Elevated Heater & B+ Voltage

[Audiophile]

Hi Folks, Its been a while! I thought I had given up but Im back messing with valves for friends....
I was wondering if someone could look at my elevated heater circuit and just check Ive got it right, I dont even know if I need it, Ive read that this should be done for a CF and Im doing an SRPP but someone said I should elevate an SRPP so I thought I would ask here where the valve people live.

The other thing is my B+, the transformer is 180v, I was expecting about 254v DC but im measuring 279v un loaded. I dont really know what all the abbreviations mean so I dont even know if this is too much for the valve? I will attach a pic of its specs, looking at it through stupid eyes I see Ua 250v in characteristics so im thinking thats recommended? then I see Ua300v Max so im thinking Im ok but is there other factors/specs to consider?

[bikerhifinut]

That circuit should work, I'd be inclined to increase the potential dodvider resistors to something like 100k and 47k to reduce the current draw through which is really wasted.

The Ua which in British terms is Va is significant in as much as its the Voltage difference between the anode and the cathode, i.e. what the valve "sees", so your CF or Srpp circuit should be ok. Personally SRPP is a bit of a dead end but some like them.

As to your extra bit of HT, I'd measure the AC volts off the transformer as I bet thats a bit high possibly due to your local mains being a little bit higher than what the primary was wound for. The difference there wouldn't concern me. Likewise check the DC heater volts UNDER LOAD and adjust if needed with the resistor between your smoothing/reservoir caps. Not sure the 100n Bypasses will do much audible difference there, you'd be better using them on the HT methinks.
When the circuit is under load you'll likely find a voltage drop anyway.
If you need to adjust the HT volts down then looking at your PSU diagram you have scope to put extra resistors between the parallel 470uF capacitors. This will have the benefit of creating an RC filter which will reduce PSU ripple more.
those 470uF in parallel seem a bit of overkill to me, 1000uF of reservoir capacity is HUGE and in my opinion a waste as I doubt you''ll notice any difference at all if you used say a pair of 220uF in total. Here's a thing, if you had 4x 100uF instead of 4 x 470uF and put a 330 ohm resistor between each capacitor instead of a single 1k as you have there, you'd probably get better filtration than what you will with that circuit. I'll need to look it up again but theres a good explanation with the formula for a series of RC filters in Morgan Jones' valve Amplifiers book.

Andy.

[Audiophile]

Wow thanks Andy, you type quicker than me! Ok I've just realised I've drawn the circuit slightly wrong but I really appreciate the advice, I should of posted this before I purchased and built my supply. I did measure the transformer AC and I think it was pretty bang on 180v but that was about a year ago, the part I've got wrong is after the 1k resistor there is only one 470uf, I should of split the rail there, the second 470uf is for the other channel, I need to draw that out again. I will indeed do as you say on the voltage devider, my biggest concern was the high voltage of B+, I've done this a few times for people and the transformer has always given me 180v DC, I purchased some more from the same source only to find they had increased the voltage to 180AC, I sent an email which was greeted with tumble weeds.... Ok I'm taking all your info on board, I asked how to reduce the voltage on DIY but I was recommended to just buy higher voltage caps that cost more money and take up more room. Maybe I should post the diagram for the srpp as I thought the 1k resistor was kind of relative to the circuit.

Alan.

[Audiophile]

I should probably also say the circuit is an I/V stage for TDA1541, aka Lampizator, I know there are many alternatives but my friends seem to like it and I'm happy building it. I use a much smaller resistor for conversion, he uses 100r and I prefer 56.

[Audiophile]

Originally Posted by bikerhifinut

That circuit should work, I'd be inclined to increase the potential dodvider resistors to something like 100k and 47k to reduce the current draw through which is really wasted.

The Ua which in British terms is Va is significant in as much as its the Voltage difference between the anode and the cathode, i.e. what the valve "sees", so your CF or Srpp circuit should be ok. Personally SRPP is a bit of a dead end but some like them.

