I have this thought because my friend has an electronic factory in China and when I talked to him, he said he can make such at a max of 5 amp, 12V with a minimum of 200 pcs at $25 each. The module has current limiter and voltage control that I think this could be good for DHT tubes. I'm not sure other than the above, is there any other criterion to determine a good DHT filament heater?
I have used this sort of power supply when testing 813SE amplifier without any extra filtering (but series NTC of course).
I found no measurable problem in THD or S/N.
These power supplies have voltage adjustment, but I don't know about current limit.
12V 10A LED Strip Power Supply 120W LED 12VDC Switching Power Supply Led Adapter | eBay
I found no measurable problem in THD or S/N.
These power supplies have voltage adjustment, but I don't know about current limit.
12V 10A LED Strip Power Supply 120W LED 12VDC Switching Power Supply Led Adapter | eBay
I believe that there is no shortage of single output low noise switched mode power supplies, no need to design a custom one. An example of a modern commonly available compact power supply of this kind is the Mean Well LRS-75-12: 75W, the output can be set between 10.2 and 13.8V at 6A, and the ripple and noise at 20 MHz bandwith is 120 mVpp. Switching frequency is 65 KHz. This UL listed industrial grade SMPS is better than most LED power supplies, and it costs way less than 25 USD. For a little more money, better specifications are available.
Its like … people are forgetting good RF suppression tech? Seems so. The unsubtle answer is "separate cans": isolating one's crazy-noisy (potentially) switching power supplies in earth-grounded faraday cages is critical. Having tight little multiturn ferrite-core chokes right at the exit point of the cage is also critical. Do all the filtering IN-can, and then suppress the emission of anything RF'ish.
You can do this for Lo volt H+ supplies (heater); you can do this for B+ or S+ (screen) supplies. Copper sheeting isn't all that expensive, and does wonders. So do hand-made small ferrite RF suppressor chokes.
OR - if you prefer - you can do the Faraday cage thing in reverse. Put all the hash outside the amplifier cage. Enclose the amp in its own Faraday cage to keep out in-the-air RF. Put the little RF suppressor beads right at the power-feed-in points. "Think like an antenna" as my uncle used to say.
Anyway, my 2¢ worth. Not much these days.
GoatGuy
You can do this for Lo volt H+ supplies (heater); you can do this for B+ or S+ (screen) supplies. Copper sheeting isn't all that expensive, and does wonders. So do hand-made small ferrite RF suppressor chokes.
OR - if you prefer - you can do the Faraday cage thing in reverse. Put all the hash outside the amplifier cage. Enclose the amp in its own Faraday cage to keep out in-the-air RF. Put the little RF suppressor beads right at the power-feed-in points. "Think like an antenna" as my uncle used to say.
Anyway, my 2¢ worth. Not much these days.
GoatGuy
The faraday cage concept is actually almost a necessity for the compliance to EN-55032 anyway (I believe). This is the latest EMC emission standard for multimedia equipments. If you look at a modern EN-55032 compliant power supply such as the above mentioned LRS-75-12, it clearly shows the construction details you mentioned. Good earth ground is mandatory to avoid nasty EMC issues.
Finally "common sense" becomes "common law" for the "common good" of the "common man".
I'm an unrepentant ex-Ham, no longer on the air, but once 40 years back "into it". You learned early-and-quickly to use Faraday cages for all sorts of things Hammish. Its "the way" of RF.
Thanks for mentioning EN–55032.
Its surprisingly easy to do the right thing with copper wire mesh caging material. Just remember to solder the "little holes' mesh wires" where you intend to screw it onto the chassis.
Do so before drilling the screw-holes. Makes it SO much easier to get clean holes drilled in the mesh. Also, it serves as a nice buffer when tying the mesh cages to the chassis for the screws (which tend to crumble the naked copper wire over time).
GoatGuy
PS: it also makes good sense to buy a small box or bag (if they still have them) of small, thin brass washers. Disks with holes. Really thin. You can solder these to the mesh too, to make both screw-holes and wire-port feed-thrus. Don't forget your grommets! Insulation is no barrier to gradual insulation wear. Grommets are. GoatGuy
Thank you for all the input. The unit recommended by PCAN ?? ?? ??? DC12V?? ?? ???-??? is a high voltage switching supply from a 12V input. The filament supply is a 317 which is adjustable for various voltage output. Since I need a switching supply for DHT filament supply, this module is not for me.
