I have finally decided to use discrete diodes for power amp rectifiers. I found two suggestions: HFA25TB60 at CAD 2.5 a pc. and IRF HFA08TA60C at CAD 5 a pc. Is there any big differences between the two, since having to use 16 pcs. per amp the price makes the difference. Any other choices?
Not sure what kind of diode these are, you should go for fast recovery HEXFRED types if you have the $$$
Personally, IN4007 does it for me 😀
Personally, IN4007 does it for me 😀
The HFA08TA60C has a softer and faster recovery. It is also a common-cathode construction, so you could conceivably use three to make a full-wave rectifier, instead of four. You said you need 16, so I must assume you need to parallel them to get the current you want. The HFA08TA60C is good for 32A so I guess you must need a lot of current 🙂
I wouldn't try to save cost on the rectifier. It will affect the performance of everything else in the amp. I was surprised by your price of CAD5/each, so I checked Digi-Key and their price would be CAD55/16 (3.44/per). Anyway I've used the HFA08TA60C and they are a happy little diode.
Is this for your Aleph-X? How much current are you going for. Yes, I'm nosy.
I wouldn't try to save cost on the rectifier. It will affect the performance of everything else in the amp. I was surprised by your price of CAD5/each, so I checked Digi-Key and their price would be CAD55/16 (3.44/per). Anyway I've used the HFA08TA60C and they are a happy little diode.
Is this for your Aleph-X? How much current are you going for. Yes, I'm nosy.
I'm not sure about HAA08TA60C, can't find it at IRF.
Maybe you meant HFA08TA60C? That's a doule diode (two in one case, cathodes already connected), but rated aat only 4A. Perhaps nit sufficient for a power amplifier.
HFA25TB60A looks great to me. Ultrafast rectification, slow recovery, 25A continuous.
Maybe you meant HFA08TA60C? That's a doule diode (two in one case, cathodes already connected), but rated aat only 4A. Perhaps nit sufficient for a power amplifier.
HFA25TB60A looks great to me. Ultrafast rectification, slow recovery, 25A continuous.
The continuous forward current rating isn't very interesting in a highly filtered power supply. A rectifier feeding a huge capacitor bank will have a tiny conduction angle. Because of this I usually pay more attention to the maximum forward pulse current.
Ok, you may be right concerning the current rating. I like to have some reserve if in doubt..😀
Could you please explain how to do a bridge rectifier with three cc-diodes? Wasting two "halves"?
Could you please explain how to do a bridge rectifier with three cc-diodes? Wasting two "halves"?
I must have done some mistake when checking prices. HFAo8TA60C is indeed cheaper that the other one (CAD2.4 if over 10). I'm using them for Aleph 2 with separate bridge per rail. So you waste 2 diodes? Are they better than HFA25TB60?
You don't necessarily "waste" the two anode pins. You need three anodes and three cathodes for any full wave rectifier. It's a shame that they don't make the same diode in common-anode constructions. Actually it's a shame they don't make a HEXFRED bridge. Anyway if it bothers you to have two unconnected pins, connect them to the other anode pin on the same device 🙂
These diodes recover faster and have less stored charge than HFA25TB60, so I would say they are better. But, I've never used the other one.
These diodes recover faster and have less stored charge than HFA25TB60, so I would say they are better. But, I've never used the other one.
So maybe it's better to use 4 instead of 3 and connect them all in parallel. Although I tried once to run two bridges in parallel, but I got impression it wasn't sounding as good as a single bridge.
I've just notice that those diodes are 8A. Personally, I wouldn't like to use anything below 25A. I had my share of blown bridges at turn on already.😉
Maximum repetative forward current: 60A. Should fulfill all needs, if jwb is right.jwb said:The continuous forward current rating isn't very interesting in a highly filtered power supply. A rectifier feeding a huge capacitor bank will have a tiny conduction angle. Because of this I usually pay more attention to the maximum forward pulse current.
I've got mixed up with numbers in the beginning.
The two types of diodes I had in mind were HFA25TB60 at CAD4.80 and 30ETH06 at 2.80. Both of them are single, 25 A. The difference in price is almost double, how performance compare? Any advice again?
