If that's true, you need a different pre. It seems the Saga is only meant to be a remote volume control, between the source and a pre with no remote control. Only another pre with gain would put out 5v of usefull signal. Why would anyone do that? Or, just rebuild the 142 input and FB back to stock with the right input impedance. Good luck with your project. Enjoy always.
Last edited:
It's a unity gain preamp. At least that's what it's called.
Also can be called a passive preamp if in passive mode.
I don't really agree with calling something without gain a preamp.
Also can be called a passive preamp if in passive mode.
I don't really agree with calling something without gain a preamp.
I agree with that definition. Look for a Numark M4 to give you a bunch of good input options and some gain to put in front of the Saga. Or behind it.
Opinion alert
On and on it goes, more and more complicated. How many stages are we up to? We're a long way from elegant and clean "less is more". You seem to care about performance, but are choosing ever more convoluted solutions that are not big wins, and losing quality with each addition. If only you had a stereo amp, you could cut out the pre-amp, and the buffer, and use your sources directly through the Schitt Sys. Elegant. (5K driving into 47K) What are your sources? Some devices let you select a mono output. BTW- you could easily use the stereo connector on your amp, do a summing network with two 47K resistors on the amp, (solves a lot of problems, makes the amp much more versatile in a stereo world) and build your upstream infrastructure to handle stereo for when you eventually get tired of mono and get some smaller speakers.
I would strongly assume the chips on the Chinese board are almost certainly counterfeit. This is the perfect product to hide them in. The authentic DV134 is only made by and available from Texas Instruments (Mouser and Digikey only carry TI versions anyway.) and is $5 a chip retail, and would bear a copyrighted TI logo. These do not. I doubt the manufacture would dump that kind of money into authentic chips and not herald that all over the Amazon listing. Worse, they are surface mount, so you can't easily change them out. I'm not saying they won't work, but if you want the real TI DV134 experience, this will likely fall short in some measurable manner.
On and on it goes, more and more complicated. How many stages are we up to? We're a long way from elegant and clean "less is more". You seem to care about performance, but are choosing ever more convoluted solutions that are not big wins, and losing quality with each addition. If only you had a stereo amp, you could cut out the pre-amp, and the buffer, and use your sources directly through the Schitt Sys. Elegant. (5K driving into 47K) What are your sources? Some devices let you select a mono output. BTW- you could easily use the stereo connector on your amp, do a summing network with two 47K resistors on the amp, (solves a lot of problems, makes the amp much more versatile in a stereo world) and build your upstream infrastructure to handle stereo for when you eventually get tired of mono and get some smaller speakers.
I would strongly assume the chips on the Chinese board are almost certainly counterfeit. This is the perfect product to hide them in. The authentic DV134 is only made by and available from Texas Instruments (Mouser and Digikey only carry TI versions anyway.) and is $5 a chip retail, and would bear a copyrighted TI logo. These do not. I doubt the manufacture would dump that kind of money into authentic chips and not herald that all over the Amazon listing. Worse, they are surface mount, so you can't easily change them out. I'm not saying they won't work, but if you want the real TI DV134 experience, this will likely fall short in some measurable manner.
So far the only thing that will be added will be the DRV134 board which will replace the tube buffer.
I could get this Schiit preamp as it has gain and I could then add a pair of resistors to sum the stereo to mono at the amp and reduce the signal some to make up for the 4X gain the preamp has, however I'd be losing one input.
https://www.schiit.com/products/saga_2
I could get this Schiit preamp as it has gain and I could then add a pair of resistors to sum the stereo to mono at the amp and reduce the signal some to make up for the 4X gain the preamp has, however I'd be losing one input.
https://www.schiit.com/products/saga_2
Last edited:
There you go. If it's in the budget. I like that it can be switched between passive or low and high gain. NICE!
Oh I could likely do it if available on Amazon using Affirm which is pay a little each month, however I currently have four sources (Schiit phono pre, PC Bluetooth & FM tuner) and at some point I may want to add a reel to reel deck so I'd need the 5th source.
I'm sure there's other remote controlled preamps, however the few Schiit products I've used have been well made, look good, sound great, are reliable and all at a reasonable price (they do have more expensive high end offerings though) so I'm more inclined to stick with that brand.
