Valiant PS question - cap input?

[AB3FL]

I was wondering is it is possible to put a cap (say 4uf) before the choke on a valiant in the HV PS.  I know that it would raise the voltage overall, but what negative effects would it have.  Running a simulation, under load, the HV would go from about 600V to about 700V.

thanks

Tom - AB3FL

[WQ9E]

Tom,

MV tubes won't like the higher peak current and it will cause more transformer heating for the same reason.

I assume you are trying to raise the plate voltage because of different modulator tubes???  Otherwise the Valiant is pretty much running at its safe power on AM given the limits of the power supply and modulation transformer.  I think the earlier Viking 1 and 2 are pretty conservatively built but the Valiant and most later units appear to be designed to produce both profit and RF.

[W5COA]

If you want to raise the plate voltage, replacing the MV tubes with solid state rectifiers will probably do it better than going to a capacitor input filter.

[ke7trp]

Yep. Get some solid state plug in rectifier diodes.  They work great in the valiant and give you a big boost in output. Maybe then you can use the cap also

C

[WD5JKO]

I wonder what the objective is of this project idea? Some quick napkin doodling says you will have > 800v after capacitor input conversion. The bigger the input cap the higher the voltage to a point. This higher voltage will certainly raise the carrier power, and the PEP output of the rig, but will also stress the tubes, and rig beyond the original design.

Switching to solid state rectifiers over the mercury vapor rectifiers only raise the B+ a few volts since the forward drop of a MV is ~ 15v (866), and 10v for Xenon (3B28). Going solid state in a Viking I or II raises the B+ a lot more since the 5R4's drop a lot more voltage.

Raising the PEP power of a Valiant will likely raise the RF out pi net loading cap issue where in stock form these are known to be a major PITA due to component failure.

Jim
WD5JKO

[WQ9E]

I agree with all Jim's points and additionally the slight additional output achieved from going to higher plate voltage would not be heard (or even measurable) on the receiving end.   

For a serious contesting station you can perhaps justify that a couple of db may be worth chasing but not for a vintage AM station.  Doubling the Valiant output (not possible without major modification) would only net 3 db (perhaps half an S unit in amateur speak).  The much despised clipper circuit in the Valiant, properly used, will provide an increase in communications ability under tough conditions far exceeding any practical power increase modification of the Valiant.  Hanging a robust legal limit linear amp on the output of the Valiant is the most practical way to increase its power in a meaningful manner.

I have a couple of Valiant 1 transmitters along  with a Valiant 2 transmitter and a second awaiting restoration and they do a good job when operated within their ratings.  None of mine has ever blown up a coarse loading switch but I have replaced the coarse loading caps in 2 of them.  They will cruise along fine at 275 CW / 200 AM but my Viking 500 which is just over double the power is far more than double the weight.  It takes a lot of iron for old style high power AM.

[KE6DF]

"If you want to raise the plate voltage, replacing the MV tubes with solid state rectifiers will probably do it better than going to a capacitor input filter."

Yep. Get some solid state plug in rectifier diodes.  They work great in the valiant and give you a big boost in output. Maybe then you can use the cap also

C

I disagree with these suggestions.

Going from MV tubes to sold state rectifiers won't make a big difference in output voltage.

The voltage drop across an MV rectifier is usually less than 15V so going to SS can't raise output more than that.

Now, going from vacuum recitifiers to SS does make a big difference because the drop across vacuum rectifiers is much higher.

[ke7trp]

HUH?


Removing 866s on a valiant typicaly ups the HV by over 100 volts.  I like to use the plug in style Diode replacements. You can also use 3b28s instead of 866s and you get alot of that 100 volts. But still have the heat. You also have the chance of shorting the fil winding in the HV transformer.


As for power output,  notes from my testing of HV voltage and various types of hv diodes is as follows:

carrier power. 
866 135 watts
3b28s  157
SS plug ins. 165

Pep output modulated:
866 350 watts
3b28s 420
SS plug ins 510.

So as you can see it makes a big difference in modulated audio output and little difference in carrier power.  Now you can load up past the recomended 330 ma and get the carrier up even more but I just loaded to the factory setting for these tests.

