As I don't have the proper tools to measure the input impedance Ill just assume it is "low" for now. Just got into electronics more in depth a couple weeks ago.
So regarding these 2 "experiments" I've been asking about. So swapping the pot from 20k to 100k and R3/R4 from 220k to 1m actually just results in increased preamp input impedance (which is desirable), right? From what I understand the pot and the resistor combines a voltage divider and the divider can be reasoned using the resistance ratios which in this case are almost equivalent - 1/10, 1/11.
Regarding the instabilities that can occur due to swapping R2 and R8 to achieve a lower output impedance. Without having any proper tools such as generators and scopes, is there something that I can observe without them to assess the results?
Another thing I was wondering. If the jfets have to be matched to achieve proper operation, would it make sense to ensure that the pairs are working in similar thermal environments? I mean jfets per channel tighly squeezed back to back together and applied thermal paste in between and wrap some copper foil around them. Or am I just overthinking this a lot ?
100k pot is easier on wimpy sources i.e. of high output impedance and small signal. No modern sources are like that. Unless its about some special situation. Increasing the pot's value adds more resistance noise and closes in the bandwidth because higher impedance drives the K170's 30pF input capacitance and any extra pF from wires and layout. Longer RC time filter.
Fortunately with DCB1 total input capacitance is small enough and there is no gain thus no Miller effect i.e. no multiplication of pF by the gain. Yes, to host a 100k pot, 1Meg DCB1 input resistors are the choice. JFets have very high input impedance on their own because of tiny gate current so they don't compromise the 1Meg input impedance setting. K170 has only about 10pA gate current leak at 10V rails used here.
About instability clues with no scope, you may hear buzz or whistling on the speakers and/or lame sound quality. Much static from a small portable AM radio when bringing it close for a test. That noise mostly goes away when powering off the offending circuit.
Millivolts DC offset will be your criterion about successful matching and application of the signal Jfets. Down to 1-2mV at the audio outputs has been achieved with the DCB1 as it is without bonding the parts. They are not as touchy as differential pairs, so their working maxed out at Idss by design should be enough to behave close.
Fortunately with DCB1 total input capacitance is small enough and there is no gain thus no Miller effect i.e. no multiplication of pF by the gain. Yes, to host a 100k pot, 1Meg DCB1 input resistors are the choice. JFets have very high input impedance on their own because of tiny gate current so they don't compromise the 1Meg input impedance setting. K170 has only about 10pA gate current leak at 10V rails used here.
About instability clues with no scope, you may hear buzz or whistling on the speakers and/or lame sound quality. Much static from a small portable AM radio when bringing it close for a test. That noise mostly goes away when powering off the offending circuit.
Millivolts DC offset will be your criterion about successful matching and application of the signal Jfets. Down to 1-2mV at the audio outputs has been achieved with the DCB1 as it is without bonding the parts. They are not as touchy as differential pairs, so their working maxed out at Idss by design should be enough to behave close.
That fixed it! 1.75v across Rset, 3.9v across both irfp240pbf and 0.64v over bc560
He, he 😉
If it also has about +/- 10V PSU voltage rails and no more than 5mV DC offset on each audio output, it is ready for use.
If it also has about +/- 10V PSU voltage rails and no more than 5mV DC offset on each audio output, it is ready for use.
9.95V and -9.98V respectively. Can I just measure the 5mV on the board outputs I don't have to wire it to the RCA sockets? Oh and if I ever get to Athens I owe you a beer!
Very low offset is nice to see, given all subsystems worked. The relay actually clicked and it was true DC offset not a digital meter's mV residual on disengaged signal outputs. Des, please listen close to the board, verify that the output relay clicks just a bit after power on.
Hi all, looking to start one of these and I have some questions. I would say my experience level is still beginner although I was able to successfully assemble a Tubelab SSE which works so I learned a lot during that process.
-is this buffer a good match for the SSE? Input impedance is 220k ohm.
-is it worth it to try the hot rod version straight away or should I build a standard version first?
-I’m looking for 2-3 inputs and 2 variable outputs, is that easy to configure?
-I’m also looking for remote control volume, I’ve seen motorized 20k alps pots out there, but not sure how to implement.
-for the standard version are chassis heat sinks required? I’m planning to DIY my chassis from aluminum, I’d like the footprint to be 11”W x 13”D
Edit: a couple more:
-is there an updated BOM or is the one on Diyaudiostore (2018) the most current?
-is 2” u channel aluminum tall enough?
-is this buffer a good match for the SSE? Input impedance is 220k ohm.
-is it worth it to try the hot rod version straight away or should I build a standard version first?
-I’m looking for 2-3 inputs and 2 variable outputs, is that easy to configure?
-I’m also looking for remote control volume, I’ve seen motorized 20k alps pots out there, but not sure how to implement.
-for the standard version are chassis heat sinks required? I’m planning to DIY my chassis from aluminum, I’d like the footprint to be 11”W x 13”D
Edit: a couple more:
-is there an updated BOM or is the one on Diyaudiostore (2018) the most current?
-is 2” u channel aluminum tall enough?
Last edited:
- Yes, as it has about 250 Ohm source impedance
- The standard "Ten Years After" version is already hot rod (not extreme but enough)
- 2-3 inputs yes, you can configure as many you like up to six on a Mezmerize. Variable outputs you need to implement on some extra proto-board you will have to devise yourself. That will create more output impedance though. Up to 1/4 the nominal value of each of your output pots. Var outputs will change the normal straight out Mez performance of course.
- Look on eBay etc. for remote volume controller kits
- No, you may use the chassis floor alone for cooling because the dissipation is moderate.
- 2018 BOM generally holds true unless you can't find some BC transistors which recently went EOL or the odd back-ordered part but look in the previous pages for close alternatives suggested. If unsure, ask again here after checking for what's currently available in the stockists.
- 2U chassis height is fine
This is helpful thank you.
Re: outputs, I’ve seen other builds where there are a 2nd set of outputs that just have 220ohm resistors between them and the first set, does that not provide a 2nd output that varies with volume?
Apologies for the basic questions I’m trying to learn as I go
Those are fixed parallel outputs for connecting yet another power amp say for bi-amplification or an active subwoofer etc.
By the way its better to feed signal for the second set of output resistors from the input side of the first set, not making them additive to those
Alright then, fixed secondary outputs are easy to implement. Active subwoofers have volume controls of their own.
Oh you mean they are FIXED, I see. Then that won’t work. I want them to vary when I change the volume with the same output level as the main set.
It won’t work to simply run a 2nd set of wires from the output on the board, will it?