Voltage sag on heater PSU - Page 2

grommeteer · 9th January 2013, 08:43 PM

To get a clearer picture can you please:
Give all component values,
Measure DC and AC voltages on all regulator pins,
Measure the current from reg-out to heater wiring when volts are low,
Test the circuit with a dummy load 22 Ohms 5W or a dial lamp 7V 0.3A. The 6922 is rated at 0.3A.
Test the heater wiring with a bench supply, a 6V battery, 5V wall wart, or a simple heater transformer. Use a fuse!
Just a thought: 1.24V is the typical Vreg of the LM338.

DF96 · 9th January 2013, 09:01 PM

Quote:

Originally Posted by Preamp

Where are those numbers from? Sounds interesting...

I can't remember, but you should be able to find them in books on power supplies. The basic idea is that the charging pulses for an electrolytic reservoir cap have a much higher RMS/average ratio than the resistive load assumed by transformer VA ratings. You have to derate by some factor around 2 or 3 (I can't remember the actual figure).

If the DC is powering a Class B/AB power amp then you can get most of that factor back again because of the duty cycle of real music. Thus a 50VA transformer can (very roughly!) just about power a 50W amp (or 25W stereo). Heaters draw continuous current so you lose this advantage.

bg3009 · 10th January 2013, 11:56 AM

Quote:

Originally Posted by grommeteer

To get a clearer picture can you please:
Just a thought: 1.24V is the typical Vreg of the LM338.

Hi, I am using 1/4 watt resistors to set the reference voltage (2 x 180R). Would there be a concern with the size of these resistors? I can't seem to find anywhere in the data sheet where it specifies the rating for these resistors.

Cheers!

DF96 · 10th January 2013, 01:19 PM

250mW power, maybe 200V voltage? Should not be a problem for a 6.3V supply.

bg3009 · 11th January 2013, 01:48 PM

Hi,
With load, the waveform to the bridge rectifier is about 9.5 volts p-p.

The voltage at the input of the power regulator is about 10.4 volts DC. The waveform pretty flat with some noise at the 1mV setting. This measurements were taken with just the 6922 plugged in. With a good 10volts supply entering the Lm338K, I am beginning to suspect the power regulator is not set up correctly. The circuit was built as per the following schematic.

The bottom section is of course the heater section that's causing grief.

Cheers!

bg3009 · 11th January 2013, 01:56 PM

Hi,

Sorry I had split my reply up. This is the waveform of the AC input at the bridge rectifier with load. I am getting a good 10.4 volts input at the power regulator.

Cheers!

grommeteer · 11th January 2013, 04:42 PM

Case of LM338K is out, but In and adj can easily be interchanged.

Robert Kesh · 11th January 2013, 05:18 PM

Quote:

Originally Posted by bg3009

Hi, I am using 1/4 watt resistors to set the reference voltage (2 x 180R). Would there be a concern with the size of these resistors? I can't seem to find anywhere in the data sheet where it specifies the rating for these resistors.

Cheers!

1.25^2/180 = a lot less than 250mW

The other resistor can sometimes need a half watt or more, but not at 6.3V

I'm wondering if your heat sink/case of regulator, is shorting with something

metalsculptor · 12th January 2013, 01:50 AM

Quote:

Originally Posted by bg3009

Because I power up through a variac, would the ramping up cause a voltage violation issue with the LM 338K?

Cheers

It would not help, the drop out current is less than 1 amp with low voltage differential. I do not know the fold back characteristics of this device but it may have gone into some sort of foldback.
Try applying full power to the circuit then switching on a heater load. The transient load from a cold heater can be determined by measuring the cold resistance, the valve datasheet may mention it as well.
There is a simple soft start circuit on the TI lm138 datasheet, this might be another option if the variac is causing trouble

Just check you have connected the 338K the right way

bg3009 · 30th January 2013, 09:01 AM

Gents,
After a good summer holiday, a fresh re-look into the circuit revealed what was wrong. Thee feedback resistors were built into a small PCB that was mounted onto the voltage setting terminal of the LM338K. A wire was to provide voltage from the output of the LM338K back to this circuit. However, due to a fault in connection, no voltage was flowing back from the output to the resistor/trimpot network. Re-did the connection and the circuit came up beautifully. Thanks every one for the input!

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Permissable voltage sag?	johnr66	Power Supplies	14	1st May 2011 06:55 PM
sudden voltage sag	forcijo	Car Audio	12	23rd July 2010 02:43 AM
B+ Voltage sag for AB1 PP	boywonder	Tubes / Valves	34	26th January 2010 01:16 AM
Distortion from heater sag?	hihopes	Tubes / Valves	33	7th October 2007 05:33 PM
Voltage sag	JonnySwitchblade	Solid State	6	8th July 2007 10:20 AM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

9th January 2013, 08:43 PM	#11
grommeteer diyAudio Member Join Date: Apr 2012 Location: Kiel, Germany	To get a clearer picture can you please: Give all component values, Measure DC and AC voltages on all regulator pins, Measure the current from reg-out to heater wiring when volts are low, Test the circuit with a dummy load 22 Ohms 5W or a dial lamp 7V 0.3A. The 6922 is rated at 0.3A. Test the heater wiring with a bench supply, a 6V battery, 5V wall wart, or a simple heater transformer. Use a fuse! Just a thought: 1.24V is the typical Vreg of the LM338.

10th January 2013, 01:19 PM	#14
DF96 diyAudio Member Join Date: May 2007	250mW power, maybe 200V voltage? Should not be a problem for a 6.3V supply.

11th January 2013, 04:42 PM	#17
grommeteer diyAudio Member Join Date: Apr 2012 Location: Kiel, Germany	Case of LM338K is out, but In and adj can easily be interchanged.

30th January 2013, 09:01 AM	#20
bg3009 diyAudio Member Join Date: Jan 2013	Gents, After a good summer holiday, a fresh re-look into the circuit revealed what was wrong. Thee feedback resistors were built into a small PCB that was mounted onto the voltage setting terminal of the LM338K. A wire was to provide voltage from the output of the LM338K back to this circuit. However, due to a fault in connection, no voltage was flowing back from the output to the resistor/trimpot network. Re-did the connection and the circuit came up beautifully. Thanks every one for the input!