Good evening all,
I am experimenting with a Frankenstein re-build of an old pro-sound subwoofer. The only voice coil I could find so far is from the car audio world, and is rated at a disturbing 1 ohms nominal impedance.
Do any of you know of a module or proven design which is stable at such a load? I thought of using a car audio amplifier, but it seems a shame to step the mains voltage down to 13.5 or whatever, only to have the power supply in the mobile amp step it back up. North of 500 watts RMS if possible!
Thanks for your help,
JF
I am experimenting with a Frankenstein re-build of an old pro-sound subwoofer. The only voice coil I could find so far is from the car audio world, and is rated at a disturbing 1 ohms nominal impedance.
Do any of you know of a module or proven design which is stable at such a load? I thought of using a car audio amplifier, but it seems a shame to step the mains voltage down to 13.5 or whatever, only to have the power supply in the mobile amp step it back up. North of 500 watts RMS if possible!
Thanks for your help,
JF
well, input and output current is the most problem, and 2R stabil amp will work on 1R, at lower power, but depends on amp, there are some that will push 1R at full power too, not really good for amp, but they can. So if you get say 2.5kw amp stabil at 2R, it will work on 1R @ say 1.5kw easy...
But all depends on many factors, just to name few: your electrical system, speaker impedance raise, design of amp,...
I'm pretty sure you would be best off with buying second hand 1R stable amp.
But all depends on many factors, just to name few: your electrical system, speaker impedance raise, design of amp,...
I'm pretty sure you would be best off with buying second hand 1R stable amp.
Back: Thank you for the reference to the Hypex products. They look very interesting but however they are a little out of my price range.
Wingman: You're right, it's hard to find a 1 ohm capable design that is not intended for automotive use. That may be a product of those SPL competitions (uhg...) in the mobile audio world.
Luka, I agree, I think the cheapest way will probably be to purchase a used car amp.
Here's a question: could I take a massive automotive amplifier, and bypass its internal power supply, feeding it's amplifier section directly wit the voltage it expects? Instead of stepping the mains voltage down to 13.5 or so with a huge transformer, only to have the amplifier step it back up, could I skip this step and supply the higher voltage directly? For example, here is a picture of the inside of a Hifonics car audio amplifier. It looks like the high-voltage power supply section is at the far end, and the amplifier near the camera. I wonder if I could figure out what the high voltage power supply provides, then create my own power supply that uses AC mains voltages?
Has anyone here tried such a project?
Wingman: You're right, it's hard to find a 1 ohm capable design that is not intended for automotive use. That may be a product of those SPL competitions (uhg...) in the mobile audio world.
Luka, I agree, I think the cheapest way will probably be to purchase a used car amp.
Here's a question: could I take a massive automotive amplifier, and bypass its internal power supply, feeding it's amplifier section directly wit the voltage it expects? Instead of stepping the mains voltage down to 13.5 or so with a huge transformer, only to have the amplifier step it back up, could I skip this step and supply the higher voltage directly? For example, here is a picture of the inside of a Hifonics car audio amplifier. It looks like the high-voltage power supply section is at the far end, and the amplifier near the camera. I wonder if I could figure out what the high voltage power supply provides, then create my own power supply that uses AC mains voltages?
Has anyone here tried such a project?
An externally hosted image should be here but it was not working when we last tested it.
Answer is YES, if you provide all the voltages, you are I think good to go, since all rectiviers will block current going into trafos...
But note something, you will be using 50/60Hz trafo [I assume], so it won't hurt if you bring DC into amp, and have some capacitance outside amp, and I would settle for some 10-20v lower voltage then this amp runs at, coz in car, voltage for amp drops a lot on 1R load, normal trafo will hold this voltage better, but then again, higher voltage, less current for same power on output
But note something, you will be using 50/60Hz trafo [I assume], so it won't hurt if you bring DC into amp, and have some capacitance outside amp, and I would settle for some 10-20v lower voltage then this amp runs at, coz in car, voltage for amp drops a lot on 1R load, normal trafo will hold this voltage better, but then again, higher voltage, less current for same power on output
J.R.Freeman said:
I just wanted to say that is my amp
I was wondering why it looked so familiar.
Anyways I am using a computer PSU to supply a BXi1606D(that amps little sibling) current.(Oh I'm running it at 4 ohms...)
You could probably do that, but is it really worth it? There are more than one rail, there's low voltage rails for the op amps, high voltage for the H bridge.
