Multi-voltage PS

dgta · 16th July 2011, 07:49 PM

I picked up an obscure brand old lab PS, mainly for its filament supply. But it has an interesting high voltage section also, see schematic.

Why would they do half-wave only? Doesn't seem like a cheaply made supply. Anyway, it would be nice if I could use it for 2 voltages at once, e.g. +375 and -125. Can't do that the way it's set up.

Is there anything wrong with simply putting a bridge rectifier on each winding section (and filtering) and having the 4 outputs in series? Trying to figure out why they didn't do something as simple as that.

! · 17th July 2011, 01:00 AM

You can do +375 and -125V.

1) Don't connect the center tap to ground. Make sure transformer casing isn't grounded too.
2) New ground is -125V lead.
3) -125V comes from -125V lead as ground and -250V lead.
4) Make sure capacitors and diodes are rated for the higher voltage.

dgta · 17th July 2011, 01:28 AM

That wasn't the point. There were 2 points to my post: 1. The half wave rectification and 2. Flexibility to use as many combinations of voltages as possible.

! · 17th July 2011, 05:14 AM

Half wave is just cheaper, can be air wired. Some say the heat/loss reduction helps, it depends on the goal of the supply whether it matters but yes you can do full bridge rect.

If you want the greatest flexibility of voltages with you will need to use switching to control which winding leg is the negative return from a full bridge rectifier, or vice versa, have the -250V be the fixed ground and switch which of the remaining 4 is positive to the rectificaton stage... leaving the lead marked as ground in the schematic, floating... and as already mentioned you will need to check the diode and capacitor ratings to be sure they will work.

dgta · 17th July 2011, 05:39 PM

No, if I do the simple circuit below, I can use all voltages and all the caps can be 125V. 1000V bridges are real cheap.

! · 17th July 2011, 08:38 PM

The way you have the circuit you could use any ONE adjacent pair of windings (taps in them) for 125V difference but any time you try to use more than 125V, your return is through more than one diode bridge.

No you cannot "necessarily" use 125V capacitors. 0V would not always be ground in all possible voltage configurations, your negative rail would be unless you are using a split supply... you did not clearly state EXACTLY what you are trying to do which makes a difference.

If you use more than 125V you need to use capacitors rated higher than 125V, they need to be rated for the voltage you are using and connected between the two taps after the respective rectifier bridges.

Everyone is entitled to disagree but you seem prone to a strange arguing stance when you don't seem to know. The easiest way to do the circuit is have a floating transformer ground lead, put the -250V lead as the new negative return, use a single bridge rectifier and a single throw 4 pole switch, with each of the other taps as a successive pole on the switch which alternately connects to a single bridge rectifier, the positive output of which is connected to a single capacitor rated for > 500V. That's 500V loaded, you will have to measure peak voltage unloaded to get the actual peak value needed as it could be significantly higher than your diagram shows.

dgta · 18th July 2011, 05:04 AM

!, I don't think you understood what I'm saying, so here is the explicit diagram. There is no "ground", as my previous (bridge) diagram doesn't show one. I purposely removed the ground as there is no need to ground any point on the xfmr secondary. As mentioned, this is a bench supply and I can connect the load(s) any way I choose.

The second row of numbers shows one possible use for an amp that requires 375 B+ and -125 for bias.

DF96 · 18th July 2011, 11:39 AM

No, that won't work. You can either wire the secondaries in series or the rectifier output in series. You can't do both. If your transformer already has the windings connected together then that is what you will have to use. You could use two bridges, one on the 125-0-125 and the other on 250-0-250, to get two bipolar DC supplies - about +- 170V and +- 340V. You don't have to ground the '0V' on the output, so you could have -170, 0, +170, +340, +510V by grounding the '-170V' output.

! · 08:53 PM

^^ and, the capacitor between two windings will be sufficient for the -125V, but you're reducing the effective capacitance wiring those in series for voltages higher than +-125V. Capacitance = 1/[(1/C1) + (1/C2) + (1/C3)]

To be fair, if you kept the circuit the way it was and it worked, then you could use 125V capacitors except as mentioned previously, if the transformer is rated 125V per winding, that is only at the load rating of the transformer. When there is lesser load the voltage would be approaching 125V *1.41 - (2*0.7) = 175V.

On the other hand if it is not the transformer itself that is rated for 125V per winding but instead the power supply "product" it was in was rated for 125VDC output (which with an unregulated design would be strange as the voltage would still fluctuate based on load) then you might not need to factor for the 175V peak voltage per transformer tap.

Either way, it seems a very inefficient use of capacitors instead of the normal way of having one, no series, between positive and negative of the transformer windings powering the load.

DF96 · 19th July 2011, 07:23 AM

The ratio between 125V and 175V is not a matter of transformer load, but RMS vs peak.

