I don´t have the answer,but be careful with the OPA637.kasra said:Im now in the asembling fase of my sulzer regulators and have some thoughts.
I have around a few opa627BP and opa637AP which i thought i might use for the sulzer regulators?
Perhaps any advantages/disadvantages between opa627/637 vs. AD797/AD817(ADs recomended by Jang/Didden)?
/kasra
It is not stable with gain less than 5.
I use AD817 in my Jung regulators and it works great...
Nicke
What to look for?
Ok, here's a potted summary from my experience having built about 30 of the Jung / Didden types, in my order of importance, and assuming bootstrapped op-amp supply...
1. Sonics
2. Open loop bandwidth / response shape / gain
3. Dynamic behaviour (speed, settling times etc.)
4. PSRR
5. Noise
6. Stability
The only thing that ultimately matters is the sonic performance, the AD817 and AD825 are, IIRC, quite similar in many regards, but sound totally different.
The AD817 looks slightly better on paper - it slews faster, has better settling times, wider bandwidth, open loop gain is higher, ol b/w is similar to the AD825, but in my opinion and that of many others who've tried it the AD825 sounds better, and is what I use primarily.
I've got some samples lined up that look potentially better than the AD825, but that's a secret for a select few for the moment
Andy.
P.S. The stability of the AD825 equipped reg's is wonderful too - I've not had a single unit oscillate, ever.
I couldn't say that for the AD797, which could be a total bitch to get stable!
P.P.S. Your choice of device could also depend upon application...
Ok, here's a potted summary from my experience having built about 30 of the Jung / Didden types, in my order of importance, and assuming bootstrapped op-amp supply...
1. Sonics
2. Open loop bandwidth / response shape / gain
3. Dynamic behaviour (speed, settling times etc.)
4. PSRR
5. Noise
6. Stability
The only thing that ultimately matters is the sonic performance, the AD817 and AD825 are, IIRC, quite similar in many regards, but sound totally different.
The AD817 looks slightly better on paper - it slews faster, has better settling times, wider bandwidth, open loop gain is higher, ol b/w is similar to the AD825, but in my opinion and that of many others who've tried it the AD825 sounds better, and is what I use primarily.
I've got some samples lined up that look potentially better than the AD825, but that's a secret for a select few for the moment
Andy.
P.S. The stability of the AD825 equipped reg's is wonderful too - I've not had a single unit oscillate, ever.
I couldn't say that for the AD797, which could be a total bitch to get stable!
P.P.S. Your choice of device could also depend upon application...
Stability
One thing to remember for non-unity gain stable op-amps is it's the noise gain of the circuit that's relevant for stability, not the DC gain.
As for the AD797, the load presented to it has a big impact upon stability - powering digital circuits often leads to instability, without further measures (lowering feedback bandwidth for example).
Steady-state noise is rarely, if ever, the most important PSU characterstic sonically.
Andy.
One thing to remember for non-unity gain stable op-amps is it's the noise gain of the circuit that's relevant for stability, not the DC gain.
As for the AD797, the load presented to it has a big impact upon stability - powering digital circuits often leads to instability, without further measures (lowering feedback bandwidth for example).
Steady-state noise is rarely, if ever, the most important PSU characterstic sonically.
Andy.
Andy, as a start, would a Jung type (or sulzer) regulator with a TL081 make sense ? These have low offset, compared to TL071s as I have learned just recently, but maybe slightly more noise. Does it make sense to make a Sulzer like regulator with 4 outputs using a single TL084 (for a NAC42.5 f.e.)?
Klaus
Klaus
sulzer etc
If you use a quad TL084 for a 4-channel reg you cannot use the bootstrap because there is only one supply pin for the 4 opamps. You can, if the output voltages are in the right range, use the supply of just one reg for the whole opamp. It would be interesting to see if there is any cross-coupling between reg outputs because of this.
To play it safe, I would use just a single 081 for each channel. Or, since we are talking cheap opamps, why not the original 5534? The sulzer was very good with that one, and they are less than a dollar apiece nowadays.
As already noted by other posters as well, I would not recommend the 797 unless you are really good at fighting oscillations or ringing.
Jan Didden
If you use a quad TL084 for a 4-channel reg you cannot use the bootstrap because there is only one supply pin for the 4 opamps. You can, if the output voltages are in the right range, use the supply of just one reg for the whole opamp. It would be interesting to see if there is any cross-coupling between reg outputs because of this.
To play it safe, I would use just a single 081 for each channel. Or, since we are talking cheap opamps, why not the original 5534? The sulzer was very good with that one, and they are less than a dollar apiece nowadays.
As already noted by other posters as well, I would not recommend the 797 unless you are really good at fighting oscillations or ringing.
Jan Didden
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.