As to your extra bit of HT, I'd measure the AC volts off the transformer as I bet thats a bit high possibly due to your local mains being a little bit higher than what the primary was wound for. The difference there wouldn't concern me. Likewise check the DC heater volts UNDER LOAD and adjust if needed with the resistor between your smoothing/reservoir caps. Not sure the 100n Bypasses will do much audible difference there, you'd be better using them on the HT methinks.
When the circuit is under load you'll likely find a voltage drop anyway.
If you need to adjust the HT volts down then looking at your PSU diagram you have scope to put extra resistors between the parallel 470uF capacitors. This will have the benefit of creating an RC filter which will reduce PSU ripple more.
those 470uF in parallel seem a bit of overkill to me, 1000uF of reservoir capacity is HUGE and in my opinion a waste as I doubt you''ll notice any difference at all if you used say a pair of 220uF in total. Here's a thing, if you had 4x 100uF instead of 4 x 470uF and put a 330 ohm resistor between each capacitor instead of a single 1k as you have there, you'd probably get better filtration than what you will with that circuit. I'll need to look it up again but theres a good explanation with the formula for a series of RC filters in Morgan Jones' valve Amplifiers book.

Andy.

I think I must of drawn that circuit when I was drunk, I was copying it and I still got it wrong, I dont put bypass caps on the heater reservoir, I missed them off the anode supply completely. I did buy a fair few of the 470uf caps so I will use them for now, Then I will buy some smaller caps and just use one 470 per supply followed by 100uf caps. Are you saying that putting resistors between the caps will drop some volts? That would be good. So as I said there should only be 3 caps per valve and in the circuit the 1k resistor is followed by the last smoothing cap, can I therefore go CRCRCR? so that gives me 3x 330r but the last R is after a cap instead of before the last cap.

[Audiophile]

I was also wondering if one quarter watt is ok for the 330r's?

[bikerhifinut]

Ok, I realise you must be at the beginning of a learning curve here. We all had to start somewhere so no worries.
Can I politely suggest that you do a bit of research as well as taking any advice or information from us here? As when you understand what's happening it will all fall into place.
As to power ratings of resistors nobody can say what the power rating should be without sufficient information. In this case we need to know the current draw from the PSU. Then its good old Ohms law. V = IR and W = Vsqd/R or I sqd x R. those basic formulae are kind of essential to know. In any case err on the side of too much, its usually wise to use a resistor thats a lot bigger in power rating than what you calculate, bearing in mind its not just the steady state current draw it's passing but theres the ripple current when the reservoir/smoothing capacitors are charging up as the waveform rises. This can be surprisingly high.

Here's a couple of sites that might help:

http://www.angelfire.com/electronic/funwithtubes/

http://www.valvewizard.co.uk/

And if you can afford it and personally I wouldn't be without my copy, I'd go out and buy "Valve Amplifiers" by Morgan Jones. You'll learn a lot from his writings and it will help you when designing your own, and/or building others designs.
As to the RC filters, you should be now looking at CRCRC if using 3 x capacitors, therefore if 1k is the correct resistance to get the voltage drop to the right value then split them into two, which is 500 ohms. Go for the nearest preferred value which would be 2 x 470 or 2 x 510 or a combination of the two, it won't make that much difference.

I hope you aren't offended at the mini lecture.

Andy

[Audiophile]

Originally Posted by bikerhifinut

Ok, I realise you must be at the beginning of a learning curve here. We all had to start somewhere so no worries.
Can I politely suggest that you do a bit of research as well as taking any advice or information from us here? As when you understand what's happening it will all fall into place.
As to power ratings of resistors nobody can say what the power rating should be without sufficient information. In this case we need to know the current draw from the PSU. Then its good old Ohms law. V = IR and W = Vsqd/R or I sqd x R. those basic formulae are kind of essential to know. In any case err on the side of too much, its usually wise to use a resistor thats a lot bigger in power rating than what you calculate, bearing in mind its not just the steady state current draw it's passing but theres the ripple current when the reservoir/smoothing capacitors are charging up as the waveform rises. This can be surprisingly high.

Here's a couple of sites that might help:

http://www.angelfire.com/electronic/funwithtubes/

http://www.valvewizard.co.uk/

And if you can afford it and personally I wouldn't be without my copy, I'd go out and buy "Valve Amplifiers" by Morgan Jones. You'll learn a lot from his writings and it will help you when designing your own, and/or building others designs.
As to the RC filters, you should be now looking at CRCRC if using 3 x capacitors, therefore if 1k is the correct resistance to get the voltage drop to the right value then split them into two, which is 500 ohms. Go for the nearest preferred value which would be 2 x 470 or 2 x 510 or a combination of the two, it won't make that much difference.

