Hi all. I'm designing a replacement power supply for a mixer which requires +/-15 @ 2.4A, +8V @ 300ma and +48V phantom @ 320ma (absolute maximums), and am purposefully overdesigning it for reliability. I'm using a 2x18VAC toroid trafo with dual secondaries, one for +15V/5A and the other for -15/5A. Both would employ bridge recifiers and 5A 3-pin regulators, tying the (+) output of one to the (-) output of the other for 0 volts. The +15V rail would be sent through an additional regulator to get +8V. The -15V rail would be used to power the cooling fan, LED displays, etc., to help equalize power use. Pretty standard stuff there really.
Now here's the big question for me, as I'm a little shaky on this. I'm attempting to combine a full-wave voltage tripler with the +15V rail's bridge rectifier to generate the pre-filtered, pre-regulated DC voltage needed for the +48V rail. I've come up with the circuit in the attached pic.
Is the schem correct, at least in principle? I don't want to use a second trafo unless I absolutely need to.
Thanks in advance!
Now here's the big question for me, as I'm a little shaky on this. I'm attempting to combine a full-wave voltage tripler with the +15V rail's bridge rectifier to generate the pre-filtered, pre-regulated DC voltage needed for the +48V rail. I've come up with the circuit in the attached pic.
Is the schem correct, at least in principle? I don't want to use a second trafo unless I absolutely need to.
Thanks in advance!
Attachments
Last edited:
Your schematic is not clear (where is the center tap, etc), but anyway, it is not the right way to do it.
You should use a single bridge for the +/15V, and a separate multiplier for the 48V.
Trying to reuse the +15V as a base voltage for the tripler wouldn't be a good idea, because of the drawbacks and because you don't need it: with the CT winding, you can spare a stage anyway.
See example:
You should use a single bridge for the +/15V, and a separate multiplier for the 48V.
Trying to reuse the +15V as a base voltage for the tripler wouldn't be a good idea, because of the drawbacks and because you don't need it: with the CT winding, you can spare a stage anyway.
See example:
Attachments
You said you have dual secondaries. You are free to connect them yourself in a CT configuration. You just have to wire them in series.
if you have a dual secondary then it can be converted to centre tapped as already explained.
However, when using a single bridge rectifier to produce a dual polarity supply, you can only use a centre tapped transformer.
If you have dual bridge rectifiers for dual polarity supplies you can only use isolated secondaries, one to each bridge rectifier.
If you need additional power from those secondaries, then you cannot use both, you must use one secondary only.
You will find that the third power supply is not completely isolated from the the "normal" dual polarity supplies.
If your phantom PSU needs isolated power, then I think you need a third winding.
However, when using a single bridge rectifier to produce a dual polarity supply, you can only use a centre tapped transformer.
If you have dual bridge rectifiers for dual polarity supplies you can only use isolated secondaries, one to each bridge rectifier.
If you need additional power from those secondaries, then you cannot use both, you must use one secondary only.
You will find that the third power supply is not completely isolated from the the "normal" dual polarity supplies.
If your phantom PSU needs isolated power, then I think you need a third winding.
Let me try to explain something, which I attempted to explain in my first post but appears to be a total failure. I know I can connect the secondaries to make a CT, but I am not doing that. I am using one secondary for the +15V rail - bridge rectifier, caps, regulator, etc. So far, so good. Now, the other secondary is being used for the -15V rail - bridge rectifier, caps, regulator, etc. (same as for the +15V rail), but configured as a floating rail so I can connect the normally positive output to ground, thereby creating the -15V rail.
I hope I have cleared that up.
So, using that scheme I have only the secondary used for the +15V rail to generate the +48V. Using a voltage tripler, or even a quadrupler (power requirement is relatively minimal), seems the best bet. Trouble is, I'm just not that familiar with them. I could always use a step-up trafo (or a power trafo in reverse) directly off the secondary to generate the higher voltage. I've done that before with a trafo I had on-hand, but I would prefer to use a tripler or quadrupler to avoid the extra trafo cost if I can. The circuit I came up with was a combination of a standard bridge rectifier and a full-wave voltage tripler with common-connected elements used for both.
I hope I have cleared that up.
