I have been doing a lot of research on the methods of obtaining a B+ supply for preamp tubes. In my case, it is for a 12ax7 tube.
I have two 5H inductors at my disposal as well as many other components.
The transformer I am using is a 250-0-250 with a filament tap (I have already designed and built my filament PSU).
I would like to know a good design that would yield the lowest noise and ripple possible. The goal is to have a B+ voltage of around 300VDC.
Doing some simple calculations brought me to the conclusion that my unloaded voltage from the center tap to the 250VAC tap is about 354Vp. After the bridge rectifier this would be around 350V ripple DC at 120Hz. Since the bridge rectifier is being used in a high voltage supply, the voltage drops don't amount for much.
Thank you for the help!
I have two 5H inductors at my disposal as well as many other components.
The transformer I am using is a 250-0-250 with a filament tap (I have already designed and built my filament PSU).
I would like to know a good design that would yield the lowest noise and ripple possible. The goal is to have a B+ voltage of around 300VDC.
Doing some simple calculations brought me to the conclusion that my unloaded voltage from the center tap to the 250VAC tap is about 354Vp. After the bridge rectifier this would be around 350V ripple DC at 120Hz. Since the bridge rectifier is being used in a high voltage supply, the voltage drops don't amount for much.
Thank you for the help!
Thank you for the reply. I heard a CLC filter works better for hum reduction and the inductor is usually replaced with a power resistor in low budget applications? Take care Nigel.
For a preamp a CRC type filter circuit is usually more than adequate. Don't stint on the capacitor values if you have a semiconductor rectifier. You can get 470uF/450V electrolytics quite cheaply these days (they have good specs too as they are used in switched mode power supplies).
You will find that several RC sections in series will have much better ripple reduction than and single section of equivalent R and C. So for example, if you had four stages of 1K and 470uF, this would be very much better than one stage of 4K and 2000uF.
Most preamps consume little more than 10mA from the HT line so the 1K value is about right for your application.
Remeber to place a 100nF cap across the input to your bridge rectifier to snub switching spikes which will get into your preamp.
Cheers
Ian
You will find that several RC sections in series will have much better ripple reduction than and single section of equivalent R and C. So for example, if you had four stages of 1K and 470uF, this would be very much better than one stage of 4K and 2000uF.
Most preamps consume little more than 10mA from the HT line so the 1K value is about right for your application.
Remeber to place a 100nF cap across the input to your bridge rectifier to snub switching spikes which will get into your preamp.
Cheers
Ian
Just one 100nF across the input? I read somewhere that it was a good practice to place small caps across all four diodes in the bridge to snub the switching spikes. Is that not necessary?
I have read that too. I have built many power supplies and never yet found it necessary. Even for a 70dB gain mic pre, the single 100nF has always been enough.
Cheers
Ian
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
That bridge should be OK. However, you need to set the source resistance of your generator to some realistic value to get a meaningful simulation - with it set a zero the 100nF will not work.
Cheers
Ian
An externally hosted image should be here but it was not working when we last tested it.
Estimated source resistance and load resistor. Do those probe values at the load resistor look realistic? This is approximately 8 seconds in.
The 383uV ripple looks a little high to me given you only have a 3mA load.
I don't use LC filters so I cannot comment on the design details but I would have thought C2 at 0.1uF was rather low. Try it at 100uF and see what you get.
Cheers
Ian
An externally hosted image should be here but it was not working when we last tested it.
Much better! Although I don't know if I can find a 470uF cap that will fit in my 2U chassis, I will go as large as I can. Can you see any improvements? I am planning on having a voltage divider on the output so I can dial in a good B+ voltage.
An externally hosted image should be here but it was not working when we last tested it.
Sorry; this is hard to read. I'm trying to find the voltage ratings of caps that I need. I want PCB mountable radial caps. The probes read 352V at steady state. I can get 400V caps. Is this safe to assume? I can dish out a bit more money, but sacrifice some space. My transformer is rated at 250VAC, but this is RMS and it isn't loaded.
Thank you, have a good day.
I use 400V caps for this kind of supply all the time but I also sometimes use 450V types. Both are commonly used in SMPSUs and are readily available at reasonable prices.
Cheers
Ian
Thanks Ian. How do you feel about this layout? I haven't done the AC input and B+ output pads yet.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Fatter tracks please !
A groundplane wouldnt go amiss, fill in with as much copper as possible.
Should I worry about having channels too narrow? My grid is 50mils, so the closest points are about 30mils apart. Is that a problem?
An externally hosted image should be here but it was not working when we last tested it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.