# 50C5 pentode LTSpice model

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

#### Mosquito

Hi everyone!
A question, someone knows if there is out there a LTSpice model, of a tube with similar behaviour to the 50C5 pentode?
No matter about pin compatibility or heater voltage, just similar maximun Va of 100/120 volt, and similar plate dissipation and optmal load around 2.5K
thanks in advance for any suggestion.
Cheers
J.

#### Mosquito

Hi Singh Santa, many thanks!
I'll try it this evening
Cheers
J

#### Mosquito

Hi
I get this tetrode model of the 50C5, but it isn't working at all.
Can someone try it and confirm (just to be sure that the error is in the model, and not in my circuit)
I suspect on some sintax error somewhere.
Thanks
J.

******************************

*Vacuum Tube Tetrode (Audio freq.)
.SUBCKT X50C5 A S G K
*
* Calculate contribution to cathode current
*
*the number at the right end determines sharpness of knee
Bat at 0 V=0.636*ATAN(V(A,K)/12)
*the URAMP(V(S,K)/# mostly determines peak plate current, grid line spacing nearly constant
*the number at the right end determines slope of grid lines
Bgs gs 0 V=URAMP(V(S,K)/6.2+V(G,K)+V(A,K)/90)
*the exponent sets the linearity of grid line spacing, and big impact on peak plate currrent
Bgs2 gs2 0 V=V(gs)^1.58
Bcath cc 0 V=V(gs2)*V(at)
*
* Calculate anode current, grid line spacing adjust and peak plate current
*
Ba A K I=1.35E-3*V(cc)
*
* Calculate screen current
*
Bscrn sc 0 V=V(gs2)*(1.1-V(at))
Bs S K I=0.6E-3*V(sc)
*
* Grid current (approximation - does not model low va/vs)
*
Bg G K I=(URAMP(V(G,K)+1)^1.5)*50E-6
*
* Capacitances
*
Cg1 G K 13p
Cak A K 8.5p

Cg1a G A 0.6p

.ENDS X50C5

**********************

Last edited:

#### mashaffer

Be sure that the order of parameters in the subckt call is the same as your symbol. I have run into problems where the nodes didn't match up and I ended up with B+ connected to the grid etc.

#### Mosquito

Ok Mike, I'll check that.
Many thanks
J

#### Alain Poitras

Hi
I get this tetrode model of the 50C5, but it isn't working at all.
Can someone try it and confirm (just to be sure that the error is in the model, and not in my circuit)
I suspect on some sintax error somewhere.
Thanks
J.

******************************

*Vacuum Tube Tetrode (Audio freq.)
.SUBCKT X50C5 A S G K
*
* Calculate contribution to cathode current
*
*the number at the right end determines sharpness of knee
Bat at 0 V=0.636*ATAN(V(A,K)/12)
*the URAMP(V(S,K)/# mostly determines peak plate current, grid line spacing nearly constant
*the number at the right end determines slope of grid lines
Bgs gs 0 V=URAMP(V(S,K)/6.2+V(G,K)+V(A,K)/90)
*the exponent sets the linearity of grid line spacing, and big impact on peak plate currrent
Bgs2 gs2 0 V=V(gs)^1.58
Bcath cc 0 V=V(gs2)*V(at)
*
* Calculate anode current, grid line spacing adjust and peak plate current
*
Ba A K I=1.35E-3*V(cc)
*
* Calculate screen current
*
Bscrn sc 0 V=V(gs2)*(1.1-V(at))
Bs S K I=0.6E-3*V(sc)
*
* Grid current (approximation - does not model low va/vs)
*
Bg G K I=(URAMP(V(G,K)+1)^1.5)*50E-6
*
* Capacitances
*
Cg1 G K 13p
Cak A K 8.5p

Cg1a G A 0.6p

.ENDS X50C5

**********************
Hi Mosquito,

I have many used 50C5 in my junkbox and I to experiment some circuits with this tube, so I was looking for Spice model for it when I found your post.

I currently use SIMetrix Simulator, free demo version, to test all the circuits I design, it is a very powerfull and easy to use software, probably the best ever made ... It enable you draw and simulate any circuit in very fast and work with all models I found since I use it.

The good news is your 50C5 model seem to work very well, so if you have a problem with your circuit in real life, it is probably due because and assembly error, tube condition or wrong ajustment. You should verified all voltages and current to make shure everything is allright and make shure you dont exceed the maximums rating of the tube, particulary the plate and screen grid dissipation and voltage.

