How to 'properly' reduce B+????

[Wharfcreek]

This is a subject that I've often wondered about as it relates to many amps and the use of certain output tubes in lieu of 'original' tubes......such as using 6BQ5s for 7189s, or 6V6 for 7408, etc. The issue being that the rated voltage of the 'original' tube is higher than the available replacements. I believe there is a Russian sub for the 7408, but I'm pretty sure the the EL84M or 6P14P-EB is a good sub for the 7189. However, what if you just want to 'reduce' that B+ voltage to something more to your own liking.....for purposes of either matching an original spec.....or to not overload a tube.....or what ever? In my case, I just completed my first 'scratch' build.....and finally got the amp up and running. However, now in the 'tweaking' stage, and I'm noting that my B+ voltage is just enough above spec that it has me worried. Spec calls for 370, but I'm running at 405.

I placed a Current Inrush Limiter in series with one leg of the PT primary....and that dropped me to about 395....or about a 10 volt reduction. Plus, I got the added benefit of an increase in warm-up time. Not really sure if this is a benefit, as the delay in Secondary HV has got to be correspondingly delaying the filament side as well. So, while not B+ surge initially.....also no filament voltage. Since the inrush limiter is controlling the primary and thus equally affecting ALL the secondaries.....I'm of the opinion that any perceived favorable affect of any time delay from this device is negated by the equal delay in tube heater warm up time.

But, back to the point: Aside from utilizing the CIL device, I have also attempted use of a power resistor off the rectifier to the first stage of the PS filter.....which incidentally is also where the OPT and Screens are attached. I then have 2 more stages of filtering followed by a fourth and final stage where the driver voltage is taken from. So, any reduction at the front of the filters as seen by a resistor TO the first stage of the filter would have to be compensated for at the fourth stage.....or so I figure. I've been advised NOT to use the power resistor at this stage....and to just feed the rectifier directly to the first stage. This way I'm not restricting voltage to the filter caps or the power output stage of the amplifier.....and the amp will sound better! I'd love to hear some opinions about that.

Just for the record....with 150 ohms of resistance (using a 7 watt power resistor) I had a B+ of about 375.....damn near right on the money! But, again, I was advised against doing this.....so I pulled it and now have the current inrush limiter in line. I'm wondering....without attempting to build a regulator stage more complicated than the amp itself.....any other ways to do this......or comments about which way is better?.....or doing 'both'??

Many thanks, Tom D.

 

[Tom Bavis]

A resistor has a similar effect whether it's in the primary or secondary. Putting one between the rectifier and first cap will take some "cut and try", since it carries both the DC current and the capacitor ripple current. So it will drop more voltage than you expect...

Inrush limiters can be a good thing with today's higher line voltage - it was designed for 117V and is getting more like 125 (only 122 at my house ). Use two if one doesn't drop it enough. Check heater voltage for a second opinion - bet it's high too.

But 405V is OK for the 7591 - just make sure that dissipation (plate voltage x idle current) is within ratings. Only 10% high - will drop if you raise the idle current.

A class AB amp will have some voltage sag in the supply as power increases, since current increases. Any series resistance will increase this effect. Rate of drop will depend on size of filter caps. That's why solid state rectifiers can give better bass - bigger filter caps can be used.

 

[sony6060]

The Fisher TA 600 uses a 150 ohm resistor in-between the tube rectifier and first capacitor being 40uF.

 

[Wharfcreek]

Tom,.... got it! Thanks! Sony.... the schemo I build from called for direct connect between the rectifier (5U4) and the first stage cap..being an 80uf/450V. But, as I said, the spec design also called for 370 volts at that point too. Probably as Tom stated, my house AC may be a bit high.....didn't measure that, but last time I did I think I was at 118...but I have a LOT of 'fluctuation' flickers in my lights at night...no doubt due to being in a VERY old area with VERY VERY old AC service. Like I said, when I did this straight through (using the PT as provided from the donor amp that applied to the schematic)...I got the 405 volts. BUT, as with the TA600 that Sony mentioned, I had a 150 ohm/7 watt resistor in between the rectifier and the first stage, and at that point I had nearly a perfect 375. That's where I got told 'Don't use a resistor there.....it won't let the amp sound as good!' I might also point out that I also have switched from the 5U4 to a 5AR4 ..... So, that's where I am. I may go back to the resistor...but I may add a second CIL to the PT primary side. Or both? TSD

 

[gadget73]

My Stromberg-Carlson PA amps use I think a 200 ohm resistor between the output of the rectifier and the first filter cap. Eventually I'd like to toss that in favor of a choke, probably with a small cap right at the tube to dial in the voltage exactly where it ought to be. If you have the space for it, that will give you a stiffer and better filtered supply, but not all amps have room for a big fat choke.

