• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Rectifier diodes for tube circuits

Jump to Latest
andyjevans

andyjevans

Member
Joined 2003
artsandmedia.com
#1
I'd like some up-to-date recommendations for diodes to use for rectification in tube amps. Are there better options than UF4007, if so which?

I could do with some help on voltage ratings as well. A typical PSU for my amps would have around 300v with a load of say 160mA, but maybe 425v off load. So would a 600v rating be safe? I have a few C3D02060F which are 600v Schottkys. In fact would Schottky or other specialist diodes make any audible difference?

I've always used tube rectification but I do use hybrid/Graetz bridges. I haven't optimised the diodes for a while. I'd also like to listen again to solid state rectification to see how it sounds.
 
#3
andyjevans said:
I'd like some up-to-date recommendations for diodes to use for rectification in tube amps. Are there better options than UF4007, if so which?
Click to expand...

The UF400x series is quite good, fast and soft recovery. The only reason to go to a higher-current one would be if you draw a lot of current and/or use very large reservoir capacitors. Big tube power amps might need that. Look for diodes that have the required voltage and current rating and are fast and soft switching. That minimizes rectifier noise.
And remember that the voltage rating must be minimum twice the final DC voltage!

Jan
 
#7
You may be able to measure it but you won't hear it.

A 1000v bridge costs about £1 and is easier to mount. The higher current ones with a fixing bolt hole don't cost much more and you can still bypass the diodes with low value capacitors should you wish to. (and you will also not hear any difference)

I do wonder whether some forumites actually listen to their builds: many seem more obsessed with measurements and specifications.


If you want to use diodes, stick with the good old 1N4007 OR 1N5408 if you want a 3amp 1000v one.
 
Last edited:
#8
I can't answer the question of whether I can hear a difference until I actually try out the alternatives. I can hear differences in tube rectifiers between the same rectifier in solid and mesh plate versions, so I'm keeping an open mind. In any case, there are plenty of other threads about the more general issue of whether you can hear differences in rectifiers and I'd like this particular thread to just be about practical suggestions.
 
#11
As I said, I use diodes in my hybrid bridges when the transformers don't have a centre tap. But I'm curious about an all-solid state rectifier hence the thread. I do mostly use tubes but it's good to have options.

Thanks for the bridge link. Could be worth trying - easy to use for sure.
 
Last edited:
#13
I have always used the CSD01060A, would be interested to know the pros a cons of those vs the C3D02060F. Both are 600v 2A rated parts, one has a lower reverse current, the other a lower capacitance. What effect would each have? ...maybe a question for the quasimodo thread.
 
#14
I could do with some help on voltage ratings as well. A typical PSU for my amps would have around 300v with a load of say 160mA, but maybe 425v off load. So would a 600v rating be safe?
Click to expand...

In a typical solid state rectified tube amp the power supply will be at it's full unloaded voltage at switch on until the tubes warm up. During this time the B+ voltage will be at its max, as will the AC voltage coming from the transformer since it will see no load after the filter caps charge up.

If you do have a no load voltage of 425 volts, the AC voltage needed to charge a cap to this value must have been at least 300 volts to generate a 425 volt DC charge. (300 * 1.414 = 424.2V). This means that at the reverse biased peak the diode will see 425 volts on the capacitor side, and -424.2 on its transformer side. The diode can see 850 PIV continuously in this situation. It can see quite a bit more in transient, or noisy power line situations. This agrees with the textbook statements that say the diode should be rated for at least 2.82 times the transformer AC voltage.

Many diodes, especially the common silicon variety like the 1N4007 have a very wide tolerance in the true PIV breakdown spec. One diode may blow up at 1050 volts while the next may live at 1600 volts. Most will eat 1200 volts forever.

Some diodes like the UF series and other "fast" rectifiers do not exhibit this extra margin. I saw some failures in early SSE amps with IXYS 1200 volt Fast Recovery Epitaxial Diodes (FRED) on a 375 volt transformer. Newer 1600 volt diodes made the issue go away.

1N4007 is the optimum diode. It is not high frequency so Shottky/fast recovery diodes are not required and make no useful difference.
Click to expand...

Look for diodes that have the required voltage and current rating and are fast and soft switching. That minimizes rectifier noise
Click to expand...

As to the 1N4007 VS UF4007 controversy, some say it makes no difference, and some say it does.....

I have built several small guitar amps that use series heater strings and derive their power from an isolation transformer with a "120 volt AC" secondary, often the Triad N68-X. These transformers are usually NOT exactly 1:1, but have a higher number of turns on the secondary to make up for transformer losses. I get 170 volts from a bridge rectifier, and 340 volts from a voltage doubler under load on those with my 122 to 125 volt line

There is a small measurable (scope) and sometimes audible difference between the two types of diodes in these amps.......unfortunately, the UF4007 is not always the better choice. Transformer wiring, lead dress, PCB layout, and total DCR in the path influence the measurements. Granted some of these amps have as many as 8 diodes in the power supply. They are far cheaper than "vacuum tube" power transformers.

The 8 diode amp makes 15 watts of output power, has a bridge for 170 volts for the screen grids and tube heaters, and a full wave doubler for main 340 volt B+. It likes UF4007's.

The 5 diode 4 watt amp runs the tube heaters on unfiltered DC, and uses filtered DC for 170 volts of B+. It likes 1N4007's.

For bigger amps that need more power I use DSA1-18D diodes. They are good for 2.3 amps at 1800 volts.
 

Attachments

  • HBAC_5T.pdf
    HBAC_5T.pdf
    29.1 KB · Views: 131
  • GTA_4T_11-10.pdf
    GTA_4T_11-10.pdf
    22.6 KB · Views: 99
  • DSA1-18D.pdf
    DSA1-18D.pdf
    36.7 KB · Views: 109
  • P1000768_x.jpg
    P1000768_x.jpg
    634.9 KB · Views: 406
  • NewGuitar_x.jpg
    NewGuitar_x.jpg
    862.3 KB · Views: 334
You must log in or register to reply here.
Top Bottom