I'm tempted to add a Schiit DAC to the PC connected to the system
https://www.schiit.com/products/modi-plus
OR
https://www.schiit.com/products/modi-multibit-2
However the PC has a M-Audio Delta 192 24 bit 192kHz sound card in it and a DAC would likely not offer much improvement.
That said when Windows 10 is no longer supported I will have to get the DAC. So maybe it would be good to go ahead and buy one given how Schiit tends to retire products relatively quickly, however the replacement product is usually better so I may wait and see what Schiit has to offer in a DAC when I replace the PC.
I'm sure there's other remote controlled preamps, however the few Schiit products I've used have been well made, look good, sound great, are reliable and all at a reasonable price (they do have more expensive high end offerings though) so I'm more inclined to stick with that brand.
I'm tempted to add a Schiit DAC to the PC connected to the system
https://www.schiit.com/products/modi-plus
OR
https://www.schiit.com/products/modi-multibit-2
However the PC has a M-Audio Delta 192 24 bit 192kHz sound card in it and a DAC would likely not offer much improvement.
That said when Windows 10 is no longer supported I will have to get the DAC. So maybe it would be good to go ahead and buy one given how Schiit tends to retire products relatively quickly, however the replacement product is usually better so I may wait and see what Schiit has to offer in a DAC when I replace the PC.
Right, just run the Saga output to the Numark or run the Numark into the Saga. Full remote volume control is still in the chain.
I could run it into the Numark provided its specs are as good as the specs of the DRV134 chip. I would not run the Numark into the Schiit as I use the source selection of the Schiit.
Now concerning the DRV134 chip since it has built in 50 ohm resistors on the outputs can I tie both + outputs together and both - outputs together for a mono output or are individual higher value resistors needed?
If I need individual resistors that will defeat the feature of having a balanced or unbalanced output depending on whether a balanced or unbalanced cable is plugged in.
Now concerning the DRV134 chip since it has built in 50 ohm resistors on the outputs can I tie both + outputs together and both - outputs together for a mono output or are individual higher value resistors needed?
If I need individual resistors that will defeat the feature of having a balanced or unbalanced output depending on whether a balanced or unbalanced cable is plugged in.
100dB S/N, .05% dist. You would have so many different ways to add sources with low outputs and high outputs. Excellant phono preamps.
...update. Just did a search for the M4 and it looks like I must have bought the last one last summer. Well, I'm glad I did. I bought it for specs and it lives up to them. Very quiet machine. Again, good luck with your project. Enjoy.
...update. Just did a search for the M4 and it looks like I must have bought the last one last summer. Well, I'm glad I did. I bought it for specs and it lives up to them. Very quiet machine. Again, good luck with your project. Enjoy.
Last edited:
Will do thanks.
That would definitely be a good choice for my main stereo if I ever get around to connecting the sources back up to it.
Saw your question on 'karma- "Would EFB be of any benefit to this amp?"
I'm sure you will get better, more thorough answers than mine, but after understanding (most) of the reading, at a high level I would indicate that you have already switched to fixed, adjustable bias, so you will see most of the benefit already. One really cool thing about EFB though, is that the adjust pin is set with a scaled version of the screen voltage, and by relation the plate voltage, so that the bias compensates as the screen and plate droop under heavy load.
I will give a warning though, not addressed by Dave Gillespie until much later. Multiple users have had the LM337 regulator fail, and he identified two fairly rare causes- filament float due to transformer leakage, and I believe adjust coming up before the output of the regulator from SS rectification, all addressed with some simple protection diodes. However, I have had two fail for much more common and obvious reasons- shorted tubes. One had a grid wire that broke free and welded to the suppressor grid while in use (I test my tubes), the other had a damn flake come off a structure in the top of the tube and shorted the grid to the plate. In both cases, as soon as the regulator (and therefore cathodes) shorted to ground, the tubes red-plated. Easy to replace the regulator, only because I caught it before it smoked my transformers.
Given where your amp is now, I think it could benefit a LOT more from the stellar floating paraphase inverter with the 'transformerless' HF feedback hack.
I'm sure you will get better, more thorough answers than mine, but after understanding (most) of the reading, at a high level I would indicate that you have already switched to fixed, adjustable bias, so you will see most of the benefit already. One really cool thing about EFB though, is that the adjust pin is set with a scaled version of the screen voltage, and by relation the plate voltage, so that the bias compensates as the screen and plate droop under heavy load.
