Greetings from the thermionic emission side of the isle
As you can probably tell by my username, I'm a toob kinda guy, so I'm just dipping my toes here.
I recently bought a pair of Extremis 6.8's to use from 300-400hz down in a tri-amp setup. Not wanting to drain the grid with a array of big hot tubes, and the AC needed to cool the room, I have decided to go over to the dark side () with a UCD180 on the woofers (SE tube elsewhere).
This brings me to my question. I found this surplus power supply (PDF attached), and I need the digi experts to tell me if it's suitable for a UCD180 mono-block driving a stable 8ohm load.
That is all.
-Casey
As you can probably tell by my username, I'm a toob kinda guy, so I'm just dipping my toes here.
I recently bought a pair of Extremis 6.8's to use from 300-400hz down in a tri-amp setup. Not wanting to drain the grid with a array of big hot tubes, and the AC needed to cool the room, I have decided to go over to the dark side () with a UCD180 on the woofers (SE tube elsewhere).
This brings me to my question. I found this surplus power supply (PDF attached), and I need the digi experts to tell me if it's suitable for a UCD180 mono-block driving a stable 8ohm load.
That is all.
-Casey
Attachments
This eBay store sells something similar (if not identical). It is a bit dearer but he provides detailed information and instruction on how to adapt the SMPS to your application.
Stadholder Tek
Modification instructions:-
Stadholder Tek
Modification instructions:-
I asked the vendor Stadholder Tek his opinion about using his SMPS with Hypex UCD 180 and he replied:
"If the SMPS is well filtered and shielded, than the class D amp does not know, and cannot know how its DC supply voltage was produced."
Now, how would you filter it?
Antonio
"If the SMPS is well filtered and shielded, than the class D amp does not know, and cannot know how its DC supply voltage was produced."
Now, how would you filter it?
Antonio
+-45v @ 2A is pretty useless for a 4 or even 8 ohm load. You will be lucky to get a clean 25 watts of output power before the SMPS runs out of current. You would only be able to use the supplies full power with a high impedance load.
Don't skimp on your UCD project. Either take the money you would have wasted on those supplies and buy a proper transformer, or save up a little more and get a proper SMPS from coldamp.
Don't skimp on your UCD project. Either take the money you would have wasted on those supplies and buy a proper transformer, or save up a little more and get a proper SMPS from coldamp.
I don't know much about SMPS and I'm hoping to learn something here.
Just looking at the numbers - the Coldamp SPS30 is a 400W +/- supply - so it would provide 3.3A on each rail.
Combining two of these Lambda supplies would provide 4A on each 45V rail. If you are only interested in voltages of around +/-45....wouldn't this be suitable?
Just looking at the numbers - the Coldamp SPS30 is a 400W +/- supply - so it would provide 3.3A on each rail.
Combining two of these Lambda supplies would provide 4A on each 45V rail. If you are only interested in voltages of around +/-45....wouldn't this be suitable?
Combining two of these Lambda supplies would provide 4A on each 45V rail. If you are only interested in voltages of around +/-45....wouldn't this be suitable?
I didn't say in the start of this thread (my bad), but that was my thinking. After reading the link on Lambda mods posted by Fin, I became convinced that they could be modded to run in parallel (though the link doesn't say this specifically). So convinced in fact, that I ordered 6 of them..3 per channel for 6 amps continuous, 9 amps peak. Thats a little over a nickel a watt, and at 80 watts per unit, wont need any forced cooling
On another front, I have decided to hold off on digital for now. I just need an amp for upper bass/ lower mid and lower (Extremis 6.8 in a transmission line), and have decided that a BPA200 chip amp will give me twice the slam for the same price. This will also allow me to run a linear post regulator down to +/-35V for more noise reduction/stiffer supply.
Thanx all for your time...I'll be back I'm sure.
-Casey
Fin said:I don't know much about SMPS and I'm hoping to learn something here.
Just looking at the numbers - the Coldamp SPS30 is a 400W +/- supply - so it would provide 3.3A on each rail.
Combining two of these Lambda supplies would provide 4A on each 45V rail. If you are only interested in voltages of around +/-45....wouldn't this be suitable?
You are overlooking the fact that the supplies from Coldamp can be requested in other voltages. A SPS30 with a voltage of +-35v @ 5.7A would be far more useful for 4-8 ohm loads than one with +-60v @ 3.3A.
I wouldn't use multiple SMPS to create a single supply if you can avoid it. Most SMPS have fairly high idle power consumption, so using multiple supplies will eat up quite a few watts even when there is no load.
The output current you are calculating that way is the continuous DC current allowable. As you said, a SPS30 outputs 400W and at +/-60V that means 3.3A continuous per rail.
However, the instantaneous current is much higher, for example, you can output a continuous sinewave of 110Vpp at 4 ohm. That means around 370W rms. If you calculate the peak current, it is 55/4=13.75A.
We can, of course, provide supplies with other voltages on request. Just drop us an e-mail to info@coldamp.com and we will be glad to help you.
Best regards,
Sergio
However, the instantaneous current is much higher, for example, you can output a continuous sinewave of 110Vpp at 4 ohm. That means around 370W rms. If you calculate the peak current, it is 55/4=13.75A.