I have this need because I failed to make my DHT preamp dated back 20 years ago because of hum. With today's technology, I THINK (only) we can fight against DHT hum problem with switching power supply.
A bit elaboration on what we are getting for $25. This will be an individual RF sealed box (just like 120W 12V switching supply from eBay) with TWO channels of 60W (total 120W) supply of switching power that can be adjusted between 1.25V to 12V therefore we can use it for different DHT tubes. There is an adjustable current limiter ranging between 100mA to 5A to avoid over supply of current at start up. The unit can plug into 100V-240V and is self adjusting to suit worldwide voltage.
According to my friend, considerable cost is making the RF sealed box because of mechanical machine tools. Break down of costs: -
RF sealed box $8 min 200.
1.25V-20V 60W with current limiter $5 per channel (two channels $10).
Universal voltage supply $7.
No minimum order quantity for the switching unit.
This really draw up my interest if we don't need the RF sealed box and universal voltage supply since there should already be 6.3V or 12V from the existing transformer. I will ask him to give me (free of charge of course) two 1.25V-12V modules and restart my DHT preamp that I stopped 20 years ago.
I have this need because I failed to make my DHT preamp dated back 20 years ago because of hum. With today's technology, I THINK (only) we can fight against DHT hum problem with switching power supply.
A bit elaboration on what we are getting for $25. This will be an individual RF sealed box (just like 120W 12V switching supply from eBay) with TWO channels of 60W (total 120W) supply of switching power that can be adjusted between 1.25V to 12V therefore we can use it for different DHT tubes. There is an adjustable current limiter ranging between 100mA to 5A to avoid over supply of current at start up. The unit can plug into 100V-240V and is self adjusting to suit worldwide voltage.
According to my friend, considerable cost is making the RF sealed box because of mechanical machine tools. Break down of costs: -
RF sealed box $8 min 200.
1.25V-20V 60W with current limiter $5 per channel (two channels $10).
Universal voltage supply $7.
No minimum order quantity for the switching unit.
This really draw up my interest if we don't need the RF sealed box and universal voltage supply since there should already be 6.3V or 12V from the existing transformer. I will ask him to give me (free of charge of course) two 1.25V-12V modules and restart my DHT preamp that I stopped 20 years ago.
Last edited:
I have tried to use a 317 module with adjustable output voltage for DHT filament supply. The sound is very dull, no low and no high at all. Bass is very muddy.
The next trial is to use the buck dc to dc converter. Since it does not have a rectification module, I am using the same 317 module as rectification to supply DC to the buck converter. The result is a little bit improvement on the high but no improvement on the bass.
I believe there is no result without some extra work therefore I build a rectification module with a simple bridge and cap. Now the highs and lows are there and it is the best setup among the three that I tested.
It is still not comparable to dedicated DHT filament heater and I would say about 80% performance only. Yet, if you compare the price and performance ratio, it worth! As a remark, I discovered the buck converter is using a 317 as current limiter and I think the sound quality could be improved by removing the current limiter. Um..... I don't think I will try to do this by myself. I rather ask my friend to send me two more buck converter without current limiter - of course, same as before, FOC.
An externally hosted image should be here but it was not working when we last tested it.
The next trial is to use the buck dc to dc converter. Since it does not have a rectification module, I am using the same 317 module as rectification to supply DC to the buck converter. The result is a little bit improvement on the high but no improvement on the bass.
An externally hosted image should be here but it was not working when we last tested it.
I believe there is no result without some extra work therefore I build a rectification module with a simple bridge and cap. Now the highs and lows are there and it is the best setup among the three that I tested.
An externally hosted image should be here but it was not working when we last tested it.
It is still not comparable to dedicated DHT filament heater and I would say about 80% performance only. Yet, if you compare the price and performance ratio, it worth! As a remark, I discovered the buck converter is using a 317 as current limiter and I think the sound quality could be improved by removing the current limiter. Um..... I don't think I will try to do this by myself. I rather ask my friend to send me two more buck converter without current limiter - of course, same as before, FOC.
Last edited:
The new buck dc-to-dc converter arrived this afternoon. This is great to live here since there are so many toys around you that you can order and receive them within overnight at extremely low cost. The modules arrived with a courier fee of US$0.70 to my Shenzhen location; yes 70 US cents within same city (but still 2 hours drive apart).