The two types of diodes I had in mind were HFA25TB60 at CAD4.80 and 30ETH06 at 2.80. Both of them are single, 25 A. The difference in price is almost double, how performance compare? Any advice again?

HBarske said:
Maximum repetative forward current: 60A. Should fulfill all needs, if jwb is right.
As I said, 25A bridges did fail, it's the inrush current at turn on that I'm afraid.
IXYS high-speed soft-recovery.
I seem to remember a few discussions within some other threads that contained information on rectifiers. There were two schools of thought, one being standard bridge rectifiers or diodes to avoid the noise that is imparted by high-speed types, and soft-recovery high-speed types that were quicker in recovery yet were able to control ringing and noise by virtue of their design. A search came up with an article that Harry posted in this thread:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=2197&highlight=ixys.
I haven't had the chance to do any comparison listening, but starting with less "noisy" devices will hopefully provide a better end result. I still wonder if this is something that you can really hear or is it a matter of psychoacoustics.
One other thing, they seem rather expensive, and you can get them at Digikey.
Later,
I seem to remember a few discussions within some other threads that contained information on rectifiers. There were two schools of thought, one being standard bridge rectifiers or diodes to avoid the noise that is imparted by high-speed types, and soft-recovery high-speed types that were quicker in recovery yet were able to control ringing and noise by virtue of their design. A search came up with an article that Harry posted in this thread:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=2197&highlight=ixys.
I haven't had the chance to do any comparison listening, but starting with less "noisy" devices will hopefully provide a better end result. I still wonder if this is something that you can really hear or is it a matter of psychoacoustics.
One other thing, they seem rather expensive, and you can get them at Digikey.
Later,
Weird. The IXYS have worse specifications than the IRF HEXFREDs. Some of the specs -- like stored charge -- are worse by an order of magnitude. However, there is no way to know without trying them yourself.
something that you can really hear
Yes you can measure it also. Soft recovery diodes were not developed for audio, there are plenty of applications where the noise of conventional diodes is an issue.
http://www.gensemi.com/appnotespdf/quik108.pdf
Yes you can measure it also. Soft recovery diodes were not developed for audio, there are plenty of applications where the noise of conventional diodes is an issue.
http://www.gensemi.com/appnotespdf/quik108.pdf
Rectifier Quartet
Peter-
Discrete rectifiers can indeed make a difference in the sound of your project...especially if you are demanding much of your power supply, as a properly biased Class A amplifier will indeed do.
I am using four SCRs in my power supply (they actually performed brilliantly as fast-acting fuses when a capacitor met its demise!). One can "soft start" the power supply by controlling the conduction time of the SCRs.
To test the bare supply, I tried omitting the regulation circuit, and temporarily installed diodes. Playing with various types, we could hear subtle differences...the descriptions may vary, but I believe the perceived difference wasn't "mass hysteria".
Put some tasty ones in there, and have fun listening!
Oh...without the SCRs, the start-up transient made everyone outside the room think I was experimenting with an Electric Chair!
😱
Peter-
Discrete rectifiers can indeed make a difference in the sound of your project...especially if you are demanding much of your power supply, as a properly biased Class A amplifier will indeed do.
I am using four SCRs in my power supply (they actually performed brilliantly as fast-acting fuses when a capacitor met its demise!). One can "soft start" the power supply by controlling the conduction time of the SCRs.
To test the bare supply, I tried omitting the regulation circuit, and temporarily installed diodes. Playing with various types, we could hear subtle differences...the descriptions may vary, but I believe the perceived difference wasn't "mass hysteria".
Put some tasty ones in there, and have fun listening!
Oh...without the SCRs, the start-up transient made everyone outside the room think I was experimenting with an Electric Chair!
😱
43CTQ100 Schottky Common Cathode Diode from IR. IF(av) = 40A (20A per leg). Vrrm = 100V.
http://icceman1.tripod.com/Schottky_Bridge.htm
http://icceman1.tripod.com/Schottky_Bridge.htm
- Status
- Not open for further replies.
- Home
- Amplifiers
- Solid State
- Using discrete diodes instead of rectifying bridges.