You are probably right that spice models are far from reality, I just measured the LSK170C model with G and S grounded and Vds 10V showed 15.8mA while LSK170B 8.4mA. At least this was OKSimulated Idss can be anything
If you decide to buy k170v JFETS, choose those with 13-15ma Idss and put them into circuit instead of BF862 without changing anything and you'll be fine (they are that similar).
I'm a beginner with spice and while probably not perfect, these models I hope at least to give a hint about how will they behave in real.
Its good to know that are that similar, however, I have doubts that without changing anything I get the 14x time amplification I need. I plan to try with both and decide after which stays anyway
If I have to lower the voltage I could do it with series diodes or leds, so the PSU question is resolved
I'm thinking how could I modify it to be DC coupled...
I really don't want to deviate from this thread which was started for this 2x gain preamp, but, could you help me to modify it to have 14x gain driven by 140 mV rms input? (I could start another thread or in private...)
I thought that the modification I wrote in #203 are enough to get 14x with certainly higher, 0.1%, THD. I did some calculations on paper and the values I wrote there seemed to fit for a 14-15mA 2SK170V.
Thanks anyway.
The input signal level is 140mV rms (current output of AD1865N, +-1mA through a 200R).
As volume control after the preamp I use an R2R volume control, which has a variable input impedance of 25K - 90K and an output impedance ~10K (I will add a B1 buffer after the volume control).
I want to drive a Quad405 (ala Keith Snook DCD-Mod3) which has an input impedance of 22k and needs 1V rms for full output.
I wanted the preamp to deliver 2V rms out to be able to test it with lower sensitivity amps as well, however, if massive modifications are needed 1V rms out is enough.
As the Quads I'm using are fairly neutral I'm looking to add some warmth to them but I don't want to use tubes, so I want to try a common source JFet preamp, that's why a 0.06 - 0.1% THD at full output is OK for me.
As volume control after the preamp I use an R2R volume control, which has a variable input impedance of 25K - 90K and an output impedance ~10K (I will add a B1 buffer after the volume control).
I want to drive a Quad405 (ala Keith Snook DCD-Mod3) which has an input impedance of 22k and needs 1V rms for full output.
I wanted the preamp to deliver 2V rms out to be able to test it with lower sensitivity amps as well, however, if massive modifications are needed 1V rms out is enough.
As the Quads I'm using are fairly neutral I'm looking to add some warmth to them but I don't want to use tubes, so I want to try a common source JFet preamp, that's why a 0.06 - 0.1% THD at full output is OK for me.
For that application I would choose another topology. Actually, what you need is a good I/V stage. You'll find plenty of them in digital line level/source sections.
Not that I'm telling you what to do, just trying to save you time
But if it had to be "this way or no way" I would make gain stage like this (attached sch.)
For negative PS (1.2V) I'd use rechargable battery cell and R3 should be 1W. Idss of Q1 is 15mA or so.
The buffer stage can be left unchanged.
Not that I'm telling you what to do, just trying to save you time
But if it had to be "this way or no way" I would make gain stage like this (attached sch.)
For negative PS (1.2V) I'd use rechargable battery cell and R3 should be 1W. Idss of Q1 is 15mA or so.
The buffer stage can be left unchanged.
Attachments
For the negative supply I think I stay with the -4.6V and use leds to drop the voltage to -1.2, but nevertheless I will try the battery first as it's easier to testFor that application I would choose another topology. Actually, what you need is a good I/V stage. You'll find plenty of them in digital line level/source sections.
Not that I'm telling you what to do, just trying to save you time
But if it had to be "this way or no way" I would make gain stage like this (attached sch.)
For negative PS (1.2V) I'd use rechargable battery cell and R3 should be 1W. Idss of Q1 is 15mA or so.
The buffer stage can be left unchanged.
The DAC End2 I build uses passive I/V with a single resistor so a good line level amp is all that I need. The original project uses SE tube stage for this.
Thanks for helping!
The recommendation on this thread is to put the volume pot in front of the BF862 Preamp. This means that the source needs to have a relatively low output impedance to counter the varying impedance of the preamp, caused by different settings of the pot.
My question is, why not use the buffer part as the input stage, followed by the pot and then the gain stage? This would be more on the lines of opamp based preamps, where the source is not strained, the dynamics are not inhibited by the attenuating pot (following voltage amplification - a kind of bottleneck scenario) and relatively low impedance power amps will also not have a problem. Your thoughts.
My question is, why not use the buffer part as the input stage, followed by the pot and then the gain stage? This would be more on the lines of opamp based preamps, where the source is not strained, the dynamics are not inhibited by the attenuating pot (following voltage amplification - a kind of bottleneck scenario) and relatively low impedance power amps will also not have a problem. Your thoughts.
Here you have a lot of wrong assumptions and it would be tiresome to correct them all - so, you are free to do what ever you like to do with this circuit, and I reserve the same amount of freedom for me
The input of the preamp has a very high impedance. Therefore there will be no variance to speak of due to different settings of the pot. For all intents and purposes, the source will simply see the end-to-end resistance of the pot.
se
Thanks Zen Mod, juma and Steve, I will try both configurations and check which sounds better. As Zen Mod rightly pointed out, every concept (and approach) is a compromise.
I have read Douglas Self on this issue in EW&WW and saw that he used an Active Volume Control in his preamp design. I also saw Jeff Mc'Caullay do the same thing in an article in EW&WW and tried the circuit. Everyone who heard it thought that it was fabulous.
Whether it is a preamp or a buffer, putting a pot before the active stage, may result in poorer SNR at lower pot settings, but it does avoid overload of the stage and an unnecessary attenuation of the signal. An active pot overcomes many of these problems. The next best thing, is probably a buffer (with a good overload margin), volume pot, followed by the gain stage. If one assumes that the position of the pot is not a problem sonically, then one needs to re-think as to why a series, shunt and LDR type of attenuator present different sonic signatures.
I have read Douglas Self on this issue in EW&WW and saw that he used an Active Volume Control in his preamp design. I also saw Jeff Mc'Caullay do the same thing in an article in EW&WW and tried the circuit. Everyone who heard it thought that it was fabulous.
Whether it is a preamp or a buffer, putting a pot before the active stage, may result in poorer SNR at lower pot settings, but it does avoid overload of the stage and an unnecessary attenuation of the signal. An active pot overcomes many of these problems. The next best thing, is probably a buffer (with a good overload margin), volume pot, followed by the gain stage. If one assumes that the position of the pot is not a problem sonically, then one needs to re-think as to why a series, shunt and LDR type of attenuator present different sonic signatures.
Try just a pot followed by a gain stage. Then if that creates a problem, you can then add the buffer in front of the pot.
J.
yes. The buffer added to the attenuator to allow the attenuator to properly drive the following cable and receiver.
no.
If the source cannot drive the pot then add an buffer to the source so that it is capable of driving the cable and pot (receiver).
Hi Andrew, you missed my underlying point, which is 'keep it simple to start with, and only consider complicating it when there is a clear problem that needs solving'.
I was also working on the assumption that the source would be reasonably low impedance, and if there was an issue it might be because the pot was too low a value, hence potentially improved by the addition of buffer in front of it. However if one works on the assumption that the source has a weak output stage and the pot is of a reasonable value, then your suggestion would of course be the preferred route. i.e. the solution depends upon the cause of the problem.
Regards,
J.