Greetings,
I'm trying to model in PSUD a PS shown in section 1 in the diagram below. Unfortunately, PSUD lacks the ability to model a "hybrid rectifier", so my first model was a tube-only bridge shown in section 2. The problem is that in this model I can never achieve required 150V at the PS output.
The model in section 3 (full-wave) produces a perfect 150.72V. My guess is that the difference is because four tubes will drop more voltage than two.
I have few questions to ask regarding these models:
1. Is it OK to use full-wave model as a base for calculations even though the real circuit is going to be a bridge? Here the implicit assumption is that the two SS diodes (UF5408) have negligibly low resistance.
2. If I increase C2 to, let's say, 220uF PSUD complains that "The rectifier IFRM (current forward repetitive maximum) of 0.53A has been exceeded with a value of 0.67A, at time 0.52S." Is IFRM value the same parameter as "Peak Plate Current Per Plate" in the 5V4 datasheet?
3. I was advised in the other thread on this forum that a choke limits the current surge and a very large cap may be used after it. Is PSUD accounting for the current-limiting effect of the choke? May I dismiss the PSUD warning or is it serious?
4. May I use two 5V4s in parallel if I want to use large C2 and C3? I tried to model with 5U4, no current warnings, but the voltage drop was higher and I could not arrive at 150V output.
5. The last one - worries me a lot: in the operating characteristics chart in 5V4 datasheet I can't see any curves below 300 RMS volts per plate. My transformer I guess is producing only 175 x 1.4 = 245V. So can I use this rectifier at all ?!
Thanks for all your help!
I'm trying to model in PSUD a PS shown in section 1 in the diagram below. Unfortunately, PSUD lacks the ability to model a "hybrid rectifier", so my first model was a tube-only bridge shown in section 2. The problem is that in this model I can never achieve required 150V at the PS output.
The model in section 3 (full-wave) produces a perfect 150.72V. My guess is that the difference is because four tubes will drop more voltage than two.
I have few questions to ask regarding these models:
1. Is it OK to use full-wave model as a base for calculations even though the real circuit is going to be a bridge? Here the implicit assumption is that the two SS diodes (UF5408) have negligibly low resistance.
2. If I increase C2 to, let's say, 220uF PSUD complains that "The rectifier IFRM (current forward repetitive maximum) of 0.53A has been exceeded with a value of 0.67A, at time 0.52S." Is IFRM value the same parameter as "Peak Plate Current Per Plate" in the 5V4 datasheet?
3. I was advised in the other thread on this forum that a choke limits the current surge and a very large cap may be used after it. Is PSUD accounting for the current-limiting effect of the choke? May I dismiss the PSUD warning or is it serious?
4. May I use two 5V4s in parallel if I want to use large C2 and C3? I tried to model with 5U4, no current warnings, but the voltage drop was higher and I could not arrive at 150V output.
5. The last one - worries me a lot: in the operating characteristics chart in 5V4 datasheet I can't see any curves below 300 RMS volts per plate. My transformer I guess is producing only 175 x 1.4 = 245V. So can I use this rectifier at all ?!
Thanks for all your help!
Attachments
Q1: I would expect your section 3 model to be reasonably accurate, barring the slight additional voltage drop due to the silicon diodes.
Q2: Yes, you do need to worry about IFRM. However, valves warm up slowly. Try using the ramp function in PSUD.
Q3: PSUD seems to be a pretty good model. If it warns you of over-current then you need to be a good engineer to ignore that warning. Conversely, if it says you are fine, then you probably are. And yes, it accounts for chokes.
Q4: You could, but it's probably not a good idea. A beefier valve would be better.
Q5: If you really wanted, you could use your 5V4 to power 15V op-amps. It would be rather inefficient, but it could be done. Typically, valve rectifiers drop between 10 and 30V across themselves (rather than 1V for silicon).
Q2: Yes, you do need to worry about IFRM. However, valves warm up slowly. Try using the ramp function in PSUD.
Q3: PSUD seems to be a pretty good model. If it warns you of over-current then you need to be a good engineer to ignore that warning. Conversely, if it says you are fine, then you probably are. And yes, it accounts for chokes.
Q4: You could, but it's probably not a good idea. A beefier valve would be better.
Q5: If you really wanted, you could use your 5V4 to power 15V op-amps. It would be rather inefficient, but it could be done. Typically, valve rectifiers drop between 10 and 30V across themselves (rather than 1V for silicon).
Thanks for the replies.
I did after you asked - it drops the voltage at the output to around 100V - the difference between choke-input and capacitor-input filters as expected, I guess.
I tried and PSUD did not complain about over-current! But here I'd like to make sure that I'm not fooling myself:
1. I had "Soft Start" option enabled in the previous simulations that produced over-current with big caps. Isn't it the function to imitate the tube warm-up?
