I'm hoping someone can tell me what difference there is between 2 diodes:
* one is called a "rectifier diode, Standard Recovery" (1N5404G).
* the other is called a "rectifier diode, Switching" (G1754-E3).
My situation is that I bought some 3a/400v rectifier diodes from RS Components. Their product no. is 774-3331 (1N5404G).
What I want is the same - but 6a instead of 3a.
RS Components have product no. 708-8016 (G1754-E3). This is 6a/400v ... but its description is "Rectifier Diode, Switching 6A max".
All I want is a higher-amp version of 774-3331 ("rectifier diode, Standard Recovery). Can a G1754-E3 be regarded as the same (just higher amp-rated) or does the word "Switching" in its description mean it performs a different function?
Thanks,
Andy
* one is called a "rectifier diode, Standard Recovery" (1N5404G).
* the other is called a "rectifier diode, Switching" (G1754-E3).
My situation is that I bought some 3a/400v rectifier diodes from RS Components. Their product no. is 774-3331 (1N5404G).
What I want is the same - but 6a instead of 3a.
RS Components have product no. 708-8016 (G1754-E3). This is 6a/400v ... but its description is "Rectifier Diode, Switching 6A max".
All I want is a higher-amp version of 774-3331 ("rectifier diode, Standard Recovery). Can a G1754-E3 be regarded as the same (just higher amp-rated) or does the word "Switching" in its description mean it performs a different function?
Thanks,
Andy
It should work for your application, except for an unlikely case where the circuit is sensitive to the faster on-off transitions.
I've not looked at the data sheets. The "switching" description probably means it is faster (in regards to turn-on & turn-off time) than a "standard recovery" rectifier.
You can view the "standard" as suitable for 50/60 Hz rectification of a linear psu, while "switching" is suitable for rectification in a switch-mode psu where frequency can measure in 100's of kHz.
Like the diode PIV rating, you can go higher (or faster) than necessary, but never lower (slower) than necessary. So make sure the diode you select is adequate in regards to PIV, repetitive surge current, etc.
I've not looked at the data sheets. The "switching" description probably means it is faster (in regards to turn-on & turn-off time) than a "standard recovery" rectifier.
You can view the "standard" as suitable for 50/60 Hz rectification of a linear psu, while "switching" is suitable for rectification in a switch-mode psu where frequency can measure in 100's of kHz.
Like the diode PIV rating, you can go higher (or faster) than necessary, but never lower (slower) than necessary. So make sure the diode you select is adequate in regards to PIV, repetitive surge current, etc.
Standard recovery don't turn off immediately
and for a brief time will conduct in reverse.
Fast recovery improve on this at the expense
of a slightly higher forward voltage.
Super Barrier Rectifiers are the more modern
choice over either of the above. They behave
in circuit much like a Schottky. But inside built
like a MOSFET with near zero (ion implanted)
gate threshold. They reverse block better than
true Schottky, and to much higher voltages.
and for a brief time will conduct in reverse.
Fast recovery improve on this at the expense
of a slightly higher forward voltage.
Super Barrier Rectifiers are the more modern
choice over either of the above. They behave
in circuit much like a Schottky. But inside built
like a MOSFET with near zero (ion implanted)
gate threshold. They reverse block better than
true Schottky, and to much higher voltages.
Last edited:
The "switching type is usually of Schottky type with the lower forward voltage drop with a clean switch-off. As noted, it is mainly used in Switching power supplies. You can use the standard types just fine at the lower frequencies, but a good rule of thumb is to jump each junction with a 100nF cap to get that clean turn-off.
