Only thing with these high bandwidth fast op-amps is that they need good decoupling.
I found the OPA627 a little soft too but hopefully it may be better with the UCD
Is there only Bruno who uses a discrete input stage with the UCD's?
I found the OPA627 a little soft too but hopefully it may be better with the UCD
Is there only Bruno who uses a discrete input stage with the UCD's?
I wonder, what about just making that input stage just a buffer for impedance matching, having the opamp set to unity gain, and then raising the gain almost double the gain of the transistor output. That too could get squirrelly, but justthought I'd throw that out there.
Ray
Ray
What would be the benefit of unity gain versus a gain of two? As far as I can see it would only limit the possible upgrade candidates for the opamp as several of them are not unity gain stable.ray bronk said:
who cares. The idea is to get rid of the frontend stage altogether. The opamp at unity gain would only serve as an impedance matcher. Plus less of any effects because of the opamp itself. wonder if that could be done. Just a thought.
The best way of course is to develop a descrete frontend and stop messing around with these opamps.
Ray
The best way of course is to develop a descrete frontend and stop messing around with these opamps.
Ray
Ray,
Just to be real here, you're saying that your two cents worth is what someone else on some other forum said.. and you've heard neither the UCD or the opa627 yourself?
Please, do show us your discrete design that you think will outperform it.
.... My two cents worth.
Just to be real here, you're saying that your two cents worth is what someone else on some other forum said.. and you've heard neither the UCD or the opa627 yourself?
Please, do show us your discrete design that you think will outperform it.
.... My two cents worth.
Hi Chris,
You must be joking. First of all, I was told this by someone who I trust, and this person has played around with a lot of chips. That chip has been around for awhile.
As for the discrete circuit, if anyone could come up with one, it would be you. It does seem here in this forum, all we are doing is playing around with chips with various sonic signatures. That's why I suggested we've got enough engineers and techy types around here, so that collectively we could or you guys could come up with a discrete circuit to replace the chips. That's all I have.
Ray
You must be joking. First of all, I was told this by someone who I trust, and this person has played around with a lot of chips. That chip has been around for awhile.
As for the discrete circuit, if anyone could come up with one, it would be you. It does seem here in this forum, all we are doing is playing around with chips with various sonic signatures. That's why I suggested we've got enough engineers and techy types around here, so that collectively we could or you guys could come up with a discrete circuit to replace the chips. That's all I have.
Ray
op-amp biasing
Ric
I now see I have mis-written my earlier post. What I should have said is - biasing from the positive rail turns the output stage into a SE class A PNP emmitter follower.
I think the mechanism is that when a pull-up/ down resistor or ccs is used on the output the devices on the opposite rail pull the output to 0 volts by switching on to match the bias current through the R or ccs. This turns the output stage from class b push pull to class A se.
Now in relation to the 2134, the datasheet for that shows its powersupply rejection is at least 20db WORSE on the positive rail than the negative rail.
This leaves the choice between biasing from the negative rail and having a Class A NPN output stage with inferior PSRR and biasing from the positive rail giving a PNP output stage with superior PSRR.
Maybe what sounds best depends on how clean the power supply is?
Rob.
Ric
I now see I have mis-written my earlier post. What I should have said is - biasing from the positive rail turns the output stage into a SE class A PNP emmitter follower.
I think the mechanism is that when a pull-up/ down resistor or ccs is used on the output the devices on the opposite rail pull the output to 0 volts by switching on to match the bias current through the R or ccs. This turns the output stage from class b push pull to class A se.
Now in relation to the 2134, the datasheet for that shows its powersupply rejection is at least 20db WORSE on the positive rail than the negative rail.
This leaves the choice between biasing from the negative rail and having a Class A NPN output stage with inferior PSRR and biasing from the positive rail giving a PNP output stage with superior PSRR.
Maybe what sounds best depends on how clean the power supply is?
Rob.
Good stuff Rob.
The reason behind a particular rail having 20db worse psrr is dependant upon how the miller compensation is internally referenced.
With the opa's it is to the positive rail, so you stand to gain more by going SE PNP and having the improved psrr.
The cleaner the supply, the less of a concern PSRR becomes, but I think you still stand to gain by having things as symmetrical as possible (and trying to pretend you have the world's cleanest auxiliary supply).
The reason behind a particular rail having 20db worse psrr is dependant upon how the miller compensation is internally referenced.
With the opa's it is to the positive rail, so you stand to gain more by going SE PNP and having the improved psrr.
The cleaner the supply, the less of a concern PSRR becomes, but I think you still stand to gain by having things as symmetrical as possible (and trying to pretend you have the world's cleanest auxiliary supply).
Class A
Hi Ray,
I don't know if I can provide greater clarity without getting into the difference between PNP's and NPN's. The latter usually have better spec's than the complimentary PNP. Hence my talk about a possible trade off.
The mechanism for equalising the output voltage in compensation for the Resistor or CCS is Feedback.
