I'm not sure what those 'RCA 12HL7 with small guts' look like. So I took photos of two different RCA 12HL7 I have.
- The first two photos are of a 'clear top' RCA 12HL7 with slightly smaller guts.
- The last two photos are of an RCA with getter flash on the side, and I think it has bigger guts. This tube looks almost exactly like some Sylvania 12HL7 I have.
- I also have a GE 12HL7 with dull gray plate that looks more like the 'bigger guts' RCA 12HL7, but not exactly the same.
Are any of these the RCA 12HL7 'small guts' version you sent back?
Sorry for the blurry photos. I took them in a hurry. I hope they work well enough. If not, please let me know and I'll try harder to post sharper photos. Thanks.
- The first two photos are of a 'clear top' RCA 12HL7 with slightly smaller guts.
- The last two photos are of an RCA with getter flash on the side, and I think it has bigger guts. This tube looks almost exactly like some Sylvania 12HL7 I have.
- I also have a GE 12HL7 with dull gray plate that looks more like the 'bigger guts' RCA 12HL7, but not exactly the same.
Are any of these the RCA 12HL7 'small guts' version you sent back?
Sorry for the blurry photos. I took them in a hurry. I hope they work well enough. If not, please let me know and I'll try harder to post sharper photos. Thanks.
Attachments
I just checked the schematic for my 12HL7-triode spud headphone amp. I had tried it with a 12HL7, triode wired, 145V on the plate, -1.8V grid bias (33 ohm cathode resistor w/ 1000uF bypass cap), 55mA plate+screen current, which works out to about 8W plate+screen dissipation. I loved the sound from this version but it burned through far too many 12HL7s, so I kept tinkering.
The final, stable version wound up with a 12GN7A-triode with 165V on the plate, -2.1V grid bias (62 ohm cathode resistor w/ 2200uF bypass cap), 35mA plate+screen current, which works out to almost 5.8W plate+screen dissipation. I run this version almost every night, and the 12GN7A tubes last about a year before needing to be replaced.
I like the sound from the 12HL7-triode a bit better, but I got tired of replacing so many of them. I might try a pair of Sylvania 12HL7 in there, to see how long they last.
The final, stable version wound up with a 12GN7A-triode with 165V on the plate, -2.1V grid bias (62 ohm cathode resistor w/ 2200uF bypass cap), 35mA plate+screen current, which works out to almost 5.8W plate+screen dissipation. I run this version almost every night, and the 12GN7A tubes last about a year before needing to be replaced.
I like the sound from the 12HL7-triode a bit better, but I got tired of replacing so many of them. I might try a pair of Sylvania 12HL7 in there, to see how long they last.
Hi, may be of help. Some spice models in triode mode. One is from measured 12HL7 there in DIY audio forum other is other measurements.
Code:
**** 12LH7_triode ******************************************
* Created on 05/05/2024 09:40 using paint_kit.jar 3.1
* www.dmitrynizh.com/tubeparams_image.htm
* Plate Curves image file: /Users/zoran/Desktop/Model_Paint_Tools/Graph/
* Data source link: DIY audio member measurements
* Graph rsz-T1.jpg 50V/horizontal, 10mA/vertical, Ug step -1V
*----------------------------------------------------------------------------------
* Heat: 6.3V/0.6A, 12.6/0.3A
* Vamax=400V, Vg2=330V, Pamax=10W, Pg2max=1W
* Heater to cathode voltage 100V, 200V with peak
* Rg= 250K auto bias, 100K fixed bias
*----------------------------------------------------------------------------------
.SUBCKT 12LH7_triode 1 2 3 ; Plate Grid Cathode
+ PARAMS:
+ RGI=2000
+ MU=28.09
+ KG1=128.96
+ KP=393
+ KVB=176.4
+ VCT=-0.5032
+ EX=1.275
*--------- CGP capacitances estimated !!! added 0.7pF to each
+ CCG=15.7P
+ CGP=1P
+ CCP=6.7P
*----------------------------------------------------------------------------------
* Vp_MAX=550 Ip_MAX=100 Vg_step=1 Vg_start=3 Vg_count=23
* Rp=8000 Vg_ac=2.1 P_max=11 Vg_qui=-5 Vp_qui=219
* X_MIN=104 Y_MIN=17 X_SIZE=751 Y_SIZE=691 FSZ_X=1665 FSZ_Y=811 XYGrid=true
