• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Fun little P-P input stage

Status
Not open for further replies.
Jump to Latest
#41
Tube Rolling

With a second type of 7591, still at 1 kHz and run without the input transformers (same as the previous measurement; other input grounded), I measured 0.4% 2nd, 0.03% 3rd at 35 peak volts out (70 p-p). And I should note that these measurements are taken with the cathode-source resisters bypassed with 100 uF. So, distortion performance very much depends on the tube. It might be interesting to try a 6SU7 here.

EDIT: Early Alzheimers. I did NOT mean 7591, I meant 5691. Duh!
 
#42
Balanced Preamplifier

This is an interesting circuit too.
 

Attachments

  • bal_pre.gif
    bal_pre.gif
    75.1 KB · Views: 1,174
#43
I played around with a circuit like that some years ago, though using the 6S*7 and 6DJ8 families of tubes rather than the ones you show here; the small-signal performance was good, but swing was rather limited. Fine for a preamp, less so for a power amp driver stage.
 
#46
all-tube x-couple cascode

fkl, my apologies. I've gone through my lab notebook and I misremembered- I did not get this built with 6SL7/6SN7, just with 6KN8 and 6922. If I have time this weekend, I'll proto up a 6SN7 version and post schematic and results. But I'd still encourage you to try the tube-FET version- it works very well indeed.
 
#48
As a line stage, yes, it ought to work quite well; I'd use the input transformer version if your sources can handle a 10K load. If you want to use a higher value input resistor with the transformer, Jensen shows a nice little RC network in their app notes that will let you do that.

With a 5691 or 6SL7, the gain was about 20 (26 dB). I've noted rise time and distortion earlier in the thread. In balanced mode, the 2nd harmonic (a few tenths of a percent) will largely cancel.
 
#50
Actually, I've downloaded a couple of PC design and schematic packages. I'll try the Microcap package you suggested- thanks for the recommendation.

My problem is my own impatience with the learning curve (I already know AutoCAD, albeit poorly). When I worked in electronics, I got spoiled by having some really bright engineers working for me; I could sketch out what I wanted, and they worried about CADing it up and building the prototypes. The biggest challenge around here these days is getting a few uninterrupted hours- I should never have gotten married and procreated, takes too much time away from my hobbies 😉
 
#52
Transformer swap

I keep looking at the simple yet smart driver circuit and I keep wondering how it would sound. Since I really don't want to dole out $70plus each for the Jensen I was wondering if I

Could use a Altec Lansing Model 15015A in place of the Jensen JT11P1? Why do I want to use the Altec? Because I have a bunch of them.

Altec Lansing model 15015A 15K to 15K bridging transformer.

Which 2N5462? Mouser has 100 lot for .09 ea I believe they are Fairchild.
 
#53
You don't really need the input transformer, but it does help the balance a little bit and provides galvanic isolation. At the moment, I'm using my amp without the tranny.

The 5462s I used were antiques, Siliconix, left over from a mass purchase back in the late '70s when I was using this circuit to modify Stereo 70s. Main thing here is to match them well using a simple jig.
 
#54
Quote:

Main thing here is to match them well using a simple jig.

Could I use a Sencore TF 26 to match some Fets?

Is there a matching jig I need to make in order to match the Fets I bought? Every one I test(without touching them) I get a gain of 500 on my meter.


I have never played with any Fets. Does the drain go to the -24v, the source to the 5691's grids, and the gate to the input?
 
#55
For a jig, I connected the drain to -10V, grounded the gate, and connected the source to ground through a 470 ohm resistor. Using a voltmeter, I measured the voltage across the resistor and paired FETs with the same readings.

Your intended connection is correct. The sources do go to the "opposite" tube's grid, but also to the cathode of its tube section through the resistor.

In my current setup, I bypassed the cathode resistor with a 100uF cap. This decreases local feedback and increases open loop gain, which may or may not be what you want. It's an interesting thing to experiment with.
 
#58
A follow up to my comments about cathode bypass caps. I said that I subjectively preferred the amp with bypass caps in place (i.e., reduced loacal degeneration). So, I thought it might be fun to do some measurements to see if I was all wet or not. I just built one of these stages for an experimental EL84 amp using a 12BZ7 dual triode and J271 p-FETs. At 12V peak out (enough to take the EL84s into grid current), 1 kHz, and with the bypass caps, 2nd HD was -46dB, 3rd was -50dB. Without the bypass caps, second dropped (as expected) to -51 dB, but third actually rose to -46 dB. Now, in a p-p amp, the second is cancelled in the output stage, but the third is actually added, so there's a significantly worse distortion performance with the local feedback!

A lovely example of measurement and listening actually correlating.
 
#60
Take a resistor from the output transformer secondary to the gate of the inverting input FET. Then connect a second resistor from that FET gate to ground. For a normal gain structure, the first resistor should be 20 times larger than the second. 20K and 1K will work fine. You may need to put a small capacitor (like 20-100pF) across the 20K resistor if you see ringing on square waves.

As usual with any feedback amp, if you get oscillation when the feedback loop is closed, interchange the gate connections on the input FETs or swap output transformer connections to the output tube plates.

Sorry, can't draw a picture, my scanner is 5000 miles away!
 
Status
Not open for further replies.
Top Bottom