Hello,
I'm starting on my next project: http://www2u.biglobe.ne.jp/~tossie/KT44UL-E.html
I'm looking for a solution for the next "problem"
The KT44 has a heater voltage of 4volts 2amps ( http://www.r-type.org/exhib/aad0021.htm )
I have some options:
1: using a separate transformer for 4Volts 4amps, difficult to obtain, too many transformers on the chassis.
2: using diodes or resistors to lower the voltage with 2.4 volt (too much dissipation)
3: using the 6.3 volts/8 amps from my transformer and lower the voltage with a low-drop (switching or analog) voltage regulator to obtain 4 volts 4 amps
Before I work on option 3, is there anyone who has another nice solution for 4 volt heater supplies
I'm starting on my next project: http://www2u.biglobe.ne.jp/~tossie/KT44UL-E.html
I'm looking for a solution for the next "problem"
The KT44 has a heater voltage of 4volts 2amps ( http://www.r-type.org/exhib/aad0021.htm )
I have some options:
1: using a separate transformer for 4Volts 4amps, difficult to obtain, too many transformers on the chassis.
2: using diodes or resistors to lower the voltage with 2.4 volt (too much dissipation)
3: using the 6.3 volts/8 amps from my transformer and lower the voltage with a low-drop (switching or analog) voltage regulator to obtain 4 volts 4 amps
Before I work on option 3, is there anyone who has another nice solution for 4 volt heater supplies
Another approach is a ~2.7mH inductor. CRC to drop any remaining excess voltage.
6.3VAC-rectifier-L-C1-R-C2-heaters.
C1 voltage is 6.3*.9-diode drop=~5V.
Or you can run the two heaters in series at 8V. Rect-C-R-C. Likely little if any R would be needed. Schottkys are recommended in all cases.
6.3VAC-rectifier-L-C1-R-C2-heaters.
C1 voltage is 6.3*.9-diode drop=~5V.
Or you can run the two heaters in series at 8V. Rect-C-R-C. Likely little if any R would be needed. Schottkys are recommended in all cases.
Unless you employ switching techniques, you have to "burn" the power (make heat) A resistor is cheap and will give you some amount of soft start in the heater circuit and prolong it's life.
Diodes, linear regulators etc... will all give you the same dissipation once all is said and done.
If you want to power your heaters with DC putting the R between caps in in CRC filter is attractive after a bridge rectifier.
The bridge rectifier will also also knock a couple of volts off for you (although you caps will want to charge to the peaks which are higher anyway).
Beware of one thing when using resistors... if you have multiple 4 volt heaters in parallel, a open failure of one heater will cause the remaing heaters to run at a higher voltage. Presumably, you would be smart enough to power down before any real damage was done.
😀 😉
Diodes, linear regulators etc... will all give you the same dissipation once all is said and done.
If you want to power your heaters with DC putting the R between caps in in CRC filter is attractive after a bridge rectifier.
The bridge rectifier will also also knock a couple of volts off for you (although you caps will want to charge to the peaks which are higher anyway).
Beware of one thing when using resistors... if you have multiple 4 volt heaters in parallel, a open failure of one heater will cause the remaing heaters to run at a higher voltage. Presumably, you would be smart enough to power down before any real damage was done.
😀 😉
An inductor does not have to burn off any voltage, use 3mH so its stays critical if one heater opens. Id use Schottkey rectifiers for rectification for DC quality reasons. One could make a sort of CRC that always drops the same voltage regardless of load by using diodes, not sure how the filtering would compare.
Example:
6.3VAC-Schottky->L-C-pn diode(s)->C- heaters. No noise* from the pn diode as it never experiences reverse recovery.
*excluding Johnson noise etc. Inductor critical value of 3mH for 1 heater estimated for 50hz power, bridge Schottky rectifier, and +10% mains overvoltage.
Example:
6.3VAC-Schottky->L-C-pn diode(s)->C- heaters. No noise* from the pn diode as it never experiences reverse recovery.
*excluding Johnson noise etc. Inductor critical value of 3mH for 1 heater estimated for 50hz power, bridge Schottky rectifier, and +10% mains overvoltage.
I gotta learn this "inductor input" PSU stuff... not getting it. And I build sine invertors and PFC rectifiers!
The shame of it all...
The shame of it all...
Running the heaters in series at 8V with CRC filtering would be the easiest solution, if one heater dies the circuit goes open and the other heater shuts down.
Schottkys and plenty of capacitance would be recommended to be sure you get 8V.
Schottkys and plenty of capacitance would be recommended to be sure you get 8V.
If you decide to use an inductor, use an iron / ferrous cored one, much lower EMI than an air core.
Think of inductor as like an electrical flywheel with a clutch whos losses are mostly DCR. Inductance corresponds and flywheel weight and diameter, frequency with speed, thus reactance goes way up at higher frequency. Its got momentum and wants to keep the same speed/current. When the voltage from the rectifier is below it it drags it up, when its above it drags it down. Thus with a perfect inductor the output equals the average voltage (peak *.636 for sine, which comes to RMS AC*.9), not to be confused with RMS voltage.
This analogue might not be perfect, but it works for me.
Think of inductor as like an electrical flywheel with a clutch whos losses are mostly DCR. Inductance corresponds and flywheel weight and diameter, frequency with speed, thus reactance goes way up at higher frequency. Its got momentum and wants to keep the same speed/current. When the voltage from the rectifier is below it it drags it up, when its above it drags it down. Thus with a perfect inductor the output equals the average voltage (peak *.636 for sine, which comes to RMS AC*.9), not to be confused with RMS voltage.
This analogue might not be perfect, but it works for me.
On my sims with schottkys and an 8 amp 6.3V transformer it takes about 44,000uF to get 8V@2A.
<25V capacitors arent very expensive.
LCRC 4V@4A would give cleaner DC and be sure to hit 4V.
<25V capacitors arent very expensive.
LCRC 4V@4A would give cleaner DC and be sure to hit 4V.
poobah said:
thanks for previous suggestions...
I tried this one, the diodes are running very hot at 2 amps with one tube.
with 20000uF I have DC about 6.1volts.
I tried LM338K as regulator, but the input-voltage is too low, need 7.5 volts or higher for stable 4volts at 2amps (and I do need 4 amps...)
I have to choose: low drop switching psu, or modify circuit to use 807 (6.3volts heater) instead of kt44
suggestions?
I had a sim problem but current was modest. Using Schottky diodes and a low impedance 6.3 v sec winding will provide around 7.5V and throwing alot of C,uF at it. 10 V rated caps at hi ripple rating is required, i.e at least 2x15,000uF. My avenue is go for a low drop out reggy (LDO) like LM2941 with external pass trans,find a place on the chassis where the heat won't offend. The in/out differential will be around 300-400mV, so 4V shouldn't be a problem.
It implies that you will have to lift off the existing contact of the secondary winding from ground otherwise one half of the rect won't work..
Things will get hot, but advise breadboard first.
The doubler technique isn't on as the dissipation will run riot.
A switchmode stepdown seems over complicated and you may not favour this approach, but technically is the best one.
richj
It implies that you will have to lift off the existing contact of the secondary winding from ground otherwise one half of the rect won't work..
Things will get hot, but advise breadboard first.
The doubler technique isn't on as the dissipation will run riot.
A switchmode stepdown seems over complicated and you may not favour this approach, but technically is the best one.
richj
- Status
- Not open for further replies.