vintage tube I/V from the year 1960

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
X.G.

X.G.

Member
Joined 2003
philbyx said:
Hello

I'm working too on a valves I/U. But I want 0v on input, to avoid stressing my dac wich is based on ad1865.

This is just ideas on paper for the moment...

V1=160V
V2=8V

I : 6ma in first tube, 7ma in CF

Tl431 for negative voltage regulation and as ccs reference.

The 15k resistor on the anode of the first tube gives 15V for 1ma full scale on the ad1865.

The simulation works.
Click to expand...

hi,Philippe

15V!....stronger!!!:D

the Q1 is not set to CCS.if Q1 is set to CCS,the circuit can work in a better way.





X.G.:)
 
philbyx

philbyx

Member
Joined 2004
phil.charlet.free.fr/index.html
X.G. said:


hi,Philippe


the Q1 is not set to CCS.if Q1 is set to CCS,the circuit can work in a better way.

X.G.:)
Click to expand...

Q1 has a regulated 2.5 Volts (the TL431 reference) on its base
With the 300 ohms on the emitter, we have :

300 x I = 2.5-Vbe
I= (2.1-Vbe)/300 = 1.8/300 = 6ma

I think that this is a CCS.

If 15V is stronger, you can decrease the value of the anode resistor.

Philippe
 
G

Guido Tent

Member
Joined 2001
TentLabs.com
X.G. said:
the Q1 seems to a CCS,but not a enogh good CCS .....the Q1 more like a volt regulator;)

sorry for my poor English,I can not explain more clear.

X.G.
Click to expand...


Hi

Your English is OK, at least I understand your point

Indeed, this configuration with the 431 is a little "strange"

If the base is kept at constant voltage, the emitter will also be at constant voltage, then the emitter current is constant too (so current source)

I would hower suggest to use the 431 a little different:

Add a few 100 ohm series resistor in the anode, the ref input connected to the anode too. Then you have a rigid 2.5V voltage available across the anode-cathode.

succes
 
philbyx

philbyx

Member
Joined 2004
phil.charlet.free.fr/index.html
X.G. said:
:xeye:

for the CCS of BJT,the circuit should be set the Ib of BJT to constant....but on here,the stability of Ib is not so good,because of the less control of TL431 here than the standard application.

regards

X.G.
Click to expand...

Thanks XG and Guido

Here is a modified circuit :

The purpose of the tl431 is to give a solid negative voltage referenced to the ground, to drive the grid and to insure 0v on the cathode.

I can adjust this 0v with R15. The 160 V needs tu be very well filtered (bad rejection of the power supply here).

I had connected the base of the ccs to the ref of the TL431 for the zener cost, but I agree with you : this reference is not very good.

Then here is a modified circuit, with also a 3k on the anode of the first tube : 3v peak for 1 ma, then 2.1 volts RMS on the output.

The Cathode follower is no more in direct link, it is because my simulation soft cannot converge if I do it.

Philippe
 

Attachments

  • dacecc2.jpg
    dacecc2.jpg
    39.7 KB · Views: 499
philbyx

philbyx

Member
Joined 2004
phil.charlet.free.fr/index.html
X.G. said:
I think.... you don't use the D7 and D12 will be better,they just raise the noise.what thay do is supplied by the TL431.

PS:you can add a cap for decoupling the D13 to reduce the noise.

IMO,your circuit is very good now.

good luck.

X.G.
Click to expand...

You are right for the cap on D13, my schematic is for the principle, and requires some refinements.

I can't have and adjust the grid voltage directly with the tl431. The Tl431 cannot feed under 2.5 Volts, I need around 2 volts on the grid,then I regulate a greater voltage, and lower it.
I thought about diodes, because it is simple. But as we are regulated, perhaps only resistive bridge like this (R25/R27 and C8 for noise)...

Philippe
 

Attachments

  • dacecc3.jpg
    dacecc3.jpg
    39.4 KB · Views: 1,346
X.G.

X.G.

Member
Joined 2003
philbyx said:




I can't have and adjust the grid voltage directly with the tl431. The Tl431 cannot feed under 2.5 Volts, I need around 2 volts on the grid...

Philippe
Click to expand...
you need around 2 volts on the grid...why?

if you don't use the D7 and D12,the volt between the grid of tube and -V is regulated because of the TL431,and the cathode volt is self-adjusted.if you use the D7 and D12,the results above is the same,but the volt between the grid of tube and -V is regulated because of the D7 and D12.Both of them can work of couse.;)

EDIT:....but the volt between the grid of tube and -V is regulated because of the D7 and D12
 
Franz G

Franz G

Member
Joined 2004
If the DAC is a true current output, 1,3V couldn't harm.
Click to expand...

Coming back to the simple idea, just a 6C45 with cathode resistor, and therefore 1.3 to 2V at the output of the TDA1541A, I consulted again the datasheet of the DAC.

Concerning the output (4mA), they add the following note:

"To ensure no performance losses, permitted output voltage compliance is +-25mV maximum."

Does this refer to the resulting output swing or possibly to bias like above?

Franz
 
G

Guido Tent

Member
Joined 2001
TentLabs.com
Franz G said:


Coming back to the simple idea, just a 6C45 with cathode resistor, and therefore 1.3 to 2V at the output of the TDA1541A, I consulted again the datasheet of the DAC.

Concerning the output (4mA), they add the following note:

"To ensure no performance losses, permitted output voltage compliance is +-25mV maximum."

Does this refer to the resulting output swing or possibly to bias like above?

Franz
Click to expand...

I suspect the swing. This would require an input impedance of 12 ohm or lower.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom