ah ok I see  

because (294V x 294V) / 470R = 183.9Watts

this can't be right  ???

That's the voltage between one side of the resistor and ground. The resistor doesn't know or care where 'ground' is. To calculate power you need the voltage across the resistor.

What could possibly make this harder is that the waveforms on either side of the resistor are relevant too (or more precisely: the waveform of the current through the resistor); this is where the RMS value I mentioned earlier plays a part (and why the measurements that syn posted may be a bit misleading).

JDB.
[besides, 'ground' is an illusion; 'power ground' doubly so]