What is SPICE

colin.hepburn · Post by **colin.hepburn** » Thu Dec 06, 2007 10:23 pm

Hi Andrew/Nick
Ran the sim takes a bit longer than I expected not sure I under stand the traces as well as the psu one is there a Goggle link that will explain the traces more is It better to leave the visible trace in auto range or run them by selecting different ones I’m starting to think I am the only one doing this thats OK but I think It would help more if the others joined in so we can all get some feed back from each other

Post by **Nick** » Thu Dec 06, 2007 11:43 pm

I am standing back to let Andrew do his valve model thing, but I would give it a bit of time. maybe others have things to do, so the step a night may not work for all.

Andrew · Post by **Andrew** » Fri Dec 07, 2007 9:01 am

Colin and others...

The sim does take time to run.

First I guess we need to be comfortable with the basic circuit, would a bit of revision be useful? I guess there's no harm especially since we're looking at the circuit in a different 'virtual' environment.

OK, the triode. The anode on the triode is the upper symbol inside the triode's envelope, the envelope is represented by the outer circle. The anode's wire goes out of the top up to the load resistor, R1. The grid is next down, its the '---' lines coming out of the left. Below that we have the cathode wired out of the bottom of the envelope to R2.

V2 is our music, or if we're on the virtual test bench, as we are, our signal generator. We can make this do square waves, sine waves, ramps etc, in Spice we just use a voltage source for both our signal gen and the PSU, which is why they have the same symbols.

The triode works into the load R1, V1 provides the DC; it is our PSU.

R2 is the cathode resistor that biases the triode by making the grid negative with respect to the grid. C1 is our cathode bypass cap.

On the wire coming out of V2 you should see the 'music' signal, out of our signal generator, in the case we set a 1kHz sine wave.

Double click on the wire out of V2 to see the input sine wave.

Double click on the wire out of the anode to see the output signal, we might connect this signal to a second stage, say directly to a 2A3 in the Loftin White or more traditionally, we put a cap there as it will block the DC letting only the AC through.

Because the grounded cathode circuit inverts the sine wave at the anode will be the other way up to the sine wave on the grid.

The anode's wire will have some DC and some AC on it. We need remember that this type of analysis will show both DC and AC behaviours, as Nick said, its an analysis over time of the circuit.

Double click on the wire at the cathode and another sine wave will be shown again with some DC.

When say DC I'm trying to say that the sine wave will not pass through 0v at its mid point, but the midpoint will be at some fixed DC voltage. In the cathode's case this will be the bias we have set the circuit at - about 1.35v DC. You will also see the sine wave caused by the input signal. If you delete the cap C1 you will see a bigger AC wave. Without the bypass cap you see more AC as the bypass caps provides a path for most of the AC to go around R2 to ground.

If you get out the ECC83 curves you should be able to see where we have biased the tube etc.

Oh, and R3 is our grid leak.

This was just supposed to be an intro. Later we can show how just to look at the various DC values in the circuit on they're own, the AC behaviour at all frequencies. More power supply issues and lots of other good stuff. How to look at the curves and pick an operating point etc.

But we should let the others catch up.

-- Andrew

Dave the bass · Post by **Dave the bass** » Fri Dec 07, 2007 11:22 am

Sorry I'm late Sir(s), I've got a note from me mam.

I'm at home today while the builders are fitting a new fireplace so I'm Spice-ing today, just downloaded it.

Now building that first PSU model.

Young Master DTB (Year 2).

colin.hepburn · Post by **colin.hepburn** » Fri Dec 07, 2007 11:24 am

Hi Andrew
Thanks for the revision update helps me understand it a little better In some ways its helpfully that the others are late in gives me a chance to go back over what I have dun so far I have had a problem with LT not finding a txt files but a reboot sorts it out and some time I get a syntax error? I read somewhere this is a common problem with LT
Cheers

colin.hepburn · Post by **colin.hepburn** » Sat Dec 08, 2007 12:02 pm

Hi Andrew/|Nick
Looking at the valve dropdown menu in LT spice I see there’s a 12B4 is this same as the 12B4-A) if not does any one have a model for the 12B4-A valve
Cheers

Andrew · Post by **Andrew** » Sat Dec 08, 2007 1:58 pm

I would guess the two are close enough.