We can, of course, provide supplies with other voltages on request. Just drop us an e-mail to info@coldamp.com and we will be glad to help you.
Best regards,
Sergio
A SPS30 with a voltage of +-35v @ 5.7A would be far more useful for 4-8 ohm loads than one with +-60v @ 3.3A.
If you read the comments at All Electronics I linked to, And The mod link from Fin, the real world output of this Lambda is around +/- 43V under load. With a handful of parts, and an afternoon you can make it anything from 30 somthing to 50 something without stressing it out. I'm leary of "switchers" in general, and plan to run a linear reg after it. So with a target of +/- 35V, 40 or so volts out of the switcher sounds about right.
I wouldn't use multiple SMPS to create a single supply if you can avoid it. Most SMPS have fairly high idle power consumption, so using multiple supplies will eat up quite a few watts even when there is no load.
This is a concern, I guess I'll find out if it's excessive or not, and if there is some kind of "stand by" scheme I can come up with.
The output current you are calculating that way is the continuous DC current allowable. As you said, a SPS30 outputs 400W and at +/-60V that means 3.3A continuous per rail.
However, the instantaneous current is much higher, for example, you can output a continuous sinewave of 110Vpp at 4 ohm. That means around 370W rms. If you calculate the peak current, it is 55/4=13.75A.
We can, of course, provide supplies with other voltages on request. Just drop us an e-mail to info@coldamp.com and we will be glad to help you.
Best regards,
Sergio
I have little doubt that the Coldamp switcher is a fine product, and if I wasn't such a tight a** (have to be), I would give it a try.
-Casey
So you are going to parallel a load of those cheap supplies in order to hope for more current?
The short answer is yes. The "cheap" is the price, not the quality..Lambda makes very good supplies.
I'm afraid that's likely to result in smoke at the least.
Only if you have no idea what your doing..in which case..well...
-Casey
I found some information on what might need to be done in order to connect some of these supplies in parallel.
It is from this Site:-High Efficiency Power Converter
Does this help in any way?
It is from this Site:-High Efficiency Power Converter
From above site......
When two (or more) power supplies of the type described here are placed in parallel, a problem can arise. If one power supply is turned on while another is left off (i.e. not switching), the output bus voltage generated by the first supply will appear at the gates of the second supply's synchronous rectifiers. Once this voltage rises above the threshold value, these synchronous rectifiers will turn on and draw current. At the least this will result in extra dissipation, but it could result in a shorted output bus. This problem can occur even if both supplies are turned on and off together if one supply's transition "gets ahead" of the other.
There are several approaches to solving this problem. One is to make sure both supplies have matched transitions. Another is to connect the supplies together with ORing diodes so that no supply can draw current from the combined output bus.
Does this help in any way?
Does this help in any way?
Bigtime
I hadn't considered this was possible, I assumed the filter caps on the output would take care of this. A simple diode on the output would prevent this. I wonder what a ORing diode is .
I'll have to read the whole thing to be sure they're not talking about paralleling the switchers before the caps.
Thanx.
-Casey
Appearently the concept of ORing diodes refers to creating the logic OR function.
Ahh...as in "this output OR that output".
Sheesh..as if there's not enough confusion out there as-is without this kind of "marketing" language.
-Casey
I contacted technical support at Lambda and asked this question:-
And this was the reply:-
Richie00boy might be right.........
Is it possible to connect two or more of the SC120 SMPS modules in parallel to provide increased current?
And this was the reply:-
Not advisable. There is no current share and you would need blocking diodes.
Richie00boy might be right.........
And this was the reply:-
Not suprising, they have a rep to maintain, and would surely advise against using there product "out of spec".
To be clear, I do not intend on hooking the outputs directly together. They would go through a load balancing resistor network first (think emitter resistors in a power amp). I need to "burn" off 8V anyway, so a half a volt or so across a resistor is no big deal before the linear reg.
I will be doing some breadboarding before I arrive at the final scheme.
-Casey
Just sorry I didn't manage to jump in before you bought them. Those two reasons where exactly what I was getting at. Sorry I didn't say at the time but I was in a rush.
Now I see using dropping/sharing resistors is intended, you might be OK, but you will worsen supply impedance.
Now I see using dropping/sharing resistors is intended, you might be OK, but you will worsen supply impedance.
Now I see using dropping/sharing resistors is intended, you might be OK, but you will worsen supply impedance.
This will be before the linear regulators, so I don't see it affecting performance in any real way.
Just to give an idea as to how robust these supplies are, here's a snip of a review from a user of these supplies...
Might as well be rated at 43 volts--that's where they're at their best. blue trimpot adjusts the 48v line from 46 to 51v. I used 2 of these supplies in parallel (that's right) to provide +/- 43V 3 A for an audio power amplifier. It works well if you adjust the tripots so both supply voltages are matched
This guy apparently strapped 2 supplies directly together (not my plan) with no ill effects.
Between blocking diodes, and sharing resistors, with a post reg., I see no reason this won't work.
-Casey
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Class D
- Will This Supply Work With UCD180