Actually, I received two simple modules. One of them is very primitive while the other one has a display showing the input and output voltage. The costs are only peanuts - US$2.5 for the simple module and US$3.5 for the module with voltage display.
The simple module does not have connectors therefore I have to solder the power and output lines on it. Besides, I also changed the caps from 220uF solid caps to Panasonic FC 1800uF 35V (I don't have larger capacity with lower voltage on hand). I further changed the trimmer to fix resistor to give a consistent output voltage. They look like this after modification: -
The US$3.5 fancy model look like this without any modification: -
I did some measurement on this fancy module, the display does not show accurate voltage at all with a deviation of about 0.08V (at 1.4V output)
After installing the modified simple module to the DHT preamp; without burning in, I would say there is improvement to the module that has current limiter. The bass is getting clear and more while the high produces crispy "dings and dongs". If I have to give a comparison to the dedicated DHT filament heater supply, I would say the bass is now 90% and the high is 85% of it.
I have also ordered some 10000uF from mouser this evening and look forward if there could be further improvement by changing the cap after the bridge rectifier to this value.
Actually, I received two simple modules. One of them is very primitive while the other one has a display showing the input and output voltage. The costs are only peanuts - US$2.5 for the simple module and US$3.5 for the module with voltage display.
The simple module does not have connectors therefore I have to solder the power and output lines on it. Besides, I also changed the caps from 220uF solid caps to Panasonic FC 1800uF 35V (I don't have larger capacity with lower voltage on hand). I further changed the trimmer to fix resistor to give a consistent output voltage. They look like this after modification: -
An externally hosted image should be here but it was not working when we last tested it.
The US$3.5 fancy model look like this without any modification: -
An externally hosted image should be here but it was not working when we last tested it.
I did some measurement on this fancy module, the display does not show accurate voltage at all with a deviation of about 0.08V (at 1.4V output)
After installing the modified simple module to the DHT preamp; without burning in, I would say there is improvement to the module that has current limiter. The bass is getting clear and more while the high produces crispy "dings and dongs". If I have to give a comparison to the dedicated DHT filament heater supply, I would say the bass is now 90% and the high is 85% of it.
An externally hosted image should be here but it was not working when we last tested it.
I have also ordered some 10000uF from mouser this evening and look forward if there could be further improvement by changing the cap after the bridge rectifier to this value.
Last edited:
This is what I bought...
DC-DC Step Down Adjustable Constant Voltage Current Power Supply Module | eBay
For the price I paid, it's a bargain. Need some additional work to get it optimise.
DC-DC Step Down Adjustable Constant Voltage Current Power Supply Module | eBay
For the price I paid, it's a bargain. Need some additional work to get it optimise.
Last edited:
Do you have a hum canceling resistors (trimmer) across the filament, similar that is used when having AC-filament voltage ?
Attachments
Last edited:
Just increasing the output capacitance on the switcher might not be the best approach.
Look at the datasheet for most buck converters E.G. 2575 or LT1076 and you will see that the manufacturer has given recommendations with regard to optional output filters.
Using them as a pre reg for a decent CC/CV linear supply is doable.
Look at the datasheet for most buck converters E.G. 2575 or LT1076 and you will see that the manufacturer has given recommendations with regard to optional output filters.
Using them as a pre reg for a decent CC/CV linear supply is doable.
What I did was to add some low ESR reservoir capacitor along with some decoupling caps. More of a C-L-C configuration at the output. This tame quite a significant HF noises that I was previously able to hear at the output of the DHT driver tube. I'm happy with the latest configuration since the noise at the speaker (100dB) is barely audible.
I guess the quality of those reservoir caps on the board are in question. I'm still trying various approach powering the filament bias. Might rip-up all those original caps on the buck converter eventually and replace it with some better ones. Still experimenting.
Another switching buck converter board that I'm bought poses more challenges to get the noise down.
I guess the quality of those reservoir caps on the board are in question. I'm still trying various approach powering the filament bias. Might rip-up all those original caps on the buck converter eventually and replace it with some better ones. Still experimenting.
Another switching buck converter board that I'm bought poses more challenges to get the noise down.
Last edited:
Thank you for the suggestion
I did use hum cancelling when supplying a.c. to the filament. For HF switching PSU or buck converter, it is less effective.
With filament bias, the ground return has to be on one end of the filament connection.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.