* showLoadLine=y showIp=y isDHT=n isPP=n isAsymPP=n showDissipLimit=y
* showIg1=n gridLevel2=n isInputSnapped=n
* XYProjections=y harmonicPlot=y dissipPlot=y
*----------------------------------------------------------------------------------
E1 7 0 VALUE={V(1,3)/KP*LOG(1+EXP(KP*(1/MU+(VCT+V(2,3))/SQRT(KVB+V(1,3)*V(1,3)))))}
RE1 7 0 1G ; TO AVOID FLOATING NODES
G1 1 3 VALUE={(PWR(V(7),EX)+PWRS(V(7),EX))/KG1}
RCP 1 3 1G ; TO AVOID FLOATING NODES
C1 2 3 {CCG} ; CATHODE-GRID
C2 2 1 {CGP} ; GRID=PLATE
C3 1 3 {CCP} ; CATHODE-PLATE
D3 5 3 DX ; POSITIVE GRID CURRENT
R1 2 5 {RGI} ; POSITIVE GRID CURRENT
.MODEL DX D(IS=1N RS=1 CJO=10PF TT=1N)
.ENDS 12LH7_triode
*$
**** ******************************************
* Created on 05/05/2024 10:46 using paint_kit.jar 3.1
* www.dmitrynizh.com/tubeparams_image.htm
* Plate Curves image file: 12HL7_TriodeStrapped
* Data source link: DIY audio forum
*----------------------------------------------------------------------------------
.SUBCKT 12LH7_triode_01 1 2 3 ; Plate Grid Cathode
+ PARAMS: CCG=15.7P CGP=1P CCP=6.7P RGI=2000
+ MU=25.5 KG1=217.26 KP=430.5 KVB=23 VCT=0 EX=1.587
* Vp_MAX=200 Ip_MAX=50 Vg_step=0.5 Vg_start=1.5 Vg_count=18
* Rp=10000 Vg_ac=2.25 P_max=11 Vg_qui=-3 Vp_qui=141
* X_MIN=127 Y_MIN=114 X_SIZE=929 Y_SIZE=506 FSZ_X=1665 FSZ_Y=811 XYGrid=true
* showLoadLine=y showIp=y isDHT=n isPP=n isAsymPP=n showDissipLimit=y
* showIg1=n gridLevel2=n isInputSnapped=n
* XYProjections=y harmonicPlot=y dissipPlot=y
*----------------------------------------------------------------------------------
E1 7 0 VALUE={V(1,3)/KP*LOG(1+EXP(KP*(1/MU+(VCT+V(2,3))/SQRT(KVB+V(1,3)*V(1,3)))))}
RE1 7 0 1G ; TO AVOID FLOATING NODES
G1 1 3 VALUE={(PWR(V(7),EX)+PWRS(V(7),EX))/KG1}
RCP 1 3 1G ; TO AVOID FLOATING NODES
C1 2 3 {CCG} ; CATHODE-GRID
C2 2 1 {CGP} ; GRID=PLATE
C3 1 3 {CCP} ; CATHODE-PLATE
D3 5 3 DX ; POSITIVE GRID CURRENT
R1 2 5 {RGI} ; POSITIVE GRID CURRENT
.MODEL DX D(IS=1N RS=1 CJO=10PF TT=1N)
.ENDS 12LH7_triode_01
*$
A headphone amp (small power amp) circuit that could see >1V audio signal at its input should have a grid bias of at least -1.5V, and -2V would be better.
Also, with an OPT having limited primary inductance, you want the rp of the output triode to be as low as possible (for reasonably good bass response). That would mean fairly high plate current.
The combination of grid bias = -2V and Ip >= 35mA will result in fairly high plate voltage.
I'd also like the resulting headphone amp to be capable of delivering at least 100mW into a 50 ohm load.
According to the 12HL7 plate curves, plate voltage of 100V and grid voltage of -2V will yield Ip of 21mA. What would rp be in that situation? What would output power be in that situation?
Last edited:
Well yes, I did that, and I found that bass response (w/o NFB) was lacking if I ran the 12HL7-triodes with low plate current. I figured that was because the rp will be higher at lower plate current, causing a bit of bass rolloff with my low-budget Edcor OPTs (measured Lpri of only about 10H). I ran the tubes with higher plate current and measured slightly better bass response (presumably due to lower rp working into OPT's Lpri) and subjectively found the bass response was now adequate when driving Fostex T50RP cans (a 60 ohm load but low sensitivity, so requires a fair amount of power). I was happy with the way the amp sounded at this point, but was dismayed when I had to replace the 12HL7s every few months. So I bought a few 12GN7A to try, and those last noticeably longer in the circuit. That suggests 12GN7A has higher real-world plate+screen dissipation than 12HL7. Oh well.
Many thanks to Zoran for the LTspice models. I set them up and ran a simulation of the circuit. It looks like in my (very very simple) circuit, the 12HL7-Triode should run at about 155V at the plate, 2.7V at the cathode (62 ohm Rk. 2200uF Ck), Ip = 43.5mA (Pdiss = 6.7W). I now have three different 12HL7 models, and they're all in the same ballpark. The models work well.