I draw the very basic circuit below, folowing the recommendation of the GE 50C5 datasheet for fixed bias SE, and the results are very close. The only difference with thoses specs is I have to use a slighty higher voltage for the bias, - 8,72V instead of -7,5V to obtain the sames plate and screen grid current. The power with and "ideal transformer" reach 2 watts RMS for an input signal of 8,72V peak (6,165V RMS) for a total harmonic distortion of 8,26% ... Of course, no transformer is perfect and there is alway some lost ...

An externally hosted image should be here but it was not working when we last tested it.

It is not easy for a simulator to act exactly like the description of the datasheets, because the limits of the mathematical equations used, and the curves of any tubes are not exacts anyway. It is also difficult to find two tubes of the same type with a perfect matching, there is always some adjustments to do to in a real circuit get close to the perfection.

If you buy used or NOS tubes on eBay, you might consider using 25C5 instead of 50C5 because they are less commun and usually cheaper. Notice my schematic need some adjustment in "real life", for example, I dont consider the voltage drop in the primary of the transformer, I got a little one in here and the resistance of the primary is 167 ohms, then at 49 ma plate current, the DC voltage drop across this coil is about 8,2 volts and the supply should be 118,2V to get 110V between the plate and the cathode ...

The 50C5 is not very powerfull but can give good results with an high-efficiency loudspeaker, like any other tubes, there is several ways to lower the distortion, for example, using a low dynamic plate resistance driver tube and a O.H. Schade's local feedback used to give a triode sound to any beampower, pentode, tetrode and also JFET and MOSFET ...

Using a autobias resistor with a bypass capacitor will give a different sound and prolongate the life of the tube. But you have to use the best coupling and cathode bypass capacitors to get the better sound. You must also consider the supply have to be higher because the voltage drop across the autobias resistor. And to evaluate the real power closer, you have to add the resistance off the primary to the impedance of the charge and also the reflected impedance of the secondary witch is often 10% of the impedance of the speaker ...

In electronics, everything must be carefully calculated and you must consider the tolerance of all the parts, specially the semiconductors in wich the current vary considerably between the differents specimens and with the case temperature ... So, good luck with your projects and have fun.

#### Alain Poitras

Try 35c5 it's similar View attachment 197344
Cheers
SS
HI singh santa,

I look in my olds RCA databooks and I get the confimation the 25C5 is absolutely electrically identical to the 50C5, except for heater ratings, of course.

I try your 35C5 Spice model with the same circuit as above but I get quite differents results. So, I go back to my books and realise the 35C5 is different than the 25C5 and the 50C5, it is less powerfull and the maximum power dissipation of the plate and screen grid are lower, respectively 4,5 and 1 watt ...

The specs for the 35C5 indicate for 110V on the plate and screen grid, the maximum power obtainable for a load of 2500 ohms is 1,5 watts RMS. But the current of the plate "with no signal" must be 40ma for the plate and 3 ma for the screen grid with a fixed bias of -7,5V ...

I run my simulator and I get those currents with -7,15V bias instead of -7,5V, and with a 7,15V peak signal, I get 3,4V RMS at 7,33% of total harmonic distortion at the 8 ohms loudspeaker, the power is then 1,45 watts RMS !

So, your Spice model work very well for the 35C5 but not for the 25C5 and the 50C5 ...

Cheers,

Alain.

#### Alain Poitras

Yesterday, I look at the specs of the 50C5 in my latest RCA tubes databook from the sixtie's and I find out they are different from the older one. I also look in the datasheet of Thung-sol and it is the same thing ... The maximum power dissipation of the plate and screen grid are higher, 7W instead of 5,5W for the plate and 1,4W instead of 1,25W for the grid, and with - 8 volts bias the output power can reach 2,3 watts.

So, it is possible that differents brands of 50C5 tubes have differents maximum rating and power capability ... Be carefull about that with NOS tubes ...

An externally hosted image should be here but it was not working when we last tested it.

Some times ago, I was wondering if it is possible to use a autotransformer instead of a conventional output transformer because there will be is no DC current in the windings then no DC saturation, resulting on better AC power handling and bass response.

There is the first test schematic I design for that idea, I buy two small autotranformers in a montreal electronic surplus store (Addison Electronique Ltd) for only 1\$ each and I will probably try that this winter when I will not be too busy ... May be this will give some design ideas to some members of this forum, of course, it is perfectly possible to use a conventional 2500 / 8 ohms output transformer instead a autotransformer but it will cost a lot more ...

An externally hosted image should be here but it was not working when we last tested it.

The circuit use a DN2540 high voltage depletion MOSFET as current limiting source, to make some kind of SRPP circuit, R6 must be a multiturns trimpot to adjust precisely the voltage on the plate of the 50C5. The 280 VDC supply is a minimum and can be much higger, in this case, the value of R3 must be calculated for a voltage drop to get just over 120 volts at R1 without the zeners at the maximum current of the screen grid wich is 8,5ma. When the current is lower, the four zeners prevent the voltage to go higher. I use BZD2330 in the schematic because it was in the SIMetrix library but I will use four 1N5363A, 30V 5W in my real circuit, I order a lot of 20 of these yesterday on eBay for less than 5\$, a very good deal ...