Chris thanks for explaining the cause of the disparity in PSRR between rails.
Rob.
Hi Ray,
I don't know if I can provide greater clarity without getting into the difference between PNP's and NPN's. The latter usually have better spec's than the complimentary PNP. Hence my talk about a possible trade off.
The mechanism for equalising the output voltage in compensation for the Resistor or CCS is Feedback.
Chris thanks for explaining the cause of the disparity in PSRR between rails.
Rob.
ray bronk said:
Have the UCD400 AD with the AD8620 set to unity gain. Works just fine, I needed less gain due to high level from the preamp. Just remove the 560ohm resistor near the opa.
I'd argue that hardly constitutes as removing the op amp. "getting rid of the front end stage"
Classd4sure
Question:
You previously wrote: "I'm not playing with the old 180's anymore at this time (got plans for them though). V6.1 UCD400 on the old 180 stereo supply"
Since your running your new UCD400AD modules off of your previous UCD 180 power supply, just curious how many volts DC your running them on?
Thanks
Question:
You previously wrote: "I'm not playing with the old 180's anymore at this time (got plans for them though). V6.1 UCD400 on the old 180 stereo supply"
Since your running your new UCD400AD modules off of your previous UCD 180 power supply, just curious how many volts DC your running them on?
Thanks
Coupling Caps Again
Hi All,
Please forgive me for beating a dead horse. I have read through all of the pages of the UCD400, UC180, and the "Hotrodding" postings. I am new to tweaking and just want to make sure I am clear on the instructions. I have a 4 module UcD180AD with 2 ST power supplies and a 800 VA toroid. I am interested in removing the coupling caps to improve the sound. These appear to be C7 and C8 on my modules and are rated at 22 uF at 63V. If I understand correctly, I can remove both of these, jumper the pins, and place a 2.2 uF 450VDC Auricap in series with the positive wire of the RCA inputs. Does replacing both of the 22 uF caps with the one 2.2 uF 450VDC Auricap give the equvalent DC protection to the speaker output? Are other caps or values recommended? I have not checked my preamp DC offset, but regardless I would like to maintain some DC output protection.
Mooglie
Hi All,
Please forgive me for beating a dead horse. I have read through all of the pages of the UCD400, UC180, and the "Hotrodding" postings. I am new to tweaking and just want to make sure I am clear on the instructions. I have a 4 module UcD180AD with 2 ST power supplies and a 800 VA toroid. I am interested in removing the coupling caps to improve the sound. These appear to be C7 and C8 on my modules and are rated at 22 uF at 63V. If I understand correctly, I can remove both of these, jumper the pins, and place a 2.2 uF 450VDC Auricap in series with the positive wire of the RCA inputs. Does replacing both of the 22 uF caps with the one 2.2 uF 450VDC Auricap give the equvalent DC protection to the speaker output? Are other caps or values recommended? I have not checked my preamp DC offset, but regardless I would like to maintain some DC output protection.
Mooglie
It depends how you look at it, but no it doesn't really give you DC protection.
If you have high DC offset from your source, it helps ensure that DC does not get amplified to a dangerous level, so you're good from that aspect. If your source has 1.2mVdc of below, you don't need the cap at all. You don't have to use an Auricap, any good signal cap will do.
Now if you should be so unlucky as to have some failure of the output stage for whatever reason, the resulting DC at the output of the amp will damage your speaker, unless you have some other means of protection at that end, like a relay to shut the supply off, or a fuse.
If you have high DC offset from your source, it helps ensure that DC does not get amplified to a dangerous level, so you're good from that aspect. If your source has 1.2mVdc of below, you don't need the cap at all. You don't have to use an Auricap, any good signal cap will do.
Now if you should be so unlucky as to have some failure of the output stage for whatever reason, the resulting DC at the output of the amp will damage your speaker, unless you have some other means of protection at that end, like a relay to shut the supply off, or a fuse.
Ok. Well, the ST power supplies have "two channels of DC protection for helping to prevent catastrophic failure. If one of the amplifiers were to fail the amplifier could short directly to one rail of the power supply. The fuses on this power supply would help protect speakers from possible damage. " So, does the mean I'm covered for the failure mode that you describe? Thus, leaving me only to worry about the DC offset of my preamp? I'll try to check the DC offset on my preamp this weekend, but expect it should be very low. Besides the Auricap that I mentioned, what uF and voltage rating should I consider for the film cap?
Many Thanks,
Mooglie
Many Thanks,
Mooglie
Re: Coupling Caps Again
I certainly may be wrong here, but wouldn't you need to replace two caps on each module?
Mooglie said:
I certainly may be wrong here, but wouldn't you need to replace two caps on each module?
Nope, because he specified RCA, single ended.
Higher voltage seems best, film, good quality. Yeah if you're supply covers you, you're on the ball as far as protection goes 🙂
Higher voltage seems best, film, good quality. Yeah if you're supply covers you, you're on the ball as far as protection goes 🙂
- Status
- Not open for further replies.