This is mostly armchair speculation, but I think the key to 1R operation is to keep the supply rails low so you don't get into impossible SOA problems. It's when you build something with conventional rails that 1R becomes difficult. With 1R you don't need high rails because something like plus/minus 20V gives you about 200W. Designs for higher power simply use higher rails and a lot of output devices, but there shouldn't be unusual stability issues.
Hey guys,
Thank you all for your comments. I find this really interesting! It's exciting to think there may be a DIY class D solution to the need for a massive power amplifier, at a substantial cost saving over commercial stuff. I guess all of you know this already, but I'm just getting into class D!
Judtoff, that's funny that of all the pictures on that site I chose one of your model amplifier. You're right, I was considering that afterward - the front end is probably low voltage and would need it's own rail. It might not be worth it to try and sort out which power goes where. The advantages of powering the output stage directly, however, would be in efficiency. But it might be easier to start from scratch.
Luka, that sounds like a good idea about running a little lower in terms of voltage across the mosfets, as compared to what they would see in a car. But the more I think about it, the more I think it might not be worth butchering the amp.
Another alternative, though not as efficient, would be to use a massive 12 volt supply. Judtoff, you mentioned a computer supply. That's a good choice as it is a switcher, and is similar to a class D amplifier in the way that it regulates, with similar gains in efficiency and output. What I *may* do, is build something similar. I was thinking of starting with a big transformer from a car battery charger - one which has the 'boost' option. Some of these claim 200 amps peak (for short durations). I could then build a switching section to regulate to 14 volts or so, and a filter section to remove residual AC hum (perhaps this would go before the regulation section).
Anyway, I'm still mulling it over! Conrad, what you mentioned is very interesting. So your idea is that running at such a low impedance of 1 ohm is simply a matter of controlling the current through each mosfet. This is exciting news, because it means I might be able to build an amplifier for my massive (and hungry!) woofer, with a class D driver IC and an appropriately sized mosfet bridge.
Here's a question for you Conrad: you mentioned multiple output devices, presumably to share the current load. How does one gang mosfets together to work as one? Is it as simple as gate to gate, source to source, drain to drain? I would imagine differences in voltage drop across the source and drain during the on state would need to be matched some how, or is this of no concern?
Thanks for all your comments guys! I'm getting worked up thinking about making a big brute of a power amp, to drive my ridiculous woofer
Jim
Thank you all for your comments. I find this really interesting! It's exciting to think there may be a DIY class D solution to the need for a massive power amplifier, at a substantial cost saving over commercial stuff. I guess all of you know this already, but I'm just getting into class D!
Judtoff, that's funny that of all the pictures on that site I chose one of your model amplifier. You're right, I was considering that afterward - the front end is probably low voltage and would need it's own rail. It might not be worth it to try and sort out which power goes where. The advantages of powering the output stage directly, however, would be in efficiency. But it might be easier to start from scratch.
Luka, that sounds like a good idea about running a little lower in terms of voltage across the mosfets, as compared to what they would see in a car. But the more I think about it, the more I think it might not be worth butchering the amp.
Another alternative, though not as efficient, would be to use a massive 12 volt supply. Judtoff, you mentioned a computer supply. That's a good choice as it is a switcher, and is similar to a class D amplifier in the way that it regulates, with similar gains in efficiency and output. What I *may* do, is build something similar. I was thinking of starting with a big transformer from a car battery charger - one which has the 'boost' option. Some of these claim 200 amps peak (for short durations). I could then build a switching section to regulate to 14 volts or so, and a filter section to remove residual AC hum (perhaps this would go before the regulation section).
Anyway, I'm still mulling it over! Conrad, what you mentioned is very interesting. So your idea is that running at such a low impedance of 1 ohm is simply a matter of controlling the current through each mosfet. This is exciting news, because it means I might be able to build an amplifier for my massive (and hungry!) woofer, with a class D driver IC and an appropriately sized mosfet bridge.
Here's a question for you Conrad: you mentioned multiple output devices, presumably to share the current load. How does one gang mosfets together to work as one? Is it as simple as gate to gate, source to source, drain to drain? I would imagine differences in voltage drop across the source and drain during the on state would need to be matched some how, or is this of no concern?
Thanks for all your comments guys! I'm getting worked up thinking about making a big brute of a power amp, to drive my ridiculous woofer
Jim
My knowledge of class D amps is limited, but I'd hope the devices are switching cleanly so dissipation is minimized. They still have to share current equally and I'd suspect some very low value current balancing resistors would be in order. Be sure trace and wire resistance doesn't contribute in an unbalanced fashion, since they'll be a significant part of the resistance. Matching should help, as will keeping the devices at exactly the same temperatures. Use good thermal design. Use good protection as AFAIK it's still a SOA problem.
