Okay if you want to look at it that way, to have the same wattage draw, the 23mA max recommended output for good quality output from the LM4562 at 17v, would be 32.5 mA @ 12v. But keep in mind that the evaluation boards come with 15v regulators on them. So that 15v this takes it to 26 mA. 26mA, and the specs on the board say they draw up to 60.
Last edited:
To stop all speculations a number from NC500 datasheet (absolute maximum ratings):
Input current Iin: 10mA
Same number can be calculated with applying the Ohm law, knowing the voltage range at the buffer outputs (or NC500 inputs) and input impedance of the NC500 😉
I can clearly see what it says. The maximum current drawn from the buffer circuit is 60mA from the + reg, and 50mA from the - reg.
This is designed for use only with the NC-500. So they are talking about current draw when connected to the NC-500's.
Last edited:
I found Richards post. He figured 50-70mA so was pretty close to Hypex's data. However he did overestimate the LM4562's max output current which is a measly 23mA for optimum performance into a 600ohm load (53 mA for instantaneous destruction). Vs the max 150mA into 75 ohm load for optimum performance from the SIL-994
Last edited:
The NC400's have that same rating there. Let's let Bruno himself clarify your misconception:
Can you read english? It is in bold in your quote!!! supply current draw vs the buffer output current. Don't you understand that those two are not the same thing and different circuits?
Please stop posting random numbers and graphs with random data and on this basis call others as having misconceptions. Your signal to noise ratio is largely dominated by noise.
And please - don't quote with images, as it is pretty impossible to appropriately re-quote them or do any other textual operations on them...
Yes Bruno clearly says that the discrete buffer circuit on the NC-400's (Vsigs) draws 40mA. The 10mA is not related to supply current.
The NC-500 has no onboard buffer like the NC-400. So you must look to the evaluation board datasheet for that spec. And it clearly says 60mA for the + and 50 mA for the -
The SIL-994 is 30x the cost of the LM4562. Would it be a travesty to admit that it just might preform a little better in this application?
Admitting this is not going to put TI out of business.
Admitting this is not going to put TI out of business.
Oh good God. 60mA is the max entire current for Isig and Iaux. That includes the draw from the DUAL op amp. The LM4562 is fine for at least 26mA per channel.
You know what the feedback resistors are for the op amp. Make an estimate of the load presented by the amp - perhaps you want to use your rough number or perhaps you want to model it with the circuit provided for this very purpose in the data sheet. You know Ohm's Law. Now make an assumption for Vsig and do the division. If you did the simple model you can sweep for Vsig. Here's a clue: there's a very good reason the feedback resistors are set at the high levels they are (when lower values would lower noise substantially).
You know what the feedback resistors are for the op amp. Make an estimate of the load presented by the amp - perhaps you want to use your rough number or perhaps you want to model it with the circuit provided for this very purpose in the data sheet. You know Ohm's Law. Now make an assumption for Vsig and do the division. If you did the simple model you can sweep for Vsig. Here's a clue: there's a very good reason the feedback resistors are set at the high levels they are (when lower values would lower noise substantially).
I see Hypex has a new datasheet on the OEM evaluation board. It confirms Richard's claims that the buffer circuit consumes up to 60mA. This would take the LM4562 right to the edge of self destruction at maximum output.
Here's what Richard said about the SIL-994:
"The 994Enh-Ticha can sink and source 150mA without breaking a sweat. 200mA and the THD at 1Khz goes from 0.0003 to about 0.0005. What this means is you can drive a very LOW IMPEDANCE. Which is what you have with this Hypex POWER AMP input."
Can you provide a link for that datasheet please?
It clearly says the 60mA is for the buffer circuit alone. Are you saying that the resistors are gobbling up most of the power? If you look at the LM4562 datasheet it says it's only good for 23mA if quality is of concern. It takes more than simply wishing it could handle more power to make it actually handle more power.
Post your model or calculations once you've done the work. This is the second time we have covered this point.
The datasheet that says the buffer circuit alone draws 60mA is enough for me. I trust Hypex publishes accurate info on their datasheets. I also trust Richard's findings having tested the buffers and amps in his lab with top of the line AP gear.
Anyways it sounds like crap compared to the SIL-994 and anyone who compares will clearly hear this.
Last edited:
You are arguing that the current demands placed on the LM4562 exceed its capabilities. Present your calculations. They're not complex.
The datasheet shows this. That's enough for me. It's adequate for simple OEM evaluations of the NC-500's. Do you really think they were going to use an exotic discrete opamp like the SIL-994 in a cheap and simple evaluation board? It's only guys like you that think manufacturer evaluation boards are the holy grail of what's possible from the products.
No but there must be some reason for all of the crying about the SIL-994 putting the LM-4562 to shame. So just in case these guys work for TI I thought I'd make that clear.