___________________________________________________Rick............
___________________________________________________Rick............
Depending on your intended use, particularly if it's for an audio component, perhaps you might look for a specific diode called a 'soft recovery diode'
The 1N5404 are a pretty standard diode, been around for a long time, works okay but not the quietest device in good quality power supplies even when run under 1A
The 708-8016 (G1754) are a fairly recent device from Vishay and are a 'much faster switching device' and a lot if these tend to produce a rather bright edge to the sound, but again, this is totally dependent on the intended use, circuit, etc, etc
Possibly, best answer is to get some of either device and try them out in your application and have a listen
The 1N5404 are a pretty standard diode, been around for a long time, works okay but not the quietest device in good quality power supplies even when run under 1A
The 708-8016 (G1754) are a fairly recent device from Vishay and are a 'much faster switching device' and a lot if these tend to produce a rather bright edge to the sound, but again, this is totally dependent on the intended use, circuit, etc, etc
Possibly, best answer is to get some of either device and try them out in your application and have a listen
I found on valve amp B+ supplies I got switching noise with ordinary diodes.
I had to use fast recovery diodes or the noise got into the audio chain.
I had to use fast recovery diodes or the noise got into the audio chain.
Thank you your comments, gentlemen.
If I can select a few of them which raise questions in my mind ... perhaps someone can provide answers?
Q1. "I got switching noise with ordinary diodes. I had to use fast recovery diodes or the noise got into the audio chain."
This I can understand. But is a "fast recovery" diode, the same "fast diode" as a "switching diode" (which I can understand must be faster than a "standard recovery" diode)? 😉
Q2. "If it's for an audio component, perhaps you might look for a specific diode called a 'soft recovery diode'."
Would this be a candidate? Vishay HEXFRED HFA08TB60 diode - 8a/600v.
The circuit I want to use them in is a simple 240v mains DC blocker, for an 800w sub amp. Rod Elliott has the circuit here (scroll down to Figure 8, near the end of the article):
Mains DC and Transformers
I am currently using 1N5404s, as Rod specified but I want to enable a higher current use than these allow - hence my search for a 6a (or more) equivalent to 1N5404.
NB: You might like to comment whether I am being unnecessarily timid. These diodes can take 3a continuous - am I right in thinking that because the Rod Elliott circuit has 2 parallel diode strings, 6a can comfortably pass through the diodes? Even an 800w (into 4ohms) sub amp is not going to draw this, continuously!
Q3. "Super Barrier Rectifiers."
Can you give me a product number? I googled 'Super Barrier Rectifier' but couldn't find an actual product no.
Thanks,
Andy
If I can select a few of them which raise questions in my mind ... perhaps someone can provide answers?
Q1. "I got switching noise with ordinary diodes. I had to use fast recovery diodes or the noise got into the audio chain."
This I can understand. But is a "fast recovery" diode, the same "fast diode" as a "switching diode" (which I can understand must be faster than a "standard recovery" diode)? 😉
Q2. "If it's for an audio component, perhaps you might look for a specific diode called a 'soft recovery diode'."
Would this be a candidate? Vishay HEXFRED HFA08TB60 diode - 8a/600v.
The circuit I want to use them in is a simple 240v mains DC blocker, for an 800w sub amp. Rod Elliott has the circuit here (scroll down to Figure 8, near the end of the article):
Mains DC and Transformers
I am currently using 1N5404s, as Rod specified but I want to enable a higher current use than these allow - hence my search for a 6a (or more) equivalent to 1N5404.
NB: You might like to comment whether I am being unnecessarily timid. These diodes can take 3a continuous - am I right in thinking that because the Rod Elliott circuit has 2 parallel diode strings, 6a can comfortably pass through the diodes? Even an 800w (into 4ohms) sub amp is not going to draw this, continuously!
Q3. "Super Barrier Rectifiers."
Can you give me a product number? I googled 'Super Barrier Rectifier' but couldn't find an actual product no.
Thanks,
Andy
As Rod mentioned, for higher current applications, use a block bridge - 35A block bridge at Jaycar - just make sure the caps will take the load current
.. Note that if no soft-start circuit is used, larger diodes are highly recommended. Regardless of whether soft-start is used, a 25A or 35A bridge rectifier assembly is a simple and relatively inexpensive way to obtain very high current diodes, already neatly packaged and insulated. When using a bridge, remember that + and - must be shorted together to obtain 4 diodes in anti-parallel (as shown in Figure 3).