I hope you aren't offended at the mini lecture.

Andy

Thanks Andy and no im not offended, thanks for the web links, ive bookmarked angelfire and ive spent hours on the valve wizard. I think there are probably some of us that fall into a category of never been able to understand how things work but with help from people like you we enjoy and do a bit of DIY, I have a few friends like me but they draw the line when they dont understand so I carry the flame and ive been winging it for about 30 years, ive had a blast, not to mention built a lot of stuff that sounds fantastic. Dont get me wrong I would love to understand how stuff works, I started a course on electronic repair but I had to quit due to my inabilty to understand maths. Im just ordering parts from RS now, I am getting some 100uf 450v as there much cheaper than anything lower voltage and the current draw is only about 2mA per tube so Im guessing .25W are ok. I will re do my power supply board. I am embarrassed to say I have only ever read one book in my 53 years, Grimms Fairy Tales.....

[Audiophile]

I dont think the 1k resistor is there to get the voltage drop to the right value, thats a very uneducated statement from me but the designer states any transformer between 100-200v AC and any .25w or higher resistor between 800 and 1.2k, I thought it may be a current limit resistor.

Alan

[bikerhifinut]

Originally Posted by Audiophile

I dont think the 1k resistor is there to get the voltage drop to the right value, thats a very uneducated statement from me but the designer states any transformer between 100-200v AC and any .25w or higher resistor between 800 and 1.2k, I thought it may be a current limit resistor.

Alan

It forms part of the RC filter, sometimes called a pi filter, that smooths the ripple on the rectified DC.
The current limiting aspect I suppose will be that it slows down current inrush into the second capacitor after the rectifier, this is a useful thing especially with Valve rectifiers that may only tolerate low values straight off their cathodes.
You can also use it to adjust the HT voltage so theres a double bang for your buck.
I was curious looking at that circuit and wondering why the designer didnt use a choke but of course thats a bit more expense. The resistor smoother works at the expense of losing a few more HT volts.
It sounds like that particular circuit, which I do not know, is very tolerant of variations in HT. I guess it's not much more than a buffer type stage?
I do not know what this lampizator is so I can't make any comments beyond that.
You really should try and get your head round the simple sums at the basic end. Ohms Law is hardly Advanced maths and its such a useful tool. I feel your groping around in the dark with no real understanding of whats going on.

Well with a couple of mA and lets assume 2mA, you can calculate the power dissipation through the resistor quite easily. Assuming you go for the 1.2k then power in watts (W) will be 2mA (.002) squared and multiplied by 1200 by the equation W = I^2xR (the ^2 is a way to wite the square function ^ means to the power of).
So, we have .002 x .002 x 1200 = .005 i.e 5mW so on the face of it 250mW resistors are fine. I'll still advise you use 2W though, and here's why.
1) a modern 1/4 watt resistor is bloody tiny, i bought some 2W metal films a couple of weeks back off RS and they were the same size as old 1/4 W resistors were.
2) smaller power rating resistors usually tend to have quite low voltage ratings, perhaps as low as 100V. Now this may not be an issue as we are concerned with the difference between one end of the resistor and the other ie the Voltage drop across it, the resistor doesn't really know anything else. BUT in a PSU especially, there will be a moment of time when the full voltage off the rectifier will be present across the resistor, until the valves conduct and while the capacitors charge up. So I err on the side of safety in these areas.
3) the calculation above is fine in terms of steady state DC conditions, but that initial resistor will also have a fairly substantial ripple current through it caused by the current drawn to recharge the reservoir and smoothing capacitors between cycles. I'd advise further reading as its quite an eye opener.
So in addition to mr Grimms tales, I think you should extend your library a bit, it doesnt have to be a paper book!
It's true that when you start to get a handle on what's going on in a circuit you get a real buzz out of knowing what you've built and also can start to spot the odd blooper that happens even with the best thought out designs.

A.