What you get with a big resistor between the rectifier and the first filter is a lower current capacity power supply. Max current capacity in the supply will be limited by the dropping resistor. If its not being pushed to the limits, it probably won't matter. If the supply is a wee bit marginal, you'll get voltage sag that can make for mushy bass. If it extreme enough, you can hook an analog voltmeter to the B+ supply and watch it dance with the music.

 

[analog addict]

Here's a useful chart...

http://www.300guitars.com/articles/rectifier-tube-voltage-drop-chart/

If you can safely run a 5AR4, switching back to a 5U4GB will drop your voltage an additional 30-40 volts. You can drop the B+ further by dropping in a 5R4. You could even consider a choke input power supply, although I think you're close enough that a rectifier switch will do the job. Also, make sure you look at the specs for the maximum filtering cap value. For a 5U4GB, it's listed as 40uF. Some people play a little fast and loose with this value, but since I barely know what I'm doing, I like to follow the prescribed values. A 5AR4 is listed as 60uF.

Personally I would do what it takes to get rid of that resistor between the rectifier and the first filtering cap. Remember that it will dissipate the voltage as heat, and heat is the enemy of any tube amp, especially under the hood.

 

[kward]

Eventually I'd like to toss that in favor of a choke, probably with a small cap right at the tube to dial in the voltage exactly where it ought to be.

Bingo. I think that's the best way to do it. However a smaller cap in that first position, even with a choke after it, will give a little more ripple than a larger cap in that same position. But the ability to dial in the voltage you want based on the size of the cap, I feel is a really good trade off.

For example (picking some arbitrary values for the sake of example): a 600V center tapped transformer with full wave 5AR4 rectification, using a CLC filter will give the following:

C1=1 uF, L=5H, C2=100 uF: Voltage = 260VRMS, ripple about 1/2 volt

but changing C1 to 32 uF and keeping everything else the same:

C1=32 uF, L=5H, C2=100 uF: Voltage = 375VRMS, ripple about 25 mV

So you do have quite the ability to "tune" the output voltage based on the size of that first cap. Just need to be watchful if the resultant ripple is something you can live with.

 

[SoundOfSound]

 

[nerdorama]

This page has an interesting solution. Scroll to the bottom for the B+ reducer. The neat thing about this solution is you can dial in the exact voltage drop you want, within reason and device dissipation, but it's a bit more complicated than a resistor.

http://www.geofex.com/Article_Folders/mosfet_folly/mosfetfolly.htm

John

 

[jaymanaa]

For some really hot amps, I've used a thermistor in both legs of the power cord wiring, and even one from the CT of the PT secondary to ground. I prefer them over resistors for the added benefit of slow start. If you size the one on your secondary CT to ground correctly (amps wise), you also get some protection for you PT. On a typical 6L6 or EL34 type amp, 1/4 watt is about right for the secondary. Of course like everything, you have to look at the spec sheets and know the passive part will do what is expected of it. Oh, having one in the secondary (high voltage) circuit can also make sure that any bias supply is up to voltage before the power tubes get high voltage, another plus.

 

[Wharfcreek]

This is all great stuff. As I embark on my evolution from 'repair/restore' to actual construction, knowing these kinds of things will be EXTREMELY helpful! So, thanks to all. I made an unfortunate mistake in my original amp layout....in that I put the socket for the tube rectifier just too damn close to the PT core plates. So, with a 5U4 plugged in, it damn near touches the PT. With a 5AR4 plugged in I get about 1/8 of space between the bottle and the transformer......at least enough for some small margin of airflow. So, I think I'm going to stick with the 5AR4 unless I can find a 5R4 that will also give me sufficient space. I'd move the socket....but I'd now have interference problems underneath with the filter cap layout. I put this all into the category of 'things you learn when you first start out......'. I'll not make this mistake again in the future.

BTW, regarding cap values..... I find it interesting that the first stage of the PS filter as per the schematic (and on the original amps) is an 80 uf from the 5U4. As Analog Addict points out....the 'spec' on that tube calls for a max of only 40 uf..... so how this has been gotten away with is beyond me...... unless it also incorporates some kind of additional consideration for the fact that both the OPT Center Taps AND the Screens to all 4 output tubes are also attached at this point as well. Screens go thru a 100 ohm resistor.....and one of the two OPT Plate connections goes thru a 10 ohm resistor. But, other than that...the all go straight to the first PS Cap and Rectifier output.

Keep it all coming! I'm learning a lot! Thanks guys!

Tom D.

 

[ferninando]

this thread goes in my "tube stuff" folder for reference.
AK moderator should make this a sticky. Much needed information.
Please.

 

[gadget73]

Bingo. I think that's the best way to do it. However a smaller cap in that first position, even with a choke after it, will give a little more ripple than a larger cap in that same position. But the ability to dial in the voltage you want based on the size of the cap, I feel is a really good trade off.