.. Note that if no soft-start circuit is used, larger diodes are highly recommended. Regardless of whether soft-start is used, a 25A or 35A bridge rectifier assembly is a simple and relatively inexpensive way to obtain very high current diodes, already neatly packaged and insulated. When using a bridge, remember that + and - must be shorted together to obtain 4 diodes in anti-parallel (as shown in Figure 3).
A little bit, perhaps, but not imprudent. Better safe than sorry. But you aren't analyzing the circuit correctly. The diodes aren't in parallel; they're anti-parallel. So they don't 'share' the current load.
Rod suggests in his article that a diode bridge can be used. Have you considered that option?
SBR10U300CT
Its a pair of 5A diodes in a 3 pin TO-220.
Fullpack isolated is an option if you prefer.
Yes you can parallel them, since they are
an array of parallel mosfets inside anyhow.
The cathode is common. For no reason I
can figure, they don't yet offer a standard
axial diode, common anode, or full bridge
in a usefully high voltage. And why not?
I really thought Diode's SBR offered offered
something "normal", guess I was wrong...
Maybe someone else offers the equivalent
in a different brand?
Silicon carbide Schottky is another maybe.
If you need more than 300V and don't mind
a rather large forward drop. Even more drop
than a fast recovery silicon diode, but the
sacrifice may be worth it for getting rid of
reverse recovery.
Perhaps Cree's C3D06060F
We just care about not hearing it 120 times a sec.
But if you switch hundreds of thousands of times
a sec, reverse losses can add up worse than the
forward losses do. Which drives the market.
It can be implied from the packaging decisions,
that diode manufacturers care nothing about
selling them for the convenience of diy audio.
Last edited:
"Switching" simply means it's a fast recovery diode suitable for use in a switched mode power supply. Fast recovery diodes work fine in ordinary line frequency power supplies, though they are never used in mass production because the standard recovery ones are a few cents cheaper.
6A is about the limit for what a wire ended diode package can handle. It needs to get rid of something like 5 watts of heat. So, your choice of wire ended 6A diodes is limited, if you go to TO220 packages or bridge rectifier blocks, you will get a wider choice.
Thank you very much, scopeboy. 🙂
6a (which I can get) is fine, methinks.
So if "Switching simply means it's a fast recovery diode suitable for use in a switched mode power supply" ... that means 708-8016 (G1754) can be used to give higher current-carrying capacity for the Rod Elliott circuit?
Thanks,
Andy
Aah, OK, thanks for the explanation (my lack of electronics understanding! 😉 ). So that diode-foursome can only pass 3a of current (continuous).
I used the DC blocker "in front of" an electric kettle ... and noticed that the diodes got too hot to touch! Since it was drawing about 9a ... I'm not surprised - but it suggested that, for 800w sub amps at least, I should increase the amperage above 3a! 😀
Attached is a pic of the device ... the way I've configured it makes it easier to use "wire-ended" diodes, rather than a bridge. The caps are 3 x 2200uF/100v Nichicon KGs in parallel (instead of a single 4,700uF) - so they can take 10a ripple current.
Regards,
Andy
Attachments
Last edited:
Yes, each current polarity is steered through a series pair.
The kettle probably uses a resistive heating element, which could well be greater than 800W. The average power of the sub's music signal won't be anywhere near 800W.
Also keep in mind that the blocker sees primary side current, considerably less than the secondary current through the sub's circuits.
Looks good. It doesn't appear to me, though, that a high-current bridge block with in-line pins would be too difficult to fit.
Axial lead diodes are rated based upon a specific lead length, soldered into a PC board. Your assembly will overheat the diodes if you push them to their specification current.