I say go for it.  They made millions of valiants. They are not rare. This conversion actualy helps the rig last longer and it clearly runs better. Make sure to put in some new electrolytics.  I used the 500v ones but the 450vs will take the SS diodes. 

If I had a 4 UF that would fit, I would probably try the cap input. My guess is that it would work just fine. The big trouble here is of course the wafer switches in the loading control and bandswitch. 

Have fun hotrodding and rememeber. If you leave that 871 blower off your Chevy, you will get better mileage and the engine will last longer!

C

[WD5JKO]

Clark,

  I don't have any direct Valiant experience. I just looked at the spec's of the tubes. Here is a 3B28:

www.shinjo.info/frank/sheets/049/3/3B28.pdf

As you can see the data sheet for a 3b28 shows an approximate voltage drop of 10-14 volts at 2 amperes passing through the tube. The drop will be somewhat less for more normal currents. A silicon diode array might have 4-5 PN junctions on each leg, which might drop 3-5 volts. This is less than the 3B28, but not by much. I do not dispute your results, but in saying that, i'm uncertain how you got that much more soup out of your Valiant with SS.

I have two examples of hot rodding a rig that is marginal at design specifications. Both belonged to W5PYT, Ozona Bob.

The first was his WRL Globe King 400. I beefed up the modulator driver so it could drive the heck out of the 5514 modulators. After doing that I could copy Bob on 10m backscatter 100% Q5 when I mobile'd away from his QTH to mine about 200 miles away. That rig really stood up and talked. Before I might copy every third word. A few weeks later, Bob was just starting a long dissertation as only he could do, and the RF tuning cap arc'd over. A sustained arc with plate current on the peg continued until a nice pair of V70D's were ruined.

The second was also with Bob, a Collins 32V2. I did the same stuff, such that those 807's could modulate cleanly well into class AB2 region. Again the boost in his talk power was impressive. Soon thereafter the RF loading capacitor module fried. That was the end of the road for that rig with Bob. He died before fixing it.

There is no right or wrong here. This is a hobby, and we can pretty much do whatever we want to our rigs so long as we stay within FCC requirements.

Just my 2 cents,

Jim
WD5JKO

[WQ9E]

Jim,

Your experience with going from MV to SS is the same as mine and not surprisingly follows what is shown in the tube specification sheets.  I noted no increase in output in my Valiant.  But flashover and other MV problems is now off the list of possible failures and it does also take the DC plate voltage off the 5 volt plate rectifier winding of the LV transformer.

I don't have any experience with 866A tubes that are approaching the end of life but maybe the voltage drop does increase and that is why Clark experienced an increase in HV.  But one of the positives of MV rectifiers is the low (and fairly constant across a reasonable range of current) voltage drop which is probably why the 83V rectifier is used in a number of tube testers.

As to modifying for more output you can certainly do so if a decrease in reliability/life is acceptable to you.  I prefer to maximize reliability but electronic hot rodding is far cheaper than mechanical in any case.  One of my friend's sons just proved that with an 18 month old Ford diesel pickup.  His driving combined with the aftermarket box resulted in a roughly $15,000 hole in the block when a rod decided to get a breath of fresh air.  You could blow up a whole lot of Valiants (and even a couple of Desk KWs) and not hit that figure.

[ke7trp]

I think it was improper tuning of the rig that caused those troubles. I really dont know how both could melt down like that otherwise.  Maybe a the rig was Driven into severe overmod and that produced the failures?  

Valiants are not good running rigs and they do fail all the time.  I had one that constantly gave me trouble when I have another that works every day.  I talk to a friend on 160 and it seems about every week, His valiant dies on the air and he has to go in and fix it again.  Sometimes a guy gets a good one and they seem to run and run. Othertimes your really in for a frusterating ownership.

I have plug in replacements in the GK500C both modulator and PS. Its been up for 2 years now running along.  The Champion 300 did not care for them. I ran them for about a year but the shock to the transformer in that rig got me worried so I took them out. Also. The champion only has 1kv oil caps in it. With the SS rectifiers installed the voltage jumped from 950 to 1200!

