[ARC5] Solid State 6AL5

Thu Feb 8 17:07:35 EST 2018

A MOV after the fuse will moderate a lot of that risk. 

Peter

> On Feb 8, 2018, at 5:03 PM, Bob kb8tq <kb8tq at n1k.org> wrote:
> 
> Hi
> 
> One of the major debates on this sort of thing has always been “how much margin
> do I need?”. There is a lot of evidence that your line voltage has various burps on 
> it that can go to pretty high levels. Just how this impacts your choice of diodes is
> something you can spend a lot of "quality time" on …..
> 
> The guys testing PC power supplies seem to routinely nuke them with 8X over voltage
> spikes ...
> 
> Bob
> 
>> On Feb 8, 2018, at 4:48 PM, Peter Gottlieb <kb2vtl at gmail.com> wrote:
>> 
>> I doubt anyone uses 1N4007’s in anything HV these days. Look at Mouser and you can find cheap 4kV diodes in stock and I see that single component 10 kV ones are around as well. 
>> 
>> 
>> Peter
>> 
>>> On Feb 8, 2018, at 4:23 PM, Dennis Monticelli <dennis.monticelli at gmail.com> wrote:
>>> 
>>> The fast diode problem is unequal release of stored charge.  Quite a bit of instantaneous current flies around when fast diodes are suddenly reversed in a PWM circuit.
>>> 
>>> Yes, a dead short in one diode of a string increases the reverse BV of the survivors, thus stressing them further.  I generally design for 50% of BV rating when using common Si power diodes in 60Hz applications.  1N4007's are so cheap these days while 1KV rated resistors are not.
>>> 
>>> I wonder how modern day HV 60Hz rectifier modules are built? Do they do anything to match BVs or leakages before assembly?  Do they use resistors?  I haven't broken one open.  My guess is that they a simple test to throw out the out-liers to insure they are using only the middle of the distribution and then dispense with resistor balancing. 
>>> 
>>> Dennis AE6C
>>> 
>>>> On Thu, Feb 8, 2018 at 12:23 PM, Peter Gottlieb <kb2vtl at gmail.com> wrote:
>>>> Yet, I still find manufacturer app notes, such as STMicro AN443, that talk about this, although in the context of fast diodes. 
>>>> 
>>>> application-notes.digchip.com/005/5-10702.pdf
>>>> 
>>>> One diode shorts in that HV bridge and the whole rest of them will not be far behind. 
>>>> 
>>>> I’m not saying you’re wrong, just that I’d like to see a manufacturer or academic reference justifying eliminating the resistors before I’d accept it in anything beyond hobby use. 
>>>> 
>>>> 
>>>> Peter
>>>> 
>>>>> On Feb 8, 2018, at 3:04 PM, Dennis Monticelli <dennis.monticelli at gmail.com> wrote:
>>>>> 
>>>>> FYI.  The practice of using equalization resistors dates way back to when early Si manufacturing had uniformity issues.  If you are using modern 1N4007's (particularly from the same batch) don't bother with the resistors.  Just design with the usual BV margins and you will be fine.
>>>>> 
>>>>> Dennis AE6C
>>>>> 
>>>>>> On Thu, Feb 8, 2018 at 12:00 PM, Peter Gottlieb <kb2vtl at gmail.com> wrote:
>>>>>> In my test set the 5R4 is set up as a full wave rectifier from a 1700 volt CT winding. At one point I thought the tube was weak and I’d replace with diodes. I have a bunch of 1N4007 but looking at the voltages I realized I needed three diodes for each plate!  And then equalization resistors, but with their max rated voltages I needed a lot of them. This started to be more trouble than it was worth and I was able to locate a replacement tube in my stash and just went with it.
>>>>>> 
>>>>>> 
>>>>>> Peter
>>>>>> 
>>>>>> > On Feb 8, 2018, at 2:47 PM, Richard Knoppow <1oldlens1 at ix.netcom.com> wrote:
>>>>>> >
>>>>>> >    I don't remember if I ever measured the resistance of the 5R4 replacement I have and don't know where it is now. However, the B+ with it installed was substantially higher than with the tube rectifier and it was small so I don't think it had anything but diodes in it.
>>>>>> >
>>>>>> >> On 2/8/2018 10:17 AM, John Watkins via ARC5 wrote:
>>>>>> >> My solid state 6AL5s have just diodes and possibly a resistor in series with them (2 diodes).  They measure 16 Ohms forward resistance using a Simpson VOM on the Rx1 scale.off the upper end on both Rx1 and Rx100 for reverse.  Nothing in mine for heater load.  That is a no connection.  I haven’t removed the amber epoxy yet to see if the diodes have a 1N number.  Will report on that when I get back to town.
>>>>>> >> John WD5ENU
>>>>>> >
>>>>>> >
>>>>>> > --
>>>>>> > Richard Knoppow
>>>>>> > 1oldlens1 at ix.netcom.com
>>>>>> > WB6KBL
>>>>>> > ______________________________________________________________
>>>>>> > ARC5 mailing list
>>>>>> > Home: http://mailman.qth.net/mailman/listinfo/arc5
>>>>>> > Help: http://mailman.qth.net/mmfaq.htm
>>>>>> > Post: mailto:ARC5 at mailman.qth.net
>>>>>> >
>>>>>> > This list hosted by: http://www.qsl.net
>>>>>> > Please help support this email list: http://www.qsl.net/donate.html
>>>>>> ______________________________________________________________
>>>>>> ARC5 mailing list
>>>>>> Home: http://mailman.qth.net/mailman/listinfo/arc5
>>>>>> Help: http://mailman.qth.net/mmfaq.htm
>>>>>> Post: mailto:ARC5 at mailman.qth.net
>>>>>> 
>>>>>> This list hosted by: http://www.qsl.net
>>>>>> Please help support this email list: http://www.qsl.net/donate.html
>>>>> 
>>> 
>> ______________________________________________________________
>> ARC5 mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/arc5
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:ARC5 at mailman.qth.net
>> 
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.qth.net/pipermail/arc5/attachments/20180208/91c33754/attachment-0001.html>