Input oscillation (Pin2) OPA2604 & OP275
I have replaced NE5532 with either OPA2604 or OP275 in an Tascam M5000 recording console (AUX Master Section). scope shows no oscillation @ outputs but indicates oscilations @ pin 2 (5).
with no oscillation at pin1 or pin7, is feedback oscillation an issue @ inputs (pin2/5)? (scope lead is grounded to signal ground)
also note that the input(s) are not strapped to kill ramdom 'air-bourne' signals.
the Tascam M5000 was a mid level pro recording console in the late 90's ($50-65k) for a 40 x 40 In-line conmsole desk using 100% NE5532 opamps. i am going through the console as a project replacing most caps and the 5532's with OP275 and/or OPA2604's. tyhe OPA2604 works ok in most locations (no oscillation at outputs. BUT as this thread states, my tech chops are not keen yet and not sure about feeedback oscialltions???
who is up speed on mid to high end analog large format recording console mods? i would dig an extended commucation on this subject(s). BTW: is anyone located near Statesboro, GA USA?
THX.
I have replaced NE5532 with either OPA2604 or OP275 in an Tascam M5000 recording console (AUX Master Section). scope shows no oscillation @ outputs but indicates oscilations @ pin 2 (5).
with no oscillation at pin1 or pin7, is feedback oscillation an issue @ inputs (pin2/5)? (scope lead is grounded to signal ground)
also note that the input(s) are not strapped to kill ramdom 'air-bourne' signals.
the Tascam M5000 was a mid level pro recording console in the late 90's ($50-65k) for a 40 x 40 In-line conmsole desk using 100% NE5532 opamps. i am going through the console as a project replacing most caps and the 5532's with OP275 and/or OPA2604's. tyhe OPA2604 works ok in most locations (no oscillation at outputs. BUT as this thread states, my tech chops are not keen yet and not sure about feeedback oscialltions???
who is up speed on mid to high end analog large format recording console mods? i would dig an extended commucation on this subject(s). BTW: is anyone located near Statesboro, GA USA?
THX.
Hi,
When you connect the ~12pF of your probe to the inverting input of the amplifier, you create a parasitic pole which can, under the right circumstances, give enough phase shift to create oscillations.
This is more likely with high bandwidth/ high impedance circuits, which is one of the reasons why both are rarely combined. Note that this should apply to pins 2 & 6, not 5.
LV
When you connect the ~12pF of your probe to the inverting input of the amplifier, you create a parasitic pole which can, under the right circumstances, give enough phase shift to create oscillations.
This is more likely with high bandwidth/ high impedance circuits, which is one of the reasons why both are rarely combined. Note that this should apply to pins 2 & 6, not 5.
LV
thx for replies . . .
i see that makes sense . . . similar to the fundamental Quantum Physics law that states loosely 'when you measure a quantity you have just changed the resulting measurment due to the intereaction(s) between matter and its forces about it . . . '!
with that in mind, basically just scope the output (pin1/7) for system oscillations, correct. the 8 opamps (16 circuits) in mind Master AUXs) all behave the same way so i should be ok with the OPA2604 or the OP275 in place of the NE5532. i prefer the OPA2604 as it is somewhat 'sweeter' even though it sports a slightly higher thermal noise number.
i have one cicuit in the Master L/R section that oscillates with the OPA2604 but is ok with a OP275. what is the best approach to mod the circuuit to allow the 2604 to run w/o oscillation @ output?
raise feedback caps to a smaller pF value??? is there a resistor to cap ration rule of thumb to follow?
the OPA2604 is more 'snakey' than a OP275 due to fast slew rates and faster freq limits i am told. that's about all the magic i know about these opamps, etc.
i am using the FC Panasonic elect caps in place of the OEM caps of '95 circa with 2x-4x the cap value increase in the new caps (newer caps have smaller foot prints these days!)
thx.
i see that makes sense . . . similar to the fundamental Quantum Physics law that states loosely 'when you measure a quantity you have just changed the resulting measurment due to the intereaction(s) between matter and its forces about it . . . '!
with that in mind, basically just scope the output (pin1/7) for system oscillations, correct. the 8 opamps (16 circuits) in mind Master AUXs) all behave the same way so i should be ok with the OPA2604 or the OP275 in place of the NE5532. i prefer the OPA2604 as it is somewhat 'sweeter' even though it sports a slightly higher thermal noise number.
i have one cicuit in the Master L/R section that oscillates with the OPA2604 but is ok with a OP275. what is the best approach to mod the circuuit to allow the 2604 to run w/o oscillation @ output?
raise feedback caps to a smaller pF value??? is there a resistor to cap ration rule of thumb to follow?
the OPA2604 is more 'snakey' than a OP275 due to fast slew rates and faster freq limits i am told. that's about all the magic i know about these opamps, etc.