For example (picking some arbitrary values for the sake of example): a 600V center tapped transformer with full wave 5AR4 rectification, using a CLC filter will give the following:

C1=1 uF, L=5H, C2=100 uF: Voltage = 260VRMS, ripple about 1/2 volt

but changing C1 to 32 uF and keeping everything else the same:

C1=32 uF, L=5H, C2=100 uF: Voltage = 375VRMS, ripple about 25 mV

So you do have quite the ability to "tune" the output voltage based on the size of that first cap. Just need to be watchful if the resultant ripple is something you can live with.

Yeah, its something I need to fiddle with. Right now it runs the 6L6's at fairly low voltages. 330v on the plate if I remember right. I could probably even get by with simply eliminating the resistor entirely, running the output tubes a little hotter, and adding more resistance to drop the voltage on the preamp stuff.

 

[kward]

This page has an interesting solution. Scroll to the bottom for the B+ reducer.

The topology looks sort of like a series-pass regulator. Putting it between the center tap and ground keeps the working voltages down to the amount you're trying to reduce. That's clever.

By the way, I'm not that familiar with bucking transformers but I wonder if one could be used here as yet another way to reduce voltage? I know they are traditionally used on primaries, but could you configure one to be wired to the secondary to reduce secondary voltage a bit? Might be too expensive to have this realistically be a viable solution.

If using a toroid, it becomes a bit easier by simply extending both secondary legs with some high voltage hookup wire and wind each leg tightly around the outside of the toroid in reverse direction of the original wind. The amount of voltage reduction would depend on the number of reverse turns added. I wouldn't probably do this unless in a pinch, and unless the toroid were under the chassis or covered, since exposing wires this way above chassis seems like a bad idea.

 

[Wharfcreek]

I need to address some further knowledge of 'chokes' as well. From what I can tell, they should almost always be used if you want to have a really rock-solid power supply. I understand that 'room' can be a problem, as well as trying to keep costs down during initial production. But, finding one today isn't always the easiest thing to do either! I don't see a lot of chokes coming up for sale. About the only one I see consistently is the one that Dynaco used....and is now re-produced. I actually thought about using that one in my 'build' project. And.... I may yet do so. Adding it to the top of the chassis wouldn't be a problem at all!! TSD

 

[wa2ise]

Personally I would do what it takes to get rid of that resistor between the rectifier and the first filtering cap. Remember that it will dissipate the voltage as heat, and heat is the enemy of any tube amp, especially under the hood.

That resistor will make life easier on the vacuum tube rectifier. I assume you have the resistor between the rectifier cathode/filament and the first filter cap. And you can make that first filter cap bigger. The resistor will make the peak cap recharging (this is the current that tops off the cap at every peak of the full wave 60Hz waveform) current lower.

 

[Wharfcreek]

Just a quick 'second thought' about chokes: If the 'resistance' of a choke is ie 75 ohms, is that not still the equivalent of having a 75 ohm resistor in the same position? I understand that the inductive aspect of the choke will improve ripple reduction. But, I guess I'm just trying to answer the question of: When is 75 ohms NOT 75 ohms? If 'never'....then that's good to know. But, if the 75 ohms of a choke isn't seen the same way in the end, that's good to know too.

 

[sony6060]

Just a quick 'second thought' about chokes: If the 'resistance' of a choke is ie 75 ohms, is that not still the equivalent of having a 75 ohm resistor in the same position? I understand that the inductive aspect of the choke will improve ripple reduction. But, I guess I'm just trying to answer the question of: When is 75 ohms NOT 75 ohms? If 'never'....then that's good to know. But, if the 75 ohms of a choke isn't seen the same way in the end, that's good to know too.

75 ohm choke still drops voltage like a resistor. The difference is the choke stores energy and will regulate better.

The reason some say a resistor in-between the rectifier and first capacitor sounds poor is that it adds resistance slowing down voltage recovery from a large short term current draw (regulation). Perhaps something to look at with large single-ended amps, but PP amps have a more constant current draw at moderate volumes vs a single-ended amp.

Rectifier tubes also have an internal voltage drop. GZ34 about 20 volts, 5U4G or GB about 40 volts and 5R4GYB about 60 volts. The rectifier tube is like a series resistor too.

 

[Wharfcreek]

Sony, that makes perfect sense.....and I appreciate the additional info. That's really what I'm looking for.....to be able to understand the 'why?' of the question.....not just the answer to the question. Taking your answer a step further, I'm assuming that since the choke is, in a sense, a 'dynamic' device..... (referring to the electrical movement of the field generated as the current passing through the choke rises and falls thus creating an on-going field-effect within the choke.....) I'm 'guessing' that the effect is that depleted current from the demand of the amp may be somewhat compensated for by the collapsing field within the choke and the developed current as a result? Not that I'm looking for an answer like a physics professor might give, but if that's the basic 'gist' of the thing....then I guess I 'get it'!! Thanks! Tom

 

[derekva]

Not turning in your homework or studying for tests always worked for me.

-D