I will give a warning though, not addressed by Dave Gillespie until much later. Multiple users have had the LM337 regulator fail, and he identified two fairly rare causes- filament float due to transformer leakage, and I believe adjust coming up before the output of the regulator from SS rectification, all addressed with some simple protection diodes. However, I have had two fail for much more common and obvious reasons- shorted tubes. One had a grid wire that broke free and welded to the suppressor grid while in use (I test my tubes), the other had a damn flake come off a structure in the top of the tube and shorted the grid to the plate. In both cases, as soon as the regulator (and therefore cathodes) shorted to ground, the tubes red-plated. Easy to replace the regulator, only because I caught it before it smoked my transformers.
Given where your amp is now, I think it could benefit a LOT more from the stellar floating paraphase inverter with the 'transformerless' HF feedback hack.
Since the DRV134 has a balanced output, would it be better to wire up the 12AX7 as two gain stages and use the transformerless HF feedback hack or will that then make the amp dependent again on the 12AX7 sections being matched?
If I had the necessary audio measurement equipment I'd love to measure the amp as it is now then make the changes and remeasure the amp.
The amp does sound real good now and I'm reluctant to make any changes to the amp circuit itself without some sort of measurements such as distortion, frequency response ETC... that would indicate something that could be improved.
If I had the necessary audio measurement equipment I'd love to measure the amp as it is now then make the changes and remeasure the amp.
The amp does sound real good now and I'm reluctant to make any changes to the amp circuit itself without some sort of measurements such as distortion, frequency response ETC... that would indicate something that could be improved.
Last edited:
On the first question I plead 'above my pay grade'.
You have a resistive load, and I assume some type of O-scope for the measurements you took. My cheap tiny little scope had a signal generator built into it that I could attenuate with a pot. I then picked up a cheap signal generator with adjustable output that is a little better. That's all you need for frequency response, power bandwidth, output impedance, gain with and without feedback connected, clipping behavior and levels, crossover notch distortion or not, square wave performance, including edge rate, overshoot, ringing, oscillation, ramping etc. You can also generally pretty well tell just by eye on a sinusoid whether there is obvious distortion as you approach clipping. That's about 90% of what you need to do. If you have it looking pretty good across all of those, having an absolute distortion figure is the only thing you are missing.
You have a resistive load, and I assume some type of O-scope for the measurements you took. My cheap tiny little scope had a signal generator built into it that I could attenuate with a pot. I then picked up a cheap signal generator with adjustable output that is a little better. That's all you need for frequency response, power bandwidth, output impedance, gain with and without feedback connected, clipping behavior and levels, crossover notch distortion or not, square wave performance, including edge rate, overshoot, ringing, oscillation, ramping etc. You can also generally pretty well tell just by eye on a sinusoid whether there is obvious distortion as you approach clipping. That's about 90% of what you need to do. If you have it looking pretty good across all of those, having an absolute distortion figure is the only thing you are missing.
That's how I've always seen distortion and other issues with an amplifier.
When I make a power measurement I do it at the maximum output voltage before clipping starts to be seen on a scope with the amp feeding a non-inductive load at its rated impedance. I then use ohm's law to calculate the power.
For testing anything balanced I'll use my dual channel scope with channel 2 inverted and if both outputs are the exact same I should see no signal at all. Any signal shown shows one section is not 100% the same as the other section. That's particularly useful for push pull tube amps.
Posted over in the More Fun With Magnavox thread.
"For those who wish to maintain the original sensitivity level of 1Vac, a compensated reduction network is shown on the schematic that should be inserted between the input jack, and the junction of the 47k/470k resistors."
So I'm wondering if the original sensitivity of this amp is 1Vac which could explain why I needed some gain. I'm not entirely sure the gain was reduced with the driver and phase splitter mod I did.
Same. I know it can be done, however how well it will work I am not sure of and I have no easy way to do that sort of test. It would require feedback either the modded way from the plates or by grounding the 4 ohm tap and using the 16 ohm tap and ground to provide the feedback. If I did that it would require replacing the 1/4" speaker jacks with isolated 1/4" jacks.
Plus it would mean that I could not test the amp without the DRV134 board or other balanced source connected.
Last edited:
Just suppose you would build a low Zout preamp with moderate gain of 5x and enough inputs (what is considered a normal preamp). Then the summing resistors at the input of the power amplifier. Or is this illogical in your way of thinking?