But others who no more detail might be able to help, might be worth asking under valves. Someone like John C. may know how close they are.

The Spice models are, in many respects, only approximations of real valves, so the question is how close do you need to be?

You can draw the Spice model's curves on screen and compare them to the datasheet for the 12B4. I can do the curves next up if you like?

cheers,

-- Andrew

Post by **Nick** » Sat Dec 08, 2007 2:09 pm

Looking on TDSL, the general numbers look the same, so I tnink you will be safe to use the model.

colin.hepburn · Post by **colin.hepburn** » Sat Dec 08, 2007 2:23 pm

Yep I did check on TDSL but is says different ratings or performance so wasn’t sure Andrew if you can do the curves next up that would be helpfully as I want to run my headphone amp after we have completed this tutorial
Cheers

Post by **Nick** » Sat Dec 08, 2007 3:54 pm

Good idea, gives an excuse to show the next sim type. I have to go shopping now, if you have time, can you give the details Andrew, if not I will do so later.

Andrew · Post by **Andrew** » Sat Dec 08, 2007 4:57 pm

I think Nick is planning (thinking) something else from what I was going to do. Nick were you planning a DC sim?

Attached is a Steve's circuit called 't-curves.asc' its shows how to make Spice show you the curves for a particular valve.

Time is short now but I've set it up to show the 12B4-A so Colin can compare.

If you click on the anode of the 12B4-A , the middle valve, get the ammeter (current) symbol not the volts symbol then click it should give you the curves - also attached as an image so you can see what I mean.

You'll need to manually set the ranges on the vertical axis, right click, select manual range you want 'Top' = 200mA and 'Tick' of of 10mA.

For now that's it, I'll do more when I get chance. But its worth comparing this to the data sheet.

cheers,

-- Andrew

Post by **Nick** » Sat Dec 08, 2007 6:50 pm

Yes, I meant a DC sim, does Steves model do somethng else?

Andrew · Post by **Andrew** » Sat Dec 08, 2007 6:51 pm

Yes, what I hope do now is plot our desired loadline on the graph as well as the max anode dissipation that our 12BA can hack. We can adjust this then to the most optimal point. This process is no different from the one, say, described in Bruce Rozenblit's book in Chap 2, we're just using a computer instead of photocopy of the data sheet, pencil and a ruler etc.

Then we can add this to the demo2 and see what it does in a DC mode sim.

Colin, your amp can be our working example if you like? What is your current load B+ etc.

OK, who's with me?

cheers,

-- Andrew

PS. How am I doing is this pitched OK? More, less etc? Feedback please.

colin.hepburn · Post by **colin.hepburn** » Sat Dec 08, 2007 7:41 pm

Hi Andrew

I now have three versions of this amp latest one is a higher current/ B+ voltage
The one at the top of this thread is what I’m using at the moment I an planning to rebuild this one to the higher current version http://www.audio-talk.co.uk/phpBB2/view ... 1&start=30 It may be better if you can PM me your email and I can send you the three sch and the design manuals that may help us better

colin.hepburn · Post by **colin.hepburn** » Sat Dec 08, 2007 8:35 pm

Hi Andrew
OK some feedback so far I I'm keeping up with the instructions that nick and your self posted although I found the PSU easer to understand partly dues to using PSU 2 but the next stage I’m not too sure that I understand what some of the traces represents i.e. how to read them in the grounded grid OK with things like square waves, sine waves but not sure what is good or bad hopefully the others will post there finding which will help me
Andrew regarding the’t-curves.asc' I'm lost here you say to click on the 12B4 the middle valve but this valve is not a 12B4 do I change it to the 12b4 or am I way off here
Sorry I'm not to good at trying to explain thing I hope you can decode this OK
Cheers

audio-talk

What is SPICE

#61

#62

#63

#64

#65

#66

#67

#68

#69

#70

#71

#72

#73

#74

#75