i am using the FC Panasonic elect caps in place of the OEM caps of '95 circa with 2x-4x the cap value increase in the new caps (newer caps have smaller foot prints these days!)
thx.
correction on my part, i.e. pin 6 not pin 5.
if a OPA2604 oscillates @ OUTPUT and NE5532 and OP275 do not, where is a good point to start to kill the oscillation?
the circuits used by the Tascam M5000 are quite complex (not always clever and better) i was told by one @ Great River Electronics. i was able to remove most all the Input coupling elect caps and then increase the output elect caps by 2-3x value (FC Panasonic ) with success. the newer OPA2604 and OP275 are trimmed better and the off-set voltageis under better control (as i understand it).
peace,
billy
if a OPA2604 oscillates @ OUTPUT and NE5532 and OP275 do not, where is a good point to start to kill the oscillation?
the circuits used by the Tascam M5000 are quite complex (not always clever and better) i was told by one @ Great River Electronics. i was able to remove most all the Input coupling elect caps and then increase the output elect caps by 2-3x value (FC Panasonic ) with success. the newer OPA2604 and OP275 are trimmed better and the off-set voltageis under better control (as i understand it).
peace,
billy
-IN is a very sensitive point. Touching it with probe tip will make it oscillate.
You may want to read this about OPA2604
http://www.diyaudio.com/forums/showthread.php?postid=1051652#post1051652
You may want to read this about OPA2604
http://www.diyaudio.com/forums/showthread.php?postid=1051652#post1051652
Interesting -- I have used OPA604 both as replacement for NE5534 in a Micromega Stage CD player, after careful listening tests ( this was the best sounding I found at the time ), and used the OPA2604 in a preamp design as well. No stability issues at all came to light, but they were correctly decoupled and the feedback network compensated accordingly.
We tend to think OpAmps can just be swapped with impunity but it's not always the case. To start altering a design to make it work with just one device may not be advisable I.M.O. Is it trying to drive a slightly capacitive load -- add say 47 ohm series output resistor to pin 6 to isolate -- thats a common problem. There are so many possible reasons why ! You may need to add series resistance to the inputs, but this is all changing the whole concept of the original design. Adding a 0.1 ceramic accross pins 7 and 4 may be worthwhile trying as well.
I still think they are one of the best for audio
Edit, Stating the obvious but you also have to remember the OPA and NE devices are like chalk and cheese. One is a Bjt device and the other FET input. If the circuit relys on characteristics of one it's unreasonable to expect it to work with characteristics of the other. Mostly we get away with it
We tend to think OpAmps can just be swapped with impunity but it's not always the case. To start altering a design to make it work with just one device may not be advisable I.M.O. Is it trying to drive a slightly capacitive load -- add say 47 ohm series output resistor to pin 6 to isolate -- thats a common problem. There are so many possible reasons why ! You may need to add series resistance to the inputs, but this is all changing the whole concept of the original design. Adding a 0.1 ceramic accross pins 7 and 4 may be worthwhile trying as well.
I still think they are one of the best for audio
Edit, Stating the obvious but you also have to remember the OPA and NE devices are like chalk and cheese. One is a Bjt device and the other FET input. If the circuit relys on characteristics of one it's unreasonable to expect it to work with characteristics of the other. Mostly we get away with it
Since the 2604 is unity gain stable, you can try a small feedback cap between the output and the (-)input.grievousangel said:
The value will depend on the impedance seen at this node. Typically, it will be in the region of some pf to tens of pf.
Try to increase the value until the oscillations cease, and use a definitive cap of about twice that value.
Re: Input oscillation (Pin2) OPA2604 & OP275
http://www.gearslutz.com
http://www.prodigy-pro.com/forum/
http://recording.org/
http://studioforums.com/eve/forums
etc.
You could also surf into pro forums likegrievousangel said:
http://www.gearslutz.com
http://www.prodigy-pro.com/forum/
http://recording.org/
http://studioforums.com/eve/forums
etc.
Hi, Mooly,
Good points.
Very close supply decoupling, resistive output damper, close+carefull layout of -IN are often forgotten.
This post in that thread also very usefull :
http://www.diyaudio.com/forums/showthread.php?postid=1052604#post1052604
Good points.
Very close supply decoupling, resistive output damper, close+carefull layout of -IN are often forgotten.
This post in that thread also very usefull :
http://www.diyaudio.com/forums/showthread.php?postid=1052604#post1052604
And if you asked Bruno Putzeys over at http://recforums.prosoundweb.com/index.php/i/6/0/ he'd quite probably tell you that a global replacement of all 5532's with 2604's isn't a good idea at all, sound-wise... but that's only one opinion, though...
- Klaus
- Klaus
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.