Two Questions, Reading leads on Opamp and Offset Trim

[pjpoes]

Hey Guys, I have a question. I replaced some older caps with OPA627 caps. Based on the information, if you look at the top of the chip, facing in the direction so that the little dimple is up, my pins seem to be reversed based on the cuircit. Am I reading it upside down. My assumption is I must be. Again, with the little notch facing up, the NC pin is the left top most pin, not the right top most pin.

Ok there is a capacitor, ceramic cap rated at 22pf that connects the offset trim to, I assume the -v pin. What does that capacitor do? Does it affect sound at all? My assumption is that no, it doesn't, its out of the signal path. Does it set the Offset, and what exactly does that mean? How does changing the value change things?

[cliffforrest]

Your first paragraph is confusing. You seem to be mixing up caps and op-amps!

If that is just Christmas cheer, then OK, have another one and come back in the morning!

Your second paragraph reveals that you need to read up and study about op-amp gain, stability and compensation before you start swapping stuff with no knowledge. Otherwise it WILL end in tears.

Read the books/net first.

[pjpoes]

Haha, thanks. I wasn't actually drinking, I've been quite sick lately. Home from work while I recover, and I haven't been thinking straight at all. I was looking at caps online for a project and writting this at the same time. Yes I ment opamps.

Thank you for the response, I wasn't really planning on swapping values, I simply was curious. I tried looking around on the net and continued to be a bit confused. None the less, I will stick with the value that was there. Thanks.

[paulb]

Yes, I think you have the pins backwards. See
http://focus.ti.com/lit/ds/symlink/opa627.pdf page 3.
Not sure why the 22 pF cap is there. Are you sure it connects to the trim pin? It would make more sense in the feedback network somewhere.

[AndrewT]

Hi,
an 8 pin DIL
has a dimple or cut-out on the left with pins facing away.
bottom left is 1, bottom right is 4, top right is 5, top left is 8.

So, the dimple/cut-out is between 1 & 8.

A dual pin out is different from a single.
The 627 is single with +V to pin7 and -V to pin4

[pjpoes]

Thanks guys, so I looked at the original opamp in that location, which was an NE5534NA, and its 1 and 4 are balance and V-, where as the OPA627 has offset trim and V-. Is Balance and Offset trim essentially the same thing? I also noticed that the NE5534NA doesn't have a not connected lead, however, there is no pad on the PCB, so its not connected.

So you have no idea what a 22pf ceramic cap between pins 1 and 4 would do?

[pjpoes]

wait I want to change that comment. Ok looking from it as you guys said to, I think it had a cap originally between 8 and 5, which on the NE was two of the Ballance pins. However pin one is not connected, so I believe it now serves no purpose. However, that does mean that the Offset trim is now connected to nothing for the BB. Was this not a good replacement? do I need to connect 1 to 5 with that cap now? Thanks.

[AndrewT]

Hi,
5534 only works for high gains. It is not compensated for low gain.

However the datasheet shows a 22pF comp cap to allow 5534 to work at unity gain.

Where exactly was the 22pF? Could it have been unity gain compensation?

5534 also has unusual pin out that few others copied.

627 is not normally compatible, but with pin 8 NC then it may work.

Add bypass at the supply pins when changing from 5534 to 627.

I wish you had given us the whole story at the beginning.

[pjpoes]

Terribly sorry, I didn't realize the pinouts were unique like that. Yes I do believe that its for unity gain compensation.

I had asked on these forums of a good replacement for that opamp, and the 627 was recomended. I decided to try it, never even thought there might be a difference, and it does work ok.

A bunch of people have suggested the bypass cap at the leads. First, do you have a recomended value? Any chance these extra 33pf wima's I have would, or do I want something of higher value? Let me know. Also, I know a lot of times companies use ceramic caps for this purpose, and given the size issue, would such a cap be adequate, or possibly a tantulum? Thanks.

[AndrewT]

Hi,
what does the manufacturer say?

Ceramic should be used very close to the chip, right on the pins if possible and with the very shortest of leads.

If the manufacturer also specifies a larger cap, then Tantalum might be suitable a little further away with slightly longer leads to fit around the ceramics and to reach the power ground (NOT the signal ground, if there is one).

Sometimes only one side of the supply is bypassed, or even just V+ to V-, it depends on what currents flow internally.