Sometimes there would be a huge BANG to the iron when keyed. Probably caught it on the wrong side of the cyle the moment I hit the key.  It seems smoother with the 866s back in it.  Power output really suffered on that rig.  With 866s it loads to 160 watts and can reach about 100%.  With the plug its it loads right to 220 watts and reaches well over 125+ mod.  

I have a valient that YEC rebuilt. Diodes saved that rig from the scrap heap.  The Transformer filiment winding was shorted.  Diodes where put in and the rig was back running. Sometimes they are required.

I understand the drop that is listed on various rectifiers. However, IN real world measurements, You will find a much higher increase in voltage on some rigs. Alot of this has to do with the config of the PS I think and maybe the age of the rectifiers you remove?

C

[K5UJ]

I have a general question about those s.s. drop  in replacements I have wondered about.  I have noticed that there are two types available for the 866A (of course now I can't remember the difference except that one type is somehow more robust than the other).  I've been told that when replacing tube rectifiers, especially h.v. ones, with s.s. stacks, you have to be careful because they don't work well in all all p.s.  I have taken that to mean that in for example an AM rig where the RF and modulators share the h.v. supply, the voltage swing can take out some s.s. drop ins?  Tnx

Rob

[ke7trp]

There are different ones in this format.  Some are short to replace the 866JR.  Some are tall to replace 866a and some are higher rated to replace the 3b28. Some are Rated at 10KV each.  Most are 250MA constant. 40 amp surge. I have some that are 1 amp each at 1kv.  They made these for 40 years so its hard to say what they are.  The good thing is that if its in this format, You can use it and most are major overkill for these applications.

The common one is 1n2637. I got lucky and got 20 of them at a ham fest for a couple bucks each.  I gave some away to fellow Amers and used the rest.

C

[WD5JKO]

I have a general question about those s.s. drop  in replacements I have wondered about.  I have noticed that there are two types available for the 866A (of course now I can't remember the difference except that one type is somehow more robust than the other).  I've been told that when replacing tube rectifiers, especially h.v. ones, with s.s. stacks, you have to be careful because they don't work well in all all p.s.  I have taken that to mean that in for example an AM rig where the RF and modulators share the h.v. supply, the voltage swing can take out some s.s. drop ins?  Tnx

Rob

Rob,

  Switching from tube rectifiers to SS is a broad topic. The conversion changes the power supply design in two ways. First any turn on delay with the tube rectifier is eliminated, and secondly the SS diodes will have a lower forward impedance. In many cases these changes can be accommodated, and as a side benefit we no longer have to power that rectifier filament(s).

  So combine a BA rig running on 125vac when the design was for 115vac, and then you SS the supply. Now the B+ will be much higher than stock, and the filter capacitors might be struggling because the surge B+ over the first 15 seconds after power on might exceed the capacitor voltage rating. The entire rig will run hotter from the higher B+ too.

   Adding SS where there is high back EMF pulses sometimes leads to diode failures. Early silicon diodes once avalanched became wounded, and would soon short out. The 1N4007 for example circa 1980 was a good example. These same diodes today are better with a more controlled avalanche characteristic. Still power supplies with choke input filters often need a suitable R-C snubber to protect the diodes. This is covered well in the Bill-Orr radio Handbooks.

   The 866 drop in replacements like what Clark posted likely have characteristics that allow for the back EMF pulses. I seem to recall that Sarkes-Tarzian diodes pioneered controlled avalanche characteristics. The rest of the old diodes needed shunt resistor and capacitor to survive.

Jim

[K5UJ]

Clark and Jim thank you very much--this helps explain a lot.  Clark i never thought about the 866jr and 3b28--of course--it all makes sense now.  You got a fantastic deal on those diode replacements!  I would have bought up all of them too.

