better op-amp than TI LM318P ?

gadget73

junk junkie
Subscriber
I've got a phono preamp with a bad LM 318P. Since I have to replace it, is there something that would be an upgrade at a reasonable price? This is at the input stage of the preamp, so low noise would be a priority.

This unit has two of these, so I'd replace both. Socketed too, which is right handy.
 
LM318 is a very old opamp, it is fast, but I don't think you need this fast for audio. There are a lot of much lower distortion opamps that are designed for high end audio.

What is the supply rail voltage as LM318 spec for +/-20V max. Some opamp cannot take +/-20V, so you have to be careful.

One opamp comes to mind is OPA604 single opamp:https://www.digikey.com/product-detail/en/texas-instruments/OPA604AP/OPA604AP-ND/251162
https://www.digikey.com/product-detail/en/texas-instruments/OPA604AP/OPA604AP-ND/251162
This one can only use up to +/-17V:https://www.digikey.com/product-detail/en/texas-instruments/LME49710NA-NOPB/296-43971-5-ND/1302099

Just go on Digikey:https://www.digikey.com/products/en...t=0&page=1&quantity=0&ptm=0&fid=0&pageSize=25
 
+/- 15v on this one

This is a mid 1980s vintage Radio Systems phono preamp, so no surprise that things have been updated in the last 30 years.

Package has to be DIP8 if that changes things. Frankly I don't pretend to know what I'm looking at when it comes to op-amp specs.
 
Then you have a wide selections of better opamps. Your limitation is actually you need a single opamp, not a dual!!!! Most come as dual.

Things I usually look for audio opamps are:

1) THD spec. Most audio opamps are 0.001% or below at 20KHz driving 600ohm load. It is important the opamp be able to drive.

2) They always specify output current. I look for at least 20mA output current.

3) It has to be unity gain stable ( not in your case though as you need gain for phono amp).

4) Minimum 6 to 10MHz bandwidth.

5) Slew rate of 10V/uS or above.

For phono preamp, I'd try the LME49710.

Hope this help.
 
Be careful. If the 318 is used with a discrete front end, as they do in the old app notes, you can't just put another opamp in there. There may be no need, as the composite 318 design was quite good. You have to check the schematic to see if they're using the offset pins as the inputs.
 
since i confess to not knowing what that means, maybe you can tell me? IC1 is the one that died on me.

pa1 schematic.jpg

I *think* the initial failure was because of power supply issues. It had a 120hz hum, which I traced to ripple on the + rail of the supply. New filter caps helped, but didn't make it go away. When I went to check for ripple again, I noticed the op-amp was hot after being on for maybe a minute. Moving the op-amp to the other channel moved the hum with it, and it still got hot. The good channel's amp works fine in both sides, so I'm pretty certain its the chip that gave up. I'm hoping that with the power supply rebuilt it will stay good.
 
since i confess to not knowing what that means, maybe you can tell me? IC1 is the one that died on me.

View attachment 948327

I *think* the initial failure was because of power supply issues. It had a 120hz hum, which I traced to ripple on the + rail of the supply. New filter caps helped, but didn't make it go away. When I went to check for ripple again, I noticed the op-amp was hot after being on for maybe a minute. Moving the op-amp to the other channel moved the hum with it, and it still got hot. The good channel's amp works fine in both sides, so I'm pretty certain its the chip that gave up. I'm hoping that with the power supply rebuilt it will stay good.
This cannot be a complete circuit. It is incomplete as you need the resistors I added as shown:

LM318.JPG

If you want to play it safe, just get the LM318 and be done with it.
 
Pins 1 and 5 are the offset null pins, which makes this a nonstandard application of the chip. Since it's still available, and cheap, I would go with the original.
 
Yes, that's the circuit I described from the National Semi app notes. No extra resistors required- current is sourced from the opamp. It works very well- I built one years ago. You have to use the 318, so just go buy another.
 
You mean just by the input bias current? That's very little. You can use only the very old opamps that draw so much bias current.

I don't necessary think just because it's in their application notes, then it's a good circuit. Been challenging the app engineer more than a few times in those semi conductor companies. Been to Linear Tech to show how their ADC fails, two of the app circuit on switching regulator oscillate.

I am sure with the new opamps, you can design without using those transistors.
 
OK, well if it has to use the original piece, I guess thats what I'll do. They're like 1.25 each, so no big deal.

Thanks guys. Me and this modern tech don't get on so well.
 
You mean just by the input bias current? That's very little. You can use only the very old opamps that draw so much bias current. ...

No, the inputs are shut off. The signal now goes in the offset terminals, where hopefully the current is higher. The circuit isn't drawn correctly and is a bit misleading because we're used to seeing the input pins on the face of the triangle. I think it was drawn correctly in the app note. No question it's an out of the box thinking circuit. Certainly with today's parts one probably wouldn't resort to things like that, but I did build it and it was very good in all respects.

Here it is- http://www.elenota.pl/datasheet-pdf...r/AN-222?sid=2801096902def3fd51cf078c0332677b
 
Last edited:
This particular unit was made in 1988 I think, but its possible the design is even older. Its a phono preamp from a radio station. Its fairly well constructed but I presume it was built to conservative standards, being intended to run 24/7. It doesn't even have a power switch.
 
Back
Top Bottom