You are better off with a bridge rectifier, with the case mounted to the chassis with thermal compound.
jn
Hey Andy,
If your 800 watt subwoofer amp is a digital amp with a switch mode power supply, you don't need to worry about a dc blocking circuit
[Incidently, for those folks looking at this 'dc blocking circuit', the current goes thru the caps and not the diodes - it's a bit misleading how it works]
However, if this in not the case and you do need to use the whole 800 watts, and it is RMS and not music watts, then you're looking for a couple of really heavy duty caps for this job made for industrial use, not high fi power supply caps - they're not made for this sort of job
Plus, the current isn't shared evenly over the 3 caps and the current capacity fades with age
If your 800 watt subwoofer amp is a digital amp with a switch mode power supply, you don't need to worry about a dc blocking circuit
[Incidently, for those folks looking at this 'dc blocking circuit', the current goes thru the caps and not the diodes - it's a bit misleading how it works]
However, if this in not the case and you do need to use the whole 800 watts, and it is RMS and not music watts, then you're looking for a couple of really heavy duty caps for this job made for industrial use, not high fi power supply caps - they're not made for this sort of job
Plus, the current isn't shared evenly over the 3 caps and the current capacity fades with age
Interesting - thanks, James.
Can you explain why? Yes, the sub amp has a SMPS.
That's what I thought too ... until, just as a test, I plugged my electric kettle into the mains output socket ... and noticed that the diodes got extremely hot. Given they are 3a diodes and the kettle was probably pulling 8 or 9a, I'm not surprised! 😉
All I can assume is there must've been current flowing through the diodes, to get them hot. BTW, the caps did not change temperature in the slightest.
Noted - thanks. Can you suggest a make for such caps and where I can buy them from?
Why so? they're in parallel?
Most of the info/explanation about the caps carrying the current and the job/use of the diodes is written really well in Rod's article -we're just so used to thinking about thew one use of diodes that it's a bit strange to think they have a different purpose here
The type of caps you're looking for are the industrial type of electrolytics that used to be found in welders, and such things - many companies still make this sort of cap, usually noted having metal cases to keep them cooler and often studs on the bottom for extra heat transfer - we used to use the Siemens ones but Rifa and BHC (same company now)also have some that'll do the job - no idea who would have these today but the guys at Rockby Electronics in Clayton might know
The SMPS is a 'chopper' circuit - it cuts the ac waveform up in thin slices and doesn't care if there's a bit of dc component on the line
Now, with Torroidal transformer supplies, it's imperative to make sure no dc gets anywhere near the transformer and dc traps should be mandatory for transformers larger than bout 150va - even R-core transformers benefit from fitting dc traps unless you're using one of those regenerator 240 volt power supplies.
Ah, caps in parallel - there's a whole world of stuff about caps - basically in this use, very few caps are identical and over time, the small differences in temperature or even the difference in the wires joining them together will cause them to age a bit differently - then the current going thru them amplifies these small differences until one day, one of them will exceed it's operating conditions and fail - it's mainly the turn on surges that causes the damage and using an inline thermistor (on the active line, for example) will help a lot - plenty of different opinions about all this and I tend to favour Rod's approach pretty much in this sort of thing
All the best with the diodes, etc
The type of caps you're looking for are the industrial type of electrolytics that used to be found in welders, and such things - many companies still make this sort of cap, usually noted having metal cases to keep them cooler and often studs on the bottom for extra heat transfer - we used to use the Siemens ones but Rifa and BHC (same company now)also have some that'll do the job - no idea who would have these today but the guys at Rockby Electronics in Clayton might know
The SMPS is a 'chopper' circuit - it cuts the ac waveform up in thin slices and doesn't care if there's a bit of dc component on the line
Now, with Torroidal transformer supplies, it's imperative to make sure no dc gets anywhere near the transformer and dc traps should be mandatory for transformers larger than bout 150va - even R-core transformers benefit from fitting dc traps unless you're using one of those regenerator 240 volt power supplies.
Ah, caps in parallel - there's a whole world of stuff about caps - basically in this use, very few caps are identical and over time, the small differences in temperature or even the difference in the wires joining them together will cause them to age a bit differently - then the current going thru them amplifies these small differences until one day, one of them will exceed it's operating conditions and fail - it's mainly the turn on surges that causes the damage and using an inline thermistor (on the active line, for example) will help a lot - plenty of different opinions about all this and I tend to favour Rod's approach pretty much in this sort of thing
All the best with the diodes, etc
- Status
- Not open for further replies.