Since I use a autobias circuit and there is a voltage drop of 8,847 volts across the R9 165 ohms resistor (wich can be two 330 ohms in parallel), the voltage on the plate and screen grid to the ground can be about 8,85 volts higher ... That's why I adjust R6 to get 128,9 V on the plate and 118,8 V on the screen grid is OK !

My choice of capacitors values with a "perfect" autotransformer give a quite flat response down to 20 Hz but this will probably not be the case with the small autotransformers I buy for 1\$ each ... The fact is the inductance of the winding must be over 50 henry to give that response, so I will have a more important drop at 20 Hz, but no matter, I make this circuit just for fun and to see if it will work well !

I will also have to design a good power supply for it, the four zeners can make a good regulation for both channel of a stereophonic system and also the driver and preamplifier but R3 have to be calculated consequently, this is my next simulation steps before physical assembly. Many hours of pure pleasure ...

Notice the SIMetrix simulator not only calculate the distortion with fourier analysis but also produce frequency response curves and really a lot of other very interesting functions, it is a fantastic software ... With this circuit and a "perfect" transformer, the output power suppose to be over 2,5 watts at less than 5% of THD, this is already very good for a 50C5. But I will add a drive to the circuit with a local feedback wich will make the distortion lower, I think near 1% THD is possible.

Thank to Mosquito with is 50C5 Spice model that make my circuit simulation possible !

#### Mosquito

Hi Alain, just reading your post today. Now I'm just starting a new life in my country of origin so, by the moment, I'm very busy and found little free time for our hobby, but your approach seems very interesting. Thanks for sharing your results and the info about the SiMetrix simulator.
Cheers
J

Last edited:

#### Alain Poitras

Hi Alain, just reading your post today. Now I'm just starting a new life in my country of origin so, by the moment, I'm very busy and found little free time for our hobby, but your approach seems very interesting. Thanks for sharing your results and the info about the SiMetrix simulator.
Cheers
J
Good luck with you new life !

I am presently doing some simulation tests with differents driver tubes using "Schade's local feedback", I try a 12AT7 model and I actually get 2,7% THD at 2,2 watts RMS, over that, the 0 volt bias clipping start and the distorsion raise to 10,26% at 3,125 watts ... Not bad for a 50C5 ... I will also try the 12AU7 as driver.

When I get the best simulation results, I will post a message on this subject about it ... I write before the 25C5 is electricaly identical to the 50C5 but there is also the 12C5, 12CU5, 17C5, the 50B5 too but the "pinout" is different, the old octal 50L6 have also similar performances but is more powerful.

I was thinking about using the same voltage for the heaters of all the tubes and I order a NOS (new old stock) 12C5 yesterday on eBay for 5,23\$US shipping included, it is very cheap because usually, NOS 50C5 are sold for up to 20\$ with shipping ...

I also choice the 12C5 because 12,6 volts transformers for the heaters are very cheap and commun, a store near my place sold 12,6V 1A model for 10\$ and 2A model for 13\$ ... It is also possible to use two of them "back to back" (the two secondary branch together) to obtain an isolated power supply of 120VAC and with a fullwave voltage doubler, to obtain a DC supply over 300 volts, perfect for a little power amplifier or a preamplifier, this way, the frame can be safely grounded ...

There is many possibility for the supply since with my circuit, the DC voltage don't have to be exact, anything between 280VDC and 400VDC will do without a heatsink for the DN2540 with TO-220 package, it can dissipate 15 watts in free air at Ta = 25°C, but of course, with some little adjustments of resistors values .

I will be back here with that project but I am also busy and not in a hurry for DIY, I keep that for the winter ... I know in your country, it is actually winter but here in Quebec, Canada, it is from December to February ... I live in the Laurentides mountains and during winter, there is often over one meter of snow, it is a real ice hell and temperature go down to - 40°C ... It is then better to stay confortably inside, beside the wood stove, studying electronic and doing DIY !

This month, it is very warm here and my first hobby is cultivating my garden, I grow vegetables like tomatoes, green beans, zukinis, cucumbers, onions, potatoes, carrots, fine herbs, and also Chili and Habanero hot peppers ... I think you know very well those peppers, I love spicy foods, but the yellow and orange Habaneros are really dangerous ... I use them only to make "volcano" salsa, taking care not to put some of their juice in my eye's or you know where else ...

Cheers,

Alain.

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