"Volt Modding" Mosfets

[TheRealPorkchop]

Yes, I know that's not the correct term and the way it's done isn't technically volting anything...

I got a radio in the other day, had been converted to from a 1969 (which I assume had blown out) to a 2030 with an actual 369 CP but there was only about .25 watts of carrier. I will assume the "final" was blown out, the guy it belonged to didn't want to fix it. I flipped the radio over, the red wire mod was done to it. So my question...


I've seen that done to radios with regular transistors in it and I've seen a wattmeter reading improvement. Can it be done to a mosfet though, safely and without issue? Doesn't seem so. If we go to all this effort to get the things biased a certain way and then some turdburger goes and throws that wire on there, that's sending the input voltage directly to that mosfet, right? Doesn't that bypass the bias crap that we sit and build and install along with the mosfet? Besides ignorance, why would someone do that OR am I wrong and you can do that and it had nothing to do with that radio not producing a strong carrier?

 

[sonoma]

you can do the red wire on the bottom mod with no problems if it is done right in the 1st place. look at it thius way. all you are really doing is making the AM side just like the SSB side. SSB has the full 13 volts going to the final so all you are really doing is making the AM side the same way. at least this is the best way I know to explain it
I am not that good at trying to explain to you what the red wire mod does.
Maybe Nomad or Andy will chime in here. they are good at writing the info out in long form..
have a good day PorkChop

 

[TheRealPorkchop]

Ok well I guess I was wrong then. I was almost certain that’s what popped it but I didn’t get to work on it since he said don’t fix it. Guess you can “volt” a mosfet then.

By doing that I assume that through the bias thing out the window?

 

[OldTech03]

Controlling voltage supply to the Mosfet was never or should not be an issue most MOSFETs will handle much more than the usual 12.5 or 13.5 volts and are actually rated much higher voltages on the Drain to the tune of 50-100 volts and the Drain current is often rated at different temps which speaks directly to the cooling issues as it does with any device. which leads to the idea that an IRF520 could develop as much as 50 or 60 watts and as much as 50+ volts on the drain as long as you can give the mosfets the needed cooling to allow that or as we sometimes say keep them at room temps. Just remember the radios with 4 mosfet finals are supplied with a constant 13.5 volts. The most important voltage will always be the Gate voltage making sure that don't exceed specs and like I said the heatsink efficiency will determine how long a mosfet running a specific voltage will last. I have experimented with these ideas and they do work as I have run these mosfets at 24vdc and the only issue is keeping them cool 4 520 will easily develope twice the power with 4 times the cooling capacity which is no small thing that would make it impractical as the heat sinks would be twice the size of the radio you also must take into consideration all other related parts must be rated much higher as well.
.

 

[TheRealPorkchop]

Ok, well I’ll ask because I’m ignorant to the subject, why do they use those companion parts to regulate bias? Is that not controlling the voltage on the gate? If you use that wire to jump the input voltage to those mosfets, aren’t you bypassing the bias parts?

The service manual says like 3.6 volts on the final mosfet, if that gives say 50w pep, then jumping that voltage to it, you’re saying the output wattage would increase, right? And if that’s so, would the heat sink in say, a Galaxy chassis be sufficient to cool them at the input voltage of 14.2 which is found in most modern trucks now?

I figured whoever “volted” that mosfet, that’s why it took a dump and stopped working. Again, I guess I was wrong in that assumption. Now you have me curious and I want to pull a new radio out the box and bolt it just to see what happens.

 

[TheRealPorkchop]

I friggin’ hate answering or posting from this stupid phone. I didn’t type Donny. Who’s the idiot doing the programming on these things?

 

[OldTech03]

It's true I did not go into the details of what you would have to look out for in the event you decided to modify a circuit but as a general rule the Gate voltages "bias" in mosfet amp circuits grab the voltage from a source that is higher than the needed gate voltage or "bias" and then regulated with a zener circuit down to that would not allow you to go above 5vdc it should probably be not allowed to go above 3.5 v but that is what the vari-resistor is for and what knowledgeable techs are supposed to take care of. the is feeding the Drain with 13.5 volts should not affect the gate voltage unless the circuit is very poorly designed which one could argue all day. In a good design, the gate voltage would be regulated via an actual regulated supply rather than a zener. Some if the associated parts in Mosfet mods like the old EKL parts were designed around there not being a regulated bias for the Gates but newer radios are now using the Zener circuits which we used back before there was an EKL kit. And yes these mosfet mods were around before EKL came to be at least in my part of the country they were because we were messing with these back in the late 80's.

 

[TheRealPorkchop]

I think I’m following you here, makes a little more sense now.

 

[Handy Andy]

The main focus is the Gate, it has to make the part stay off when it should - and on when it's needed - then make sure it returns back to off when not needed.

