I ask this as a question..is it possible to get 6.3 vdc regulated from 6.3v AC after factoring in 1.4v drop in the bridge rectifier, perhaps 1.5v in the low drop out regulator and 10% droop on the 100hz raw input ? Wouldn't you need to start with approx 8v ac ?
Last edited:
... there is nothing to gain from it
There can be something else to gain from it: a soft start is very easy to implement with a linear regulator. As the filaments fail mostly at power on because of the inrush current, this increases the life of the tube.
Possible, but painfully so: the power factor of rectifier + filter cap is in the region of 0.5, compared to 1 for direct connection, and this will be more demanding on the winding.I ask this as a question..is it possible to get 6.3 vdc regulated from 6.3v AC after factoring in 1.4v drop in the bridge rectifier, perhaps 1.5v in the low drop out regulator and 10% droop on the 100hz raw input ? Wouldn't you need to start with approx 8v ac ?
If you use a standard diode bridge, you will most likely be too short of some hundreds of mV.
You can do it by using germanium or schottky diodes, or even better synchronous rectification.
I have described one or two such solutions in the PSU section.
Some form of soft start will also reduce the risk of killing the regulator, which seems to be quite common. DC heaters can be much harder to get right than AC heaters. Better for people to learn how to do AC wiring properly. Phono preamp inputs are the only valves which may need a DC heater.Vincent77 said:There can be something else to gain from it: a soft start is very easy to implement with a linear regulator. As the filaments fail mostly at power on because of the inrush current, this increases the life of the tube.
Hi!
I never had a heater failure on an indirectly heated tube due to inrush current.
Different story for directly heated tubes, especially those with thoriated tungsten filaments there a soft watrt will help to avoid filament damage
Thomas
There can be something else to gain from it: a soft start is very easy to implement with a linear regulator. As the filaments fail mostly at power on because of the inrush current, this increases the life of the tube.
I never had a heater failure on an indirectly heated tube due to inrush current.
Different story for directly heated tubes, especially those with thoriated tungsten filaments there a soft watrt will help to avoid filament damage
Thomas
Possible, but painfully so: the power factor of rectifier + filter cap is in the region of 0.5, compared to 1 for direct connection, and this will be more demanding on the winding.
Absolutely right!
You will draw more than 1.6x average DC current from the transformer, in short ripple bursts that can reach 5 times the DC average!
So if you need DC heating, make sure that the transformer winding is rated at 2x the DC current you need with enough voltage headroom, and the capacitor withstands the ripple current, or else you will get *a lot* of unwanted heat.
Can you give the title of that thread ? I did look but you have a a lot of posts !Elvee: I have described one or two such solutions in the PSU section.
I have to admit to not making a valve amp since 1962 (mullard 5/10) so I found the concept interesting. I don't remember the idea of dc heaters being discussed much then. Where could we discuss it anyway it in those pre internet days
If only the first amplifier stage heaters were heated by dc, the problems of rectifying and stabilising would be a lot less I guess. (when starting with only 6.3v ac) And I always thought that was where most hum problems began.
Supplying all the filaments and using a very large reservoir cap perhaps could be source of 100hz buzz from the small angle charging spikes ?.Just thinking aloud..
Last edited:
Hi!
I never had a heater failure on an indirectly heated tube due to inrush current.
Different story for directly heated tubes, especially those with thoriated tungsten filaments there a soft watrt will help to avoid filament damage
Thomas
Hi Thomas,
Indeed, filament failure is not very frequent, but there are some small tubes that make a flash at power on. I hate to see them flashing, especially expensive NOS.
So, if someone uses a linear regulator for the heater, it only costs a transistor+R+C to make the thing turn on slowly, protecting the filament... It's certainly better than a brutal turn on.
Here is one:Can you give the title of that thread ? I did look but you have a a lot of posts !
http://www.diyaudio.com/forums/power-supplies/177458-6-3-filament-5vac-winding-2.html#post2435043
Another one:
http://www.diyaudio.com/forums/power-supplies/200015-6-3vdc-form-6-3v-ac-psu-2.html
Last edited:
.............................Hi Thomas,
Indeed, filament failure is not very frequent, but there are some small tubes that make a flash at power on. I hate to see them flashing, especially expensive NOS.
So, if someone uses a linear regulator for the heater, it only costs a transistor+R+C to make the thing turn on slowly, protecting the filament... It's certainly better than a brutal turn on.
I often choose a linear regulator with a foldback current limit that is a little higher than the hot heater demand current..The tungsten cold filament resistance....will force it into foldback current limiting...and will rise slowly. The most simplest foldback i.e transistor across current lim resistor works by altering the low resistance series. See pic.
This simple circuit has an in/out drop out equivalent to many commercial 3/5 pin packages. Ignore the common mode inductor at Vin. I use it right through my power amps and has an excellent CM rejection ratio.
Worth experimenting.
richy
Attachments
I ask this as a question..is it possible to get 6.3 vdc regulated from 6.3v AC after factoring in 1.4v drop in the bridge rectifier, perhaps 1.5v in the low drop out regulator and 10% droop on the 100hz raw input ? Wouldn't you need to start with approx 8v ac ?
It's doable with a 1084 regulator as stated above and these 2A 40V Schottky diode bridges from Digikey with only a 0.5V forward voltage: part# 641-1213-1-ND. Surface mount, but I was able to solder them into a point-to-point circuit with a little ingenuity.
http://www.digikey.com/product-detail/en/CDBHD240-G/641-1213-2-ND/1678583
Last edited:
Hi!
Yes I also have some tubes which do that. But it's normal and doesn't harm them. realy no need to worry about
Thomas
Indeed, filament failure is not very frequent, but there are some small tubes that make a flash at power on. I hate to see them flashing, especially expensive NOS. .
Yes I also have some tubes which do that. But it's normal and doesn't harm them. realy no need to worry about
Thomas
Another great way to drop a few volts with very little effort is to put a 1n5401 in line. Each one drops about 1.2 volts. An oversized inrush current resistor like a CL-90 will soft start and provide some drop as well.
http://www.mouser.com/images/gesensing/images/CL Series.JPG
http://www.mouser.com/images/gesensing/images/CL Series.JPG
How about a different approach ?
DC Boost Buck Converter Step Down Step Up Voltage Regulator Power Supply Circuit | eBay
If not that one , search ebay for dc-dc converters. There are plenty to choose from.
DC Boost Buck Converter Step Down Step Up Voltage Regulator Power Supply Circuit | eBay
If not that one , search ebay for dc-dc converters. There are plenty to choose from.
I have a Mullard ECC83 that has been doing this every time I switch the amp on for the past 20 years. I gave up worrying about it 19 years ago...Indeed, filament failure is not very frequent, but there are some small tubes that make a flash at power on. I hate to see them flashing, especially expensive NOS.
Yes I also have some tubes which do that. But it's normal and doesn't harm them. realy no need to worry about
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Tubes / Valves
- Simple DC heating question