OPA2134 will fit. Others have mentioned LF412 being a better choice for servo duty and I think it's a drop in replacement
Nice!
I guess the new one is perfect
The new layouts are working out very well but they may improve again once the audio analyzer group buy ships. At present it's an hour trip to visit Anatech to take measurements but he's also a great wealth of knowledge. http://www.diyaudio.com/forums/equi...alyzer-ak5397-ak5394a-ak4490.html#post4405373
AnTek Xfmrs
Seriously, I never checked my 2 AS4445, are the labels incorrect or windings? My labels are 2 x 15.
Thx
MI
Seriously, I never checked my 2 AS4445, are the labels incorrect or windings? My labels are 2 x 15.
Thx
MI
The labels on my transformers matched the windings. The website listed them as 2x15VAC aux windings but the transformers said 12 & 18VAC.
the 2134 will pin swap for an 072.
The 2134 will give offsets about 20% of what a typical 072 can achieve, but at 20 times the cost.
the 412 will not be quite as good as the 2134 but it will be less than half the cost.
I have all three on the shelf waiting for a suitable amplifier to require it.
But most of my amplifiers manage offsets over a reasonable temperature range of <+-5mVdc. They are all AC coupled at both input and NFB.
The 2134 will give offsets about 20% of what a typical 072 can achieve, but at 20 times the cost.
the 412 will not be quite as good as the 2134 but it will be less than half the cost.
I have all three on the shelf waiting for a suitable amplifier to require it.
But most of my amplifiers manage offsets over a reasonable temperature range of <+-5mVdc. They are all AC coupled at both input and NFB.
We're talking about servo applications, right?
Then why 412 is not as good as 2134? 412A has got pretty much the same offset characteristics as 2134, jFET input stage (high input impedance), etc - more than enough for instrumentation applications, like servo. Even 412 with no "A" is very good.
2134 is excellent in audio applications - that's no question.
Then why 412 is not as good as 2134? 412A has got pretty much the same offset characteristics as 2134, jFET input stage (high input impedance), etc - more than enough for instrumentation applications, like servo. Even 412 with no "A" is very good.
2134 is excellent in audio applications - that's no question.
DC servo performance is determined mostly by input offset and drift of input offset.
Typical 412 is not quite as good as the much newer typical 2134 and the typical tl07x is much worse.
All three (134, 412, 07x) are the required very low input offset current.
typical values as follows
. . . . . . . . . 07x . . 412 . . 134
input offset . . 3 . . . 1 . . . 0.5mV
drift in offset. 18. . . 10 . . . 2uV/C
Typical 412 is not quite as good as the much newer typical 2134 and the typical tl07x is much worse.
All three (134, 412, 07x) are the required very low input offset current.
typical values as follows
. . . . . . . . . 07x . . 412 . . 134
input offset . . 3 . . . 1 . . . 0.5mV
drift in offset. 18. . . 10 . . . 2uV/C
Last edited:
In post 2046 I wrote: "Today I fired up the IPS with 100 ohm connected to NFB and ND-/PD+. I measured 0.61V over 100 ohm. Over R12/R13 I could easily set 3.8 mA."
Today I connected Spooky IPS with the OPS. R106 was turn to max(200 ohm).
Bias over one of the emitter: close to 100mV. This is far to high. What can I do to get the bias down. I remember in my first Slewmaster with the Kypton IPS bias were also to high. This was fixed changing R18/R19 on Kypton. Is it any "same cure" for the Spooky. My power is +/-67V DC.
In my Spooky R23 and R24 are 6.8 kohm. I noticed today in another post that it was recommended to change to 3.9 kohm. What is correct according to my +/-67V power?
Does this have any influence on my high bias??
Eivind S
Today I connected Spooky IPS with the OPS. R106 was turn to max(200 ohm).
Bias over one of the emitter: close to 100mV. This is far to high. What can I do to get the bias down. I remember in my first Slewmaster with the Kypton IPS bias were also to high. This was fixed changing R18/R19 on Kypton. Is it any "same cure" for the Spooky. My power is +/-67V DC.
In my Spooky R23 and R24 are 6.8 kohm. I noticed today in another post that it was recommended to change to 3.9 kohm. What is correct according to my +/-67V power?
Does this have any influence on my high bias??
Eivind S
R106 is 500 ohm now. Needed a little break, so I have no message to return so far. Hope that the bias will be more correct.
Vrystaat and I have sent each other some PM and discussed some alternatives to some resistors values. It is about Spooky.Some posts operate with alternative value on R23/24 to be 3.9kohm. Schema(and Jkuetermans lay out) say 6.8 kohm. R14-R17; here schema and the lay out say 47 ohm. Alternative say 68 ohm.
My power is +/- 67V DC. What will be the best alternative in my case?
Let me tell that the red led in my Spooky light up very dim and not before the voltage is on top. Any connection here to my question about alternative resistors values?
Eivind S
Vrystaat and I have sent each other some PM and discussed some alternatives to some resistors values. It is about Spooky.Some posts operate with alternative value on R23/24 to be 3.9kohm. Schema(and Jkuetermans lay out) say 6.8 kohm. R14-R17; here schema and the lay out say 47 ohm. Alternative say 68 ohm.
My power is +/- 67V DC. What will be the best alternative in my case?
Let me tell that the red led in my Spooky light up very dim and not before the voltage is on top. Any connection here to my question about alternative resistors values?
Eivind S
If the idle current will still be higher than expected with R106 = 500R, when the trimmer is in the highest R value position (which I don't expect to be the case, but still) - slightly increase the value of the resistor in series with the trimmer. Higher overall R (together with the trimmer) -> lower idle current.
I got this bias results over the different five emitter resistors (0.22 Ohm):
9.6 mV
11.2 mV
11.1 mV
18.6 mV
13.5 mV
My goal was 15 mV = 30 mV over a emitter pair.
What will be a recommended "middelvalue". 9.6mV is a little bit to low, but bringing it up to 15 mV might bring the 18.6 mV up to high????( I am not sure). As it is now, the heatsinks is not warm. It is the biggest heatsink from HiFi 2000; 400mmx210mm, so I think it can handle a lot more. Any good advice?
Eivind S
9.6 mV
11.2 mV
11.1 mV
18.6 mV
13.5 mV
My goal was 15 mV = 30 mV over a emitter pair.
What will be a recommended "middelvalue". 9.6mV is a little bit to low, but bringing it up to 15 mV might bring the 18.6 mV up to high????( I am not sure). As it is now, the heatsinks is not warm. It is the biggest heatsink from HiFi 2000; 400mmx210mm, so I think it can handle a lot more. Any good advice?
Eivind S
9.6mV and 18.6mV don't look good to me. Did you sort and match the power transistor Vbe and Hfe? To make a powerful amp that requires 5 pairs I would match the power transistors to ensure an even current sharing.
You will always get more accurate results if you measure the voltage not over a single resistor, but over two resistors in each pair (although those pairs segregation is nominal - any npn + pnp may be considered as a pair - you can measure the ones opposed to each other). Set 33-35mV as an average voltage for those pairs and you'll be fine.
The heatsink must be moderately warm after 20-30 minutes at idle.
The heatsink must be moderately warm after 20-30 minutes at idle.