Yes, since I think most people would use the term "linear" also for the basic unregulated transformer + diodes + capacitors PSU, as opposed to a SMPS. The examples in Elliott's page seem to be quite similar to the simple PSUs people build for chipamps. It would be quite interesting to see measurements from actual PSUs.
^ Well I don't, that explains a couple differences in opinion. I see no use to affixing the adjective "linear" if it is not linear regulated, and if you buy one described as "linear" made by a reputable business, it is practically an industry standard that it means linear regulated... at least in the US, maybe not everywhere it is true? Examples
Voltage measurements are not as significant as noise frequency in many cases. Consider that if your supply has a somewhat high 2V ripple at 60Hz (mains AC line frequency in the US), the rate change is only 120V/s. Suppose a conservative 50KHz switching PSU with 36mV (p-p). 50,000 * 36mV = 1800V/s.
Voltage measurements are not as significant as noise frequency in many cases. Consider that if your supply has a somewhat high 2V ripple at 60Hz (mains AC line frequency in the US), the rate change is only 120V/s. Suppose a conservative 50KHz switching PSU with 36mV (p-p). 50,000 * 36mV = 1800V/s.
Last edited:
OK, I see, but Wikipedia seems to distinguish between unregulated and regulated linear supplies, and a quick Google search came up with some 13 900 hits for "unregulated linear power supply".
I think the relevant comparison needs to be with the PSUs people actually build for use with chipamps. A smallish transformer, a rectifier, and a remarkably small capacitor bank seems to be the standard equipment.
I think the relevant comparison needs to be with the PSUs people actually build for use with chipamps. A smallish transformer, a rectifier, and a remarkably small capacitor bank seems to be the standard equipment.
Wow, this sure got going.
By linear I meant the standard Lambda-type rectified and regulated ps.
Sorry about the term.
My point was that there are a lot of cheap good switchers available, so why not try them?
No transformer to source, you know.
And regarding buying a pre-built amp; chip amps are already built, so using a pre-built power supply is not much different, in my opinion.
I thought the fun of these amps was simplicity and low cost.
It's all good, as long as it's fun.
By linear I meant the standard Lambda-type rectified and regulated ps.
Sorry about the term.
My point was that there are a lot of cheap good switchers available, so why not try them?
No transformer to source, you know.
And regarding buying a pre-built amp; chip amps are already built, so using a pre-built power supply is not much different, in my opinion.
I thought the fun of these amps was simplicity and low cost.
It's all good, as long as it's fun.
Absolutely, try everything you can get your hands on and pick by sound which is ok for use.
As for transformer size in DIY projects, I find that DIY builds tend to, on average, use significantly larger transformers per the amp target or rated power than you find in all but fairly high end/expensive manufactured gear.
As for remarkably small transformer banks, I don't agree with the philosophy in using them but see the edit to my prior reply, that it is not necessarily the ripple voltage that matters as much as the rate when the amp is solid state with an (integrally active) feedback loop (typical chipamp).
As for transformer size in DIY projects, I find that DIY builds tend to, on average, use significantly larger transformers per the amp target or rated power than you find in all but fairly high end/expensive manufactured gear.
As for remarkably small transformer banks, I don't agree with the philosophy in using them but see the edit to my prior reply, that it is not necessarily the ripple voltage that matters as much as the rate when the amp is solid state with an (integrally active) feedback loop (typical chipamp).
Last edited:
Oops, above I meant "as for remarkably small capacitor banks"... If I'm building an unregulated, what you're calling unregulated linear PSU, I would not use only a relatively low amount of capacitance, instead using a few hundred uF very near the amp chip then at least 1,000uF per required PSU amps needed to meet the target build power.
If other people like it a different way, good for them as it saves them a little money, time and space... but their often cited reasons that it kills the high end definition to have larger caps is in error I believe, that what they really needed was a couple smaller ones pretty close to the chip pins (a ceramic or film cap even closer).
If other people like it a different way, good for them as it saves them a little money, time and space... but their often cited reasons that it kills the high end definition to have larger caps is in error I believe, that what they really needed was a couple smaller ones pretty close to the chip pins (a ceramic or film cap even closer).
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.