18th July 2011, 08:46 PM	#9
! diyAudio Member Join Date: Jun 2005 Location: Midwest	^^ and, the capacitor between two windings will be sufficient for the -125V, but you're reducing the effective capacitance wiring those in series for voltages higher than +-125V. Capacitance = 1/[(1/C1) + (1/C2) + (1/C3)] To be fair, if you kept the circuit the way it was and it worked, then you could use 125V capacitors except as mentioned previously, if the transformer is rated 125V per winding, that is only at the load rating of the transformer. When there is lesser load the voltage would be approaching 125V 1.41 - (20.7) = 175V. On the other hand if it is not the transformer itself that is rated for 125V per winding but instead the power supply "product" it was in was rated for 125VDC output (which with an unregulated design would be strange as the voltage would still fluctuate based on load) then you might not need to factor for the 175V peak voltage per transformer tap. Either way, it seems a very inefficient use of capacitors instead of the normal way of having one, no series, between positive and negative of the transformer windings powering the load. Last edited by !; 18th July 2011 at 08:53 PM.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Phillps CD-840, multi voltage tranformer needed	leesonic	Digital Source	2	10th February 2011 09:06 PM
PS voltage without load	duke_tj	Power Supplies	5	16th September 2007 06:27 AM
multi-channel hum, one transf, multiple PS	harvardian	Chip Amps	1	12th December 2006 04:32 PM
Uneven voltage on PS	chipco3434	Power Supplies	11	22nd February 2006 02:10 PM
higher voltage PS	Steven Koay	Chip Amps	5	10th September 2003 11:29 AM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

17th July 2011, 01:00 AM	#2
! diyAudio Member Join Date: Jun 2005 Location: Midwest	You can do +375 and -125V. 1) Don't connect the center tap to ground. Make sure transformer casing isn't grounded too. 2) New ground is -125V lead. 3) -125V comes from -125V lead as ground and -250V lead. 4) Make sure capacitors and diodes are rated for the higher voltage.

17th July 2011, 01:28 AM	#3
dgta diyAudio Member Join Date: May 2011	That wasn't the point. There were 2 points to my post: 1. The half wave rectification and 2. Flexibility to use as many combinations of voltages as possible.

17th July 2011, 05:14 AM	#4
! diyAudio Member Join Date: Jun 2005 Location: Midwest	Half wave is just cheaper, can be air wired. Some say the heat/loss reduction helps, it depends on the goal of the supply whether it matters but yes you can do full bridge rect. If you want the greatest flexibility of voltages with you will need to use switching to control which winding leg is the negative return from a full bridge rectifier, or vice versa, have the -250V be the fixed ground and switch which of the remaining 4 is positive to the rectificaton stage... leaving the lead marked as ground in the schematic, floating... and as already mentioned you will need to check the diode and capacitor ratings to be sure they will work.

17th July 2011, 08:38 PM	#6
! diyAudio Member Join Date: Jun 2005 Location: Midwest	The way you have the circuit you could use any ONE adjacent pair of windings (taps in them) for 125V difference but any time you try to use more than 125V, your return is through more than one diode bridge. No you cannot "necessarily" use 125V capacitors. 0V would not always be ground in all possible voltage configurations, your negative rail would be unless you are using a split supply... you did not clearly state EXACTLY what you are trying to do which makes a difference. If you use more than 125V you need to use capacitors rated higher than 125V, they need to be rated for the voltage you are using and connected between the two taps after the respective rectifier bridges. Everyone is entitled to disagree but you seem prone to a strange arguing stance when you don't seem to know. The easiest way to do the circuit is have a floating transformer ground lead, put the -250V lead as the new negative return, use a single bridge rectifier and a single throw 4 pole switch, with each of the other taps as a successive pole on the switch which alternately connects to a single bridge rectifier, the positive output of which is connected to a single capacitor rated for > 500V. That's 500V loaded, you will have to measure peak voltage unloaded to get the actual peak value needed as it could be significantly higher than your diagram shows.

18th July 2011, 11:39 AM	#8
DF96 diyAudio Member Join Date: May 2007	No, that won't work. You can either wire the secondaries in series or the rectifier output in series. You can't do both. If your transformer already has the windings connected together then that is what you will have to use. You could use two bridges, one on the 125-0-125 and the other on 250-0-250, to get two bipolar DC supplies - about +- 170V and +- 340V. You don't have to ground the '0V' on the output, so you could have -170, 0, +170, +340, +510V by grounding the '-170V' output.

19th July 2011, 07:23 AM	#10
DF96 diyAudio Member Join Date: May 2007	The ratio between 125V and 175V is not a matter of transformer load, but RMS vs peak.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)