So, using that scheme I have only the secondary used for the +15V rail to generate the +48V. Using a voltage tripler, or even a quadrupler (power requirement is relatively minimal), seems the best bet. Trouble is, I'm just not that familiar with them. I could always use a step-up trafo (or a power trafo in reverse) directly off the secondary to generate the higher voltage. I've done that before with a trafo I had on-hand, but I would prefer to use a tripler or quadrupler to avoid the extra trafo cost if I can. The circuit I came up with was a combination of a standard bridge rectifier and a full-wave voltage tripler with common-connected elements used for both.
Why are you doing that? Even if your 48V requirement is not taken into account, it has a number of disadvantages.
The only reason for doing it would be the ability to use a positive regulator, but that would be a bad reason, especially with a toroidal transformer.
Many thanks to everyone for the replies. I'd say it's apparent I've been looking in the wrong places for help with this design before now. My fall-back position always was to use separate power transformers for each voltage. I have to get the mixer back in service very soon and don't have enough time now to figure this out so it looks like that's what I'll have to do. I'm more of a builder than a designer anyway, and I really have to get started building. 32-channel mixers aren't cheap and I need to use it on New Year's Eve. I was hoping to reduce my PSU costs, but it is what it is. Again, thanks!
Last edited:
Voltage Doubler / Tripler Analysis and Design Method
I came across this post while trolling the web for ideas on implementing a tripler for a project I am thinking about. While I realize the post has been inactive for a while and my post is not exactly to the original point, I was inspired to do some analytical work and to come up with a design methodology for a tripler like the one discussed here. The attached file contains the analysis and the design curves for both a full and half wave version of the tripler and shows how this is extended to cover a doubler as well.
Please let me know if anyone sees an error in the analysis.
Otherwise, hopefully someone finds this useful!
I came across this post while trolling the web for ideas on implementing a tripler for a project I am thinking about. While I realize the post has been inactive for a while and my post is not exactly to the original point, I was inspired to do some analytical work and to come up with a design methodology for a tripler like the one discussed here. The attached file contains the analysis and the design curves for both a full and half wave version of the tripler and shows how this is extended to cover a doubler as well.
Please let me know if anyone sees an error in the analysis.
Otherwise, hopefully someone finds this useful!
Thank you & Other Sources
Thank you Elvee!
Subsequent to this post another friend of mine pointed out that that there are design curves available in the AARL Handbook (see Fig. 11.10 at arrl-hand-book ¸ßƵ ÌìÏß ÉäƵ Éè¼Æ ѧϰ 11_°Ù¶ÈÎÄ¿â) which were taken from an article by Schade Analysis of Rectifier Operation). Schade's curves were based on some even earlier work by Waidelich ("The Full-Wave Voltage-Doubling Rectifier Circuit," Proc. of the IRE, p.554- 558, Oct. 1941.) I found all of their analyses as clear as mud so I did my own.
Their circuit models are more similar to the one you simulated and posted earlier. I did enter your circuit and ran it myself as it was useful as a reference for my analysis once I removed the parasitics to make it more ideal and in line with my model. I though about adding the parasitics into my analysis, but when I stopped to think about it the parasitics will have to small to begin with otherwise the performance of the multiplier will not only be poor, but there will be a notable interaction with the recharging of the main power supply caps (the ones on VCC & -VCC in my diagram). Hence, I felt that including the parasitics and complicating the analysis was not worth the effort given the practical constraints.
Thank you Elvee!
Subsequent to this post another friend of mine pointed out that that there are design curves available in the AARL Handbook (see Fig. 11.10 at arrl-hand-book ¸ßƵ ÌìÏß ÉäƵ Éè¼Æ ѧϰ 11_°Ù¶ÈÎÄ¿â) which were taken from an article by Schade Analysis of Rectifier Operation). Schade's curves were based on some even earlier work by Waidelich ("The Full-Wave Voltage-Doubling Rectifier Circuit," Proc. of the IRE, p.554- 558, Oct. 1941.) I found all of their analyses as clear as mud so I did my own.
Their circuit models are more similar to the one you simulated and posted earlier. I did enter your circuit and ran it myself as it was useful as a reference for my analysis once I removed the parasitics to make it more ideal and in line with my model. I though about adding the parasitics into my analysis, but when I stopped to think about it the parasitics will have to small to begin with otherwise the performance of the multiplier will not only be poor, but there will be a notable interaction with the recharging of the main power supply caps (the ones on VCC & -VCC in my diagram). Hence, I felt that including the parasitics and complicating the analysis was not worth the effort given the practical constraints.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.