VVR not working

[dude]

I searched here and elsewhere but not much info on troubleshooting a VVR. All values and pot good, used Dana's board but didn't check the zener (not sure how, same as diode?) I'm assuming the Mosfet blew, is there a way to check it? Worked for a few years then...?


al

[sluckey]

It's cheap. Just replace it.

[dude]

Not the NTE2973, Mouser wants $12, might be $14 now.
Any suggestions on other replacements that might be cheaper and where.


It's in a 5E3, voltage in and out is the same turning the pot, always stays on full won't lower. Had this problem with other amps and replacing it worked, just hoping it wasn't the mosfet before replacing. OK, I'll replace it, need some other parts too but the NTE2973, one Dana recommended years ago, ain't cheap. 


One other thing, posting at times I get very small lettering but preview usually doesn't post showing the small print, what gives?


Thanks in advance,
al

[dude]

Sluckey, I found some info Dana Hall posted at Gear Page a few years ago about NTE2973 substitutes. Guy's over there were harassing him on the NTE2973 being over spec'd and over priced. He justified why he uses that Mosfet, (basically because he tried several and NTE had the best results from his experience) for $5 more the NTE2973 works the best, he ought to know he basically invented the VVR with his brother Richie, I think. Mouser doesn't carry them anymore but ebay has some at reasonable prices.


I have four amps with his VVR installed two have worked for over five years, no problems, the other two worked for a year or two and blew. I've seen some difference installation info and would like to ask your opinion. Some are wired from the standby to the "in" on the VVR and out to the first filter (doing the whole amp). I think I remember in Dana's instructions that he has a cap between the standby and "in" on the VVR. You think this cap is necessary or possibly filters out ripple to the VVR, making for a better install..?


Thanks for all your advice,
al   

[sluckey]

Better find a substitute. NTE2973 is no longer being manufactured. DummyLoad and/or tubenit probably have some sub info.

[2deaf]

STW12NK90Z has built-in Zeners that protect it from over-voltage at the gate (but they don't take the place of the external 12V Zener in the crude current-limiting circuit).  Five bucks and readily available.

[dude]

Yeah, that mosfet was mentioned on another forum, but Dana Hall said to use the external zener too. Thanks, I’ll try that one too, I was able to get two NTE2973’s off eBay for $24 with shipping, $10.50 each, they are available but I think you’re right discontinued, at least at the big supply houses.                                                                                                               Thanks, al

[Washburnmemphis]

Do you have a standby switch in your build? 


If you do, the next question is do you have a  filter cap on the Rectifier side of the switch?


The cap will help protect the mosfet.  Just don’t exceed the cap limits for the rectifier.

[kagliostro]

There are lot of good cheap Mosfet for VVR use, you don't really need an NTE2973

--

London Power uses FQA8N100C (at least on some kit)

https://docs-emea.rs-online.com/webdocs/0d27/0900766b80d27314.pdf

around 4 €

https://it.rs-online.com/web/p/transistor-mosfet/6714972/

Franco

[dude]

Thanks for all the info. I never used a filter before the VVR on any of my amps, some schematics show a filter most don't.  Some of the mosfets on my amps blew after year or so, some still going after five years...? My two cents: I contribute the mosfet's  life expectancy is related to heat, depending on where you install it could be the culprit...? I've always use an old PC's heatsink and small 2" fan opposite the mosfet, and still some blew, maybe too close to PT blowing hot air on heatsink? I've only fed these 12vdc fans with 5 to 6vdc, from now on I'll take more caution on where I put the mosfet.


al     

[brewdude]

I have built 4 or 5 amps from 5 to 30 watts cathode biased with a VVR. 
I have always used a filter cap preceding the VVR, only heat sink is the attachment to the chassis through the provided insulator without any thermal conducting paste and with no fan, and they all still work... am I just lucky?

[Deric]

I had this issue with an AC-15 ish build.  Amp was wired:


Rectifier> VVR> Standby> Filter cap


Blew 2 FETs then added a cap between the rectifier and VVR.  Problem went away.  A year or so later I did some testing and thought it felt different with the extra cap.  Pulled it out.  A week later blew the FET.  Finally just moved the first filter to before the VVR (instead of adding a cap). 


Rectifier> Filter cap> Standby> VVR



Sounds and feels the same ( I think??) and has been rock solid for a couple of years.


I've been using IRFPE50  $3.85 at Mouser


Good luck.

[Ed_Chambley]

I build these things on a rectangangular double sided prototype board and I collect the heat sinks from old dead computer power supplies.  I always use one.  But then again, I put a 30mm x 30mm fan in all builds and add them to others


I do not prefer a VVR, but I have put them in a lot of amps and with a heat sink I never had a failure.  At first I ordered a Bias Tracking version for a JTM45 install.  Could not keep it cool.  Mounted to the chassis.  Got a new MOSFET from Dana and he quickly sent a new one and asked me to use thermal grease and not use a pad.  He sent another as I am sending him photos of how it is installed and works for s short time.  Finally, I mounted it to a heat sink and it is still working today I guess.


So now when I build them up, I saw a piece of the heat sink small enough to glue to the board and attach the MOSFET.  I leave the fins, just cut it 3/4 inch tall.  When installed it looks as if it just hangs there, but it is so small the pot holds it great.



I too use the IRFPE50.  I attached the page London Power recommends for pacement of the board.  This is in either case, Cathode Bias or Fixed.

[zendragon63]

I apologize in advance if this is a thread hijack. I use 'VVRs' on nearly every amp I build and moved from the NTE2973 (plus the 12V Zener) to the STP6NZ90Z--which has the internal zener.


2deaf , in reply #5:



STW12NK90Z has built-in Zeners that protect it from over-voltage at the gate (but they don't take the place of the external 12V Zener in the crude current-limiting circuit).  Five bucks and readily available. 
 
 
 

Yet to have a failure in at least a dozen 20W to 40W amps, so why would the additional external 12V zener be necessary? Could it be just the sequence of

Rectifier-> Cap->VVR (which is what I use). Just curious at what I am ignorant of (which is plenty ). Thanks in advance. Regards

dennis

[2deaf]

. . . so why would the additional external 12V zener be necessary?

The external Zener is part of a current-limiting scheme as opposed to the voltage-limiting scheme of the internal Zeners.  A discharged capacitor appears as a direct short when it is first charged and this can run some pretty serious current through your MOSFET.  Even though the bold-face specifications of your MOSFET make it look like it should be able to handle it, a closer examination may indicate that it cannot.  So some sort of current-limiting circuitry might be prudent.

The MOSFET is configured as a Source-follower in the VVR so that the Source winds up several volts lower than the Gate.  They place a 10r resistor in series with the Source which will drop some more voltage depending on how much current is flowing through it.  Let's say the Gate-to-Source voltage is 5V and there is 700mA flowing through the 10r resistor creating a 7V drop.  Now the voltage across the 12V Zener is 12V and it breaks over allowing current to flow through the 100K resistor that is between the wiper and the Gate.  This creates a voltage drop across the 100K resistor that decreases the voltage at the Gate which decreases the voltage at the Source, thereby reducing the current flowing through the MOSFET.

[zendragon63]

Thank you very much 2deaf. Makes sense to me. What a great forum. Regards


dennis