2. In the stepped load function ("ramp-up") I've given an initial load of 20mA and full load of 160mA after 1 sec (I was always hitting the over-current in the first 0.5 sec). Obviously, these numbers are a total guesstimation. What are the better ways to specify the stepped load?
3. In the simulations it was apparent thet over-current is correlated with the value of the caps - my understanding is the bigger the cap the more current it sucks on the initial charge. On the other hand, lowering the load helped to rectify
the situation. What is the interplay of things here - does it mean that lower load influences caps charge? If not, I'm confused 
4. PSUD doen't show the switch from the initial 20mA to 160mA (full load) and the resulting voltage that I get is ~135V. From the previous simulations I know that the I get the required 150V at full load so I don't have to worry, do I?
Thanks!
I did after you asked - it drops the voltage at the output to around 100V - the difference between choke-input and capacitor-input filters as expected, I guess.
I tried and PSUD did not complain about over-current! But here I'd like to make sure that I'm not fooling myself:
1. I had "Soft Start" option enabled in the previous simulations that produced over-current with big caps. Isn't it the function to imitate the tube warm-up?
2. In the stepped load function ("ramp-up") I've given an initial load of 20mA and full load of 160mA after 1 sec (I was always hitting the over-current in the first 0.5 sec). Obviously, these numbers are a total guesstimation. What are the better ways to specify the stepped load?
3. In the simulations it was apparent thet over-current is correlated with the value of the caps - my understanding is the bigger the cap the more current it sucks on the initial charge. On the other hand, lowering the load helped to rectify
the situation. What is the interplay of things here - does it mean that lower load influences caps charge? If not, I'm confused 4. PSUD doen't show the switch from the initial 20mA to 160mA (full load) and the resulting voltage that I get is ~135V. From the previous simulations I know that the I get the required 150V at full load so I don't have to worry, do I?
Thanks!
Bridge rectifiers 5V4
Your third example will have less losses than first by one diode tube drop. But you have made an assumption that the DC resistance of your centre tapped transformer is the same as your full wave bridge version. Typically each halve will have twice the secondary DC losses thus limiting your peak current. You have to transform your primary back to the secondary to get a true DC loss for your equation. However a centre tapped secondary winding with 2 tube diodes is still the prefered method.
Your figure of 220 Ohms for you 10 H choke seems very high. I can only assume its fairly small physically.A large choke weighing about 8 or 10 lbs would be in the order of 10 to 20 Ohms.
Finnally, why large caps. the choke does the job so much better.
If you use a properly designed input choke then the rectified DC is nearly equal to the RMS AC input.so 150 volts in to about 130 out with the advantage that the rectifier sees a near RMS current.
No nasty peaks. Final ripples of sub 1 mV easily acheived.
Your third example will have less losses than first by one diode tube drop. But you have made an assumption that the DC resistance of your centre tapped transformer is the same as your full wave bridge version. Typically each halve will have twice the secondary DC losses thus limiting your peak current. You have to transform your primary back to the secondary to get a true DC loss for your equation. However a centre tapped secondary winding with 2 tube diodes is still the prefered method.
Your figure of 220 Ohms for you 10 H choke seems very high. I can only assume its fairly small physically.A large choke weighing about 8 or 10 lbs would be in the order of 10 to 20 Ohms.
Finnally, why large caps. the choke does the job so much better.
If you use a properly designed input choke then the rectified DC is nearly equal to the RMS AC input.so 150 volts in to about 130 out with the advantage that the rectifier sees a near RMS current.
No nasty peaks. Final ripples of sub 1 mV easily acheived.
You've got the load current plus the cap charging current. Presumably the sum of the 2 takes you over th limit with the higher load.
You should be able to see the switch from 20mA to 160mA in PSUD. Are you running the simulation for long enough. i.e. beyond the time when the switch occurs.
Larger caps reduce the Q of the resonant circuit formed by the choke and the cap. With a smaller cap the power supply will ring when the current draw changes. This is easily seen with the step function in PSUD. (If you get it to work properly see above)
Rectifier Choke resonance?
If you want that argument then make the capacitor infinitely large.
As I said a choke input filter since it is carying the rectified AC component passes a very near sinusoidal current into the first cap so you do not have the usual saw tooth waveform on the first cap and the ripple is significantly lower.
Secondly, the choke 'swings' or changes value with load change to provide very good regulation.
If you want that argument then make the capacitor infinitely large.
As I said a choke input filter since it is carying the rectified AC component passes a very near sinusoidal current into the first cap so you do not have the usual saw tooth waveform on the first cap and the ripple is significantly lower.
Secondly, the choke 'swings' or changes value with load change to provide very good regulation.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.