2x250 IRS2092 amp build

[[email protected]]

Re: 2x250 IRS2092 amp build

Thanks so much for being patient with me. I will make the changes.

[neildavis]

Re: 2x250 IRS2092 amp build

My psu is exactly like the top one here. GD does not go to the amp, only to ground like the top pic.
http://techtalk.parts-express.com/at...3&d=1400451374

Here's maybe an easier way to see the problem with the original configuration: If you connect a preamp or some other device that has a chassis ground connected to its signal ground (which is often the case), then the "Gnd" at the IRS2092 ("G" in the lower drawing and signal ground for the preamp) will be connected to GD through "earth". That doesn't cause any sparks or high current, because those caps don't have any voltage on them anyway. However, it becomes obvious that the caps "in the middle" aren't contributing to the power supply filtering.

So you want to get rid of that GD connection. If you just remove it, your caps would be in series, as others have pointed out. So you need to do some rewiring to get all of those caps in parallel. What you have now will actually work, but you are only using half of the capacitors to filter the heavy supply current and the chassis isn't properly connected to signal ground.

I only posted because it looked like there was still some confusion as to why the original configuration wasn't correct. Hong has the right circuit now. In this case, simpler is better

[[email protected]]

Re: 2x250 IRS2092 amp build

My psu is exactly like the top one here. GD does not go to the amp, only to ground like the top pic.

http://techtalk.parts-express.com/attachment.php?attachmentid=46993&d=1400451374

[Wolf]

Re: 2x250 IRS2092 amp build

From +v to ground and ground to v-. They are polarized so should the cap on the v- be turned around?

No- don't flip it around.
Later,
Wolf

[[email protected]]

Re: 2x250 IRS2092 amp build

Hong,
Yes technically if you are looking from +V to -V there are two caps wired in series. But every thing in the electronics world is almost always referenced to common (or ground). In that case there is only one cap (it is actually two 20,000uF in parallel) from the +V rail to ground and one cap from the -V rail to ground.

From +v to ground and ground to v-. They are polarized so should the cap on the v- be turned around?


In my case, with dual rectifiers, should the ground be omitted and the caps be wired individually?

[[email protected]]

Re: 2x250 IRS2092 amp build

OMG I am hankerin' to build an amp like this. Lowell, the work is amazing and we are licky to have the big guns to prevent our ill fate. I would love to build a 500WPC amp. Are there any out there?

Pete Schumacher ® did a 500wpc with two bridged chip amps. Its not in his sig anymore though.
At 4ohm, the 2x250 is close to 500wpc.

[MSaturn]

Re: 2x250 IRS2092 amp build

Neil can easily tell you how to get the capacitance you want .. diagram wise. You may want to look in to tracing your current wiring so everyone is clear on what's happening and what should happen.

[mzisserson]

Re: 2x250 IRS2092 amp build

OMG I am hankerin' to build an amp like this. Lowell, the work is amazing and we are licky to have the big guns to prevent our ill fate. I would love to build a 500WPC amp. Are there any out there?

[[email protected]]

Re: 2x250 IRS2092 amp build

By the way, I measured 48vac on both secondaries and 65.3 vdc on v- and v+ with no load.

[[email protected]]

Re: 2x250 IRS2092 amp build

Lowell,
I'm curious if you can hear any difference between this amp and the other one you built using the two 16,000 uF computer grade electrolytic caps? They both use the same amp module. This one's cap bank should have lower ESR. I'm wondering if that is audible or not. Thanks.

Its really hard to say at this point Craig. In my small room, they seem the same. The motivation for this build was to have more power behind it. The first ran out of juice in my home theater during high transient movie scenes. It has 16000 per rail after a 650va toroid. This was meant to have 37000 after a 1000va toroid. The truth is, I don't really know how much capacitance I have. It will be simple to wire all the caps individually in parallel but not sure about grounding it. The schematic Hong posted shows a ground with between caps just like mine. Was said that it was right but mine is wrong. So confusing.

[hongrn]

Re: 2x250 IRS2092 amp build

Hi Bob,

All my PSUs built using one transformer and a single bridge rectifier are wired correctly, as indicated in the top diagram of the attachment in post #63. The confusion comes from the Sure board instructions calling for 2 separate PSUs, so I wired two of them together, as shown in this attachment. The rail voltages come out correctly, and in my case, they were -56V/0/+56V. The amp board plays loud and clean, and there is no hum of any sort. There may be a floating ground some where in there, but it did not affect the amp at all. Lowell went down the same road I did (sorry man)...

Hong: So...Did you have a power supply that was wired incorrectly or was there just some confusion with schematics....Same question for Lowell.

Bob...... Slowly recovering

[robertcottiers]

Re: 2x250 IRS2092 amp build

Hong: So...Did you have a power supply that was wired incorrectly or was there just some confusion with schematics....Same question for Lowell.

Bob...... Slowly recovering

[PWR RYD]

Re: 2x250 IRS2092 amp build

Lowell,
I'm curious if you can hear any difference between this amp and the other one you built using the two 16,000 uF computer grade electrolytic caps? They both use the same amp module. This one's cap bank should have lower ESR. I'm wondering if that is audible or not. Thanks.

[donradick]

Re: 2x250 IRS2092 amp build

I agree!

Sure am glad that we have real experts here who can advise us.
In addition, I'm now convinced of the pitfalls.. I will never build a power supply with 2 bridge rectifiers.....

[hongrn]

Re: 2x250 IRS2092 amp build

Almost there Bob.

When this thread is brought to a conclusion I will calmly check into a rehab facility....:D


Insane in the membrane...