Hi Conrad,
Thanks for your comments. By SOA do you mean Safe Operating Area? You've hit on some good points. That's a very good point about the resistance in the signal path being a big contributor to overall load. That shouldn't be too hard to deal with, but it is important to keep in mind. Thermal issues shouldn't be too hard to cope with, though my understanding of thermodynamics is rudimentary. I'm hoping that using multiple current-sharing fets should help dissipation issues, though I suppose spacing them out farther apart increases the chances of temperature differential between two widely spaced components. Hmm... something to think about for sure.
The biggest question mark for me at this point is configuring a bank of fets to share current somewhat evenly. It seems this is key, and if a strategy can be solidified, the sky's the limit.
Regarding the current balancing resistors, what values would be appropriate? Should they be matched to the individual fet's source-drain on resistance? I was speaking with a fellow technologist about this problem of current sharing, and he's on the same track as you, with each device having a small resistor at it's output (or input I would imagine would be ok too). However, that's another dissipation to consider. And I've heard of designs using balancing measures at the gate.... I don't see how this could work, but if it's possible it sounds like an attractive option.
On another note, would any of you know of a source for schematics for large high-powered car audio subwoofer amplifiers? It would be interesting to see how they handle current sharing. They certainly employ multiple fets. If we can figure this out, I would be interested in coming up with a design to share with all, as this type of application seems hard to come by in the DIY world so far.
JF
Thanks for your comments. By SOA do you mean Safe Operating Area? You've hit on some good points. That's a very good point about the resistance in the signal path being a big contributor to overall load. That shouldn't be too hard to deal with, but it is important to keep in mind. Thermal issues shouldn't be too hard to cope with, though my understanding of thermodynamics is rudimentary. I'm hoping that using multiple current-sharing fets should help dissipation issues, though I suppose spacing them out farther apart increases the chances of temperature differential between two widely spaced components. Hmm... something to think about for sure.
The biggest question mark for me at this point is configuring a bank of fets to share current somewhat evenly. It seems this is key, and if a strategy can be solidified, the sky's the limit.
Regarding the current balancing resistors, what values would be appropriate? Should they be matched to the individual fet's source-drain on resistance? I was speaking with a fellow technologist about this problem of current sharing, and he's on the same track as you, with each device having a small resistor at it's output (or input I would imagine would be ok too). However, that's another dissipation to consider. And I've heard of designs using balancing measures at the gate.... I don't see how this could work, but if it's possible it sounds like an attractive option.
On another note, would any of you know of a source for schematics for large high-powered car audio subwoofer amplifiers? It would be interesting to see how they handle current sharing. They certainly employ multiple fets. If we can figure this out, I would be interested in coming up with a design to share with all, as this type of application seems hard to come by in the DIY world so far.
JF
For this kind of details, we prefer to be contacted privately. The new module is focused in power capability and reliability, but still with very good sound (global feedback, synchronous design), and it is loaded with protection features (overcurrent/short-circuit, overvoltage, undervoltage, overtemperature with integrated sensor). Size is only a bit bigger than BP4078. It also has a header for small "plugins" we will also offer for quickly and easily implementing filters, crossovers, limiters, volume control, etc.
Price is not fixed yet, but something around 160 euro for 1 unit. Recommended power supply is our SPS80HV, that, can feed two of these modules as we will offer a "reinforced" version capable of up to 1500W).
To be available in the start of last trimester '09, together with another two interesting products.
Price is not fixed yet, but something around 160 euro for 1 unit. Recommended power supply is our SPS80HV, that, can feed two of these modules as we will offer a "reinforced" version capable of up to 1500W).
To be available in the start of last trimester '09, together with another two interesting products.
luka said:One more thing, I think amp7 is 1R stable, just not sure how this would be done, but will be able to find out soon
Hi there Luka,
Is Amp7 something you're working on? It sounds interesting!
Thanks for all the commercial responses. A commercial module would be useful, and it's good to see there are a few more 1 ohm stable options that what I had originally found.
However it would be interesting to find a DIY solution to this problem. If we could get a hold of the schematic for a high-powered current sharing output stage car amplifier, maybe we could work together to make something similar.
Do any of you have resources to share regarding mosfet current sharing or car audio amplifier schematics?
Jim
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.