Ever wonder the numbering on some MOSFET like the 13N10? The way Fairchild used to look at it - The last two digits refer to the voltage rating in tens of volts - 12N08 is a PREVIOUS Generation part with only 80 Volt Rating from S to D - so 13N10 refers to 100 volts breakdown. IRF though - doesn't use that convention - but the higher the last two digits are - the greater the Breakdown voltage is.

This also has a reciprocal effect - the higher breakdown voltages means thicker layering of the part - so in RF sense - this limits it's ability to pass a signal without propagation issues - reflections - timing problems. Light is a finite speed - RF thru a medium like Silicon is also only "so fast" as it travels a given distance thru a junction before the cycle reverses.

Bipolar had the same problem with junctions - their thickness, surface area and doping - made limits for the amount of power and the speed they can pass and control and how well they could handle INDUCTIVE loading and the back EMF from the device it controlled (Remember TRIACS? Dimmer switches mostly but also variable speed driver units for motors and is what makes Switching Power supplies more favorable - in AC power units)

Earlier MOSFET designs had more "coupling" problems - where the device would turn on - then stay on - and never turn off until you removed the power from all the leads (Source, Drain and Gate) - hence self-destruct. All due to it's construction - the electrical field passing thru the device - once it turned on, latched it on. More like SCR's you'd find it in Power supplies across the load from the source, supplied/designed to sense overload and "fire" - acting like a dead-short, to trigger a breaker - pop the breaker to remove the power fault.

They are the offspring of the original FETs' you use in RX IF mixers the typical Grant or Cobra or PC-78 uses (SK192)

Nowadays, they are constructed differently - they have a "built in" protection to allow the device to turn off when the gate goes neutral - (no applied voltage Positive Or Negative). That is where we have our problem - how much do you use to "turn them on" then, once they're done - how much power can remain at the Gate so they can turn off when you don't need it?

On top of that is -

How much Gate voltage or drive is needed to make the MOSFET work linearly? As in we know Class AB biasing in Bi-polars, you need that for SSB to even work - Base gets "just enough" so when RF is applied - it is ready to amplify, but once that RF is gone - we're back to waiting around for that next wave to hit...and we have to stay off until that wave comes.
MOSFET is different - it can sit there even with some noise on that gate - because the electrical field the Gate has to produce is not yet strong enough to pass a barrier to reach the region it needs to make active to turn the device on - so it just sits there looking back at you until a level is reached and WHAM - we're off and running. Just don't forget they generate a field as they pass power across the junctions - so that too affects how these devices act when they are on, and how well we can turn them off.
So we have to step back and look at how Tubes can operate - two plates, one called Cathode, other the Plate and place a screen called a Grid in between the two - house it in a vacuum and make it work. Apply a voltage and figure out what goes where...​

Now you can begin to understand the datasheets importance - because not every part of the same lot, manufacturer or amongst the brands - have or use the same "gate voltage" to trigger them on.

It's why you see Min, TYP, Max columns on datasheets - it is the variances of the same part amongst their own testing and a quality issue we have to pay attention to.

Why some people tend to overblow the "matched set" offer - is due to this, and overrated to a point, but it can pose a problem for us if we don't allow for it.

More than likely what happened in that radios' situation - the Gate "blewup" the part because of lack of understanding how that part works, not from overvolt - an ERF can easily handle up to 100Volts and not budge. So for that part to "blow" means something else - the sensitivity as well as level of drive needed for such a part - they are EXTREMELY high gain devices - worse than a Bipolar - but also better than a Bipolar due to the efficiency or VERY SMALL SIGNAL levels can make BIG POWER Changes. But cannot operate as "Linearly" as a bipolar because it has no true influence on the power passing thru the device that Bipolar's can.

They've come a long way - but there's still a lot more to go...

 

[Shockwave]

One other thing to consider if you're "volting" the final in an SSB radio for higher AM power is that the audio or modulation density will suffer. The reason these radio switch to 6 volts on AM is so that the series pass modulator can swing that voltage from 6 volts carrier to zero on negative audio peaks and 12 on positive peaks. It's 6 volts DC with 12 volts peak to peak AF modulation.

When you volt the final you remove the ability to independently modulate this stage. All remaining audio must come from the only stage left with modulation, the driver. The better approach is to supply the 12 volts through a choke that looks like an open to the AF. Then you can feed the output of the audio chip through a coupling cap back into the final. That will modulate the 12 volt line feeding the final.

 

[loosecannon]

nice to see you around Shockwave!

i remember discussing this exact issue with you years ago when you turned me on to the "transformer modulated 148" mod.

when i started comparing volted finals that only modulated the driver to radios that modulated both the driver and final, i found out all about modulation depth and what the lack of it sounds like.

LC

 

[Martian]

Lack of understanding voltage vs current controlled devices.

There is no magic one all be all gate voltage so that one you need to put out of your mind first. If there was then there would be no need for proper bias regulation.

As far as the power supply to it... volting has consequences for BJT or MOSFETs that is virtually never considered.
And that is the additional heat.