New design with 1KV power transformer

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
After a long pause, I have decided to build a new 50W guitar amp using some existing parts I have. The power transformer I have appears to be 500-0-500 (1 kilovolt!). I intend to use two 6550s in PP class AB1. Obviously, I dont intend to run these any higher than 600V B+. So I am not sure of the approach to take for the power supply design. One possibility is to use a regulated supply using a pass tube or two and simply dial in the voltage I need. I have built several such supplies based on designs originating from Stephie Bench's old web site, so I know how to do that. The problem is that, to bring the voltage down 3-400V with a current draw of 200mA means a monster amount of dissipation in the pass tube(s). Any suggestions on how to deal with this would be welcome. I could just use another transformer, but I want to see if I can solve this problem first.
 
Best to get a proper power transformer.

600v is more than the typical 6550 can handle, certainly the new production tubes. Even the older tubes will be sweating a lot.

Consider a pair of 807s or 1625s with that plate voltage - but then you will have trouble finding proper output transformers. :D

No free lunch.

_-_-
 
If you use a dual half-wave rectifier, you'll be able to significantly drop the voltage without excessive losses, and you will get the same benefits as a full wave choke input supply: low internal resistance and good noise rejection.

The main drawback is the necessity to use two separate chokes (and two additional diodes) but that is not an excessive price to be paid in regard of the advantages
 

Attachments

  • DualHalfWa.png
    DualHalfWa.png
    122.6 KB · Views: 237
If you use a dual half-wave rectifier, you'll be able to significantly drop the voltage without excessive losses, and you will get the same benefits as a full wave choke input supply: low internal resistance and good noise rejection.

The main drawback is the necessity to use two separate chokes (and two additional diodes) but that is not an excessive price to be paid in regard of the advantages

You made that circuit up, don't know where you pulled the
dual half-wave rectifier
label from, probably out of thin air.

What your drawing shows is a conventional full wave rectifier, with two useless diodes , D2 and D4 , and needlessly complicated by splitting the choke in two.

That something can be simulated does not mean it's "right" .
 
I have not seen that circuit before.
Neither have I :D
Am I right that it gives half the DC voltage of the conventional full-wave choke input PSU?
Yes

You made that circuit up, don't know where you pulled the label from, probably out of thin air.
Yes and no: the label is just a factual description of the functions encompassed by the circuit

What your drawing shows is a conventional full wave rectifier, with two useless diodes , D2 and D4 , and needlessly complicated by splitting the choke in two.
Try to think a little further

That something can be simulated does not mean it's "right" .
Of course not, but in my sims I take great care to stick to reality and if you built this circuit as it is drawn (with higher PIV diodes), it would behave as a dead ringer for the sim
 
Quote:
What your drawing shows is a conventional full wave rectifier, with two useless diodes , D2 and D4 , and needlessly complicated by splitting the choke in two.

Try to think a little further

I have, don't think I didn't. ;)
It took me about 1 minute, fair enough, because I gave you the benefit of the doubt so I double checked.:D

Let's analyze what's happening there:

a) start with the upper secondary winding, called L3 by you.
When it's top end is positive respect to the lower one, D1 is forward biased and passes current into the load, meaning it works as a rectifier, it does a useful job.
D2 is reverse biased, it passes no current , it's not rectifying.

b) now on the next hemi-cycle: voltage across L3 reverses, D1 becomes reverse biased, no current passes through L3 winding, no current passes through D2, it's not rectifying.

If D2 never passes current and even more important, does not rectify (which is its function in a PSU rectifier) it qualifies as useless in my book (and in anybody else's :p )

Same analysis applies to the lower winding, L4.
D4 is also useless.

c) now to the dual choke you suggest:

what you should do is to join D1 and D3 cathodes (what 99.99999999% of people does, by the way) and then filter that 100/120Hz full wave with a single choke.

What you suggest, using two chokes, which to make matters worse will have to deal with 50/60Hz halfwaves (MUCH harder to filter) is nonsense, both economically and technically.
 

Attachments

  • Elvee-NonRectifiers.png
    Elvee-NonRectifiers.png
    122.6 KB · Views: 93
Last edited:
Let's analyze what's happening there:

a) start with the upper secondary winding, called L3 by you.
When it's top end is positive respect to the lower one, D1 is forward biased and passes current into the load, meaning it works as a rectifier, it does a useful job.
OK that far
D2 is reverse biased, it passes no current , it's not rectifying.

b) now on the next hemi-cycle: voltage across L3 reverses, D1 becomes reverse biased, no current passes through L3 winding,

no current passes through D2, it's not rectifying.
Things begin to go astray:
The current through D2/L3 continues to flow and has to loop back somewhere. It does so through the load, and thus participates to the output. The energy stored in L3 is transferred to the load, exactly like in a buck converter

If D2 never passes current and even more important, does not rectify (which is its function in a PSU rectifier) it qualifies as useless in my book (and in anybody else's :p )
Without D2, the circuit would work, but very poorly: the output voltage would vary from 700V at zero-load to zero at a high load

Same analysis applies to the lower winding, L4.
D4 is also useless.

c) now to the dual choke you suggest:

what you should do is to join D1 and D3 cathodes (what 99.99999999% of people does, by the way) and then filter that 100/120Hz full wave with a single choke.
This becomes the classical circuit, having an output voltage of a little under 500V, which is too much in this case
What you suggest, using two chokes, which to make matters worse will have to deal with 50/60Hz halfwaves (MUCH harder to filter) is nonsense, both economically and technically.
Although the circuit is composed of two half-wave rectifiers joined together, the output will see a 100/120Hz ripple, the transformer will see no unbalance, and the power factor will be ~1

Think about inductors. They are not resistors. They can, temporarily, conduct current in the opposite direction to the applied voltage. That is how the conventional full-wave choke input PSU works. That is how Elvee's modified version works too - unless I have missed something. The apparently useless extra diodes do have a function.
Exactly: if the inductors are large enough to ensure operation in continuous mode, the output voltage will be √2Vac
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.