Rob

[KE6DF]

Clark,

  I don't have any direct Valiant experience. I just looked at the spec's of the tubes. Here is a 3B28:

www.shinjo.info/frank/sheets/049/3/3B28.pdf

As you can see the data sheet for a 3b28 shows an approximate voltage drop of 10-14 volts at 2 amperes passing through the tube. The drop will be somewhat less for more normal currents. A silicon diode array might have 4-5 PN junctions on each leg, which might drop 3-5 volts. This is less than the 3B28, but not by much. I do not dispute your results, but in saying that, i'm uncertain how you got that much more soup out of your Valiant with SS.
Jim
WD5JKO

The 866a has a voltage drop of 14V max -- just like the 3b28.

http://www.scottbecker.net/tube/sheets/061/2/2V-400A.pdf

Perhaps those that see a big increase in output going from 866a's to solid state are replacing old, worn out tubes?

Dave

[ke7trp]

To test that, I put in a new set of 866a's from tubes and more dot come against the SS diodes.  Over 100 volts gain with the SS diodes on the champion.

C

[WD5JKO]

The SS replacements Clark posted are 1N2637. You can zoom in on the picture to verify that. Some Googling brought me to the specifications:

http://www.americanmicrosemi.com/catalog/1N2637.html

1N2637 Specifications
I(O) Max.(A) Output Current: 250m
@Temp (шC) (Test Condition): 70п
V(RRM)(V) Rep.Pk.Rev. Voltage: 10k
V(FM) Max.(V) Forward Voltage: 28
@I(FM) (A) (Test Condition): 250m
@Temp. (шC) (Test Condition): 100
I(RM) Max.(A) Reverse Current: 350u
@V(R) (V)(Test Condition): 6.4k
Maximum Operating Temp (шC): 100’
Package Style: Tube
Mounting Style: T


So, a 3B28 drops ~ 10v at 2 amps (?)
So a 866 drops ~ 15v at 2 amps (?)
and a 1N6237 drops ~ 28v at 250ma

I'm confused.  

Jim
WD5JKO

[WQ9E]

Jim,

They must have used a lot of individual diode junctions to reach the 10 Kv PIV rating.  Maybe they also used a small amount of series resistance to protect the diodes against high peak current.

I just received a couple NIB last week from ebay.  I will measure the parameters later this week for curiosity. 

[ke7trp]

Some testing of 5 volt rectifiers and voltage drops.

http://www.guitaramplifierblueprinting.com/rectifiers.html

C

[The Slab Bacon]

But........................... those are all straight vacuum rectifiers. The %u4 is good for a little higher current, where the 5R4 is good for a little lower current but higher voltage.

they didnt have the voltage drop / current ratings for mercury vapor (866, etc) or any of the gasseous type rectifiers (3B28, etc). If one looks them up you will find the figures are MUCH better. and that the voltage drop doesnt become real apparent until you get up there in the current range.

I played around one day with an 872 down low at 12v (Ep), it actually worked quite well down there. Go Figger................................ The reason that 866s were used in Valiants is for their current handling capability and NOT their voltage capability. Just about any rectifier will work at 650-700v.

[ke7trp]

The trouble with reading the data sheets is that the voltage drop is not tested under load.  As the load is increased the voltage drop is higher and higher.  I could not answer the reason why the diodes have a much higher plate voltage then the diodes so we tested them all again. 

The internal resistance is also another factor. The diodes drop are not changing under load.  The Voltage Drop of the data sheets where not spot on either. For example, The diodes had less then listed and the 866s had more then listed. 

The Carrier power is not changing by a huge factor with this swap as listed. However the modulated output really jumps up. Those who have tested these and not found much increase in power need to go back and test modulated output using a scope.  You can clearly see a big gain in output.

C

[WD5JKO]

I found some good references concerning gasious rectifiers:

www.electricstuff.co.uk/oldbooks.html

Here is an image from chapter six, see attached.

Notice the ignition voltage, the arc voltage, and the conducting time T1.

After zero crossing we have to get to the ignition voltage before we conduct, then we have a constant drop (about 15v for MV) regardless of current, and then when current drops too low the arc quenches. Solid state diodes do not have a ignition voltage, but do drop out when the P/N junction(s) are no longer forward biased.

The extra time (portion of a 1/2 wave cycle) it takes to ignite the MV is a disadvantage that SS does not have. Still, as the B+ voltage goes up above 1Kv, the difference will lessen.

Jim
WD5JKO