• 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.

DIY tube preamp with 5687

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I would like to build 5687 tube simple pre amp as in diyparadise. Could I pump out the B+ to 500V or more (e.g. around 600V, as coupling caps limit at 630 V).

If i use 500-600 V as B+, what value, type and wattage of suggested Rp and Rk? What about the filament, should it be floating (I use DC), or should be connected to 0/GND, e.g. to pin 8?

Thanks,

Ervin L
 
Administrator
Joined 2004
Paid Member
The 5687 works very nicely with 100V - 200V on the plate, and probably would be rather unhappy at 500 - 600V as it is not really designed for operation at such high voltages. (Maximum rated is 330V)

Pre-amplifiers typically only need to deliver a few volts of signal to the power amplifier so these kinds of voltage levels are not required in this application.

Note that a single tube with a pair of plate to line transformers could make a very nice, simple stereo line stage.
 
I would like to build 5687 tube simple pre amp as in diyparadise. Could I pump out the B+ to 500V or more (e.g. around 600V, as coupling caps limit at 630 V).

If i use 500-600 V as B+, what value, type and wattage of suggested Rp and Rk? What about the filament, should it be floating (I use DC), or should be connected to 0/GND, e.g. to pin 8?

Thanks,

Ervin L
5687 in mu follower for pre amp, sound very good. and top voltage will be 400-450V (actually 200v for each one). low voltage high current always sound better than high voltage with less current in tube circute no matter for power amp or pre amp
regard tony ma
 
I agree with Arnulf, why that high of a B+?


Maybe the goal is something crazy like to load the tube with 10X its internal resistance.

at 20ma the Rp is about 2K and 20ma through a 20K plate resistor will get rid of 400V of that really quickly :)

I came up with 500VB+
20K load resistor
200 ohm cathode resistor
gives 120V P-K -3.8V bias and 18ma of current.

dave
 
Hi all,

Thanks for the suggestion. I decide to start 5687 simple AF, with B+ of 300, Rp 22k 2 watt, Rk 470 ohm 0.5 watt, sounds good as starter, but with some hums (need to explore).

Actually i would like to ask whether to supply tube with very high voltage, low current (high Rp), or lower voltage but with higher current, which one will have better/less noise.

Looking at some datasheet (other tube), higher voltage give less THD, do not know with this tube.

Now is Rp/Rk ideal, or could I lower the Rp (to which extend), to get more current?

Currently, I measure V plate +/- 105 volt, V katode +/- 5 volt with 470 ohm (around 10 mA?). Don't know how to calculate Ip with V bias/grid, as internal resistance will vary based on voltage and V bias.

Thanks,

Ervin L
 
The 5687 works very nicely with 100V - 200V on the plate, and probably would be rather unhappy at 500 - 600V as it is not really designed for operation at such high voltages. (Maximum rated is 330V)

Pre-amplifiers typically only need to deliver a few volts of signal to the power amplifier so these kinds of voltage levels are not required in this application.


Hey-Hey!!!,
There is quite good reason to use high B+; one can use larger value load resistors. Take my 12B4 LTP linestage for instance. B+ is nearly 500V, plate resistors are 20k and plate voltage is ~110V so I can pass ~18 mA/tube. One cannot simultaneously run high plate loads and high plate current without high B+.

Dave has also seen this possibility, and IMO it has much merit.
cheers,
Douglas
 
Last edited:
Hey-Hey!!!,
There is quite good reason to use high B+; one can use larger value load resistors. Take my 12B4 LTP linestage for instance. B+ is nearly 500V, plate resistors are 20k and plate voltage is ~110V so I can pass ~18 mA/tube. One cannot simultaneously run high plate loads and high plate current without high B+.

Use CCS to get the high current at lower B+ (and consequently the best linearity possible). PSRR is an added bonus. You cannot get more than mu worth of amplification out of a tube anyway so whatever extra B+ you deliver, it all goes to waste.
 
Use CCS to get the high current at lower B+ (and consequently the best linearity possible). PSRR is an added bonus. You cannot get more than mu worth of amplification out of a tube anyway so whatever extra B+ you deliver, it all goes to waste.

I suppose how one defines 'waste' would be important. The plate votlage at which 'best linearity' is delivered has nothing to do with how high B+ is; one can still drop most of it across the plate load resistor. There are also designs which can not operate with CCS plate loads. My balanced output LTP is one; CCS on cathodes and anodes is asking for 'issues'. Add to that the OP may not wish to engage in CCS loading; seems a 99% sol'n to just use higher B+ and approach mu quite closely. You'll get indistunguishably close to mu with either a high load or CCS...if that is even an issue he's worried about.
Cheers,
Douglas
 
Maybe the goal is something crazy like to load the tube with 10X its internal resistance.

at 20ma the Rp is about 2K and 20ma through a 20K plate resistor will get rid of 400V of that really quickly :)

I came up with 500VB+
20K load resistor
200 ohm cathode resistor
gives 120V P-K -3.8V bias and 18ma of current.

dave

Hey Dave!

Well, there are some purists out there who would swear that a big wirewound resistor would get you closer to a horizontal load line with more "purity" than a CCS in the plate. Right? Yes, it's wasteful, that's for sure. But I tried that, and it does sound good.
 
There is a lot of opinion involved when it comes to what sounds 'better,' and of course there's always personal preference. I was only alluding to that. I think the plate-load CCS is harder for the less experienced (like me) to build successfully, so there's an attraction to using things like giant-valued plate load resistors to achieve lowest distortion. It's a simple and a brute force, burn off the volts way to go, and it is wasteful. Anyway, that's all I meant. They both work, right?
 
I think the plate-load CCS is harder for the less experienced (like me) to build successfully, so there's an attraction to using things like giant-valued plate load resistors to achieve lowest distortion. It's a simple and a brute force, burn off the volts way to go, and it is wasteful.

When in doubt, copy :D Seriously, these are basic building blocks that have been rehashed over and over again.

Have a look at Nelson Pass' Aleph circuit (patented, but free for domestic use), which implements CCS with all the bells and whistles. You can drop portions of it (the "Aleph" feedback part, or the bootstrap part for instance) and still get a functional building block, albeit with slightly lower performance.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.