BTW, there are some typos in Zoran's models. The tube type is misspelled as "12LH7" in a few places. Change that to the correct "12HL7" and all is well. Minor quibble.
Many thanks to Zoran for the LTspice models. I set them up and ran a simulation of the circuit. It looks like in my (very very simple) circuit, the 12HL7-Triode should run at about 155V at the plate, 2.7V at the cathode (62 ohm Rk. 2200uF Ck), Ip = 43.5mA (Pdiss = 6.7W). I now have three different 12HL7 models, and they're all in the same ballpark. The models work well.
BTW, there are some typos in Zoran's models. The tube type is misspelled as "12LH7" in a few places. Change that to the correct "12HL7" and all is well. Minor quibble.
Last edited:
re: Rongon
The 1st two RCA pics are the small ones, like I sent back. They didn't have as nice of triode curves as the bigger ones, and clearly lower Watts. I also have written in my notes for 12HL7 the mu factor (from just a few samples of big ones) RCA 40, GE 30, and Sylvania 25
Take a look at some 11HM7 pics on Ebay, they look just like the smaller versions.
From u = gm rp and gm = 21000 @ 25 mA
and gm approx. 21000 Sqrt (55/25) at 55 mA = 31148
rp at 55 mA would be approx. 1000000 30/31148 = 963 Ohms
The 1st two RCA pics are the small ones, like I sent back. They didn't have as nice of triode curves as the bigger ones, and clearly lower Watts. I also have written in my notes for 12HL7 the mu factor (from just a few samples of big ones) RCA 40, GE 30, and Sylvania 25
Take a look at some 11HM7 pics on Ebay, they look just like the smaller versions.
From u = gm rp and gm = 21000 @ 25 mA
and gm approx. 21000 Sqrt (55/25) at 55 mA = 31148
rp at 55 mA would be approx. 1000000 30/31148 = 963 Ohms
Last edited:
Thanks smoking-amp.
I figure 12HL7-triode or 12GN7A-triode with Ip = 45mA will have rp of about 1100 ohms or so. I figured with an OPT having primary inductance of only 8 to 10 Henries, I'd need that really low rp to get any bass response out of it.
I looked at some 11HM7 on ebay. Yup, they look pretty much identical.
I suspected the 'clear top' RCA 12HL7s were weaklings. I went through quite a few of them. I got them for $1 each, years ago, so I guess I can't complain. Maybe I'll try them at line level, with no more than 3W Pdiss.
I also have some other RCA 12HL7s with bigger innards and the getter flashing on top, along with some Sylvania and a couple of GE. I'll give them a try and see how long they last. Maybe they'll be OK.
Have you measured 12GN7A in triode? Is it a lot less linear than 12HL7-triode?
I figure 12HL7-triode or 12GN7A-triode with Ip = 45mA will have rp of about 1100 ohms or so. I figured with an OPT having primary inductance of only 8 to 10 Henries, I'd need that really low rp to get any bass response out of it.
I looked at some 11HM7 on ebay. Yup, they look pretty much identical.
I suspected the 'clear top' RCA 12HL7s were weaklings. I went through quite a few of them. I got them for $1 each, years ago, so I guess I can't complain. Maybe I'll try them at line level, with no more than 3W Pdiss.
I also have some other RCA 12HL7s with bigger innards and the getter flashing on top, along with some Sylvania and a couple of GE. I'll give them a try and see how long they last. Maybe they'll be OK.
Have you measured 12GN7A in triode? Is it a lot less linear than 12HL7-triode?
Last edited:
Here are some triode curves for a Zenith 12GN7, 50V/div Horiz., 10mA/div Vert., 2V steps
While looking for the 12GN7, I also came across one of the small RCA 12HL7 tubes I had saved, along with an RCA 11HM7 tube. They look the same.
I'm going to trace these to see how they compare with 12HL7. I also came across some 12HL7 tubes by International that have been marked over with 12GN7 on the box (over the 12HL7 printing, tube has 12HL7 printing too, but doesn't look right ). Will look at these on the tracer too. I wonder what else has been marked 12HL7 (probably at end of tube production)?
While looking for the 12GN7, I also came across one of the small RCA 12HL7 tubes I had saved, along with an RCA 11HM7 tube. They look the same.
I'm going to trace these to see how they compare with 12HL7. I also came across some 12HL7 tubes by International that have been marked over with 12GN7 on the box (over the 12HL7 printing, tube has 12HL7 printing too, but doesn't look right ). Will look at these on the tracer too. I wonder what else has been marked 12HL7 (probably at end of tube production)?
Last edited:
single ended triode amps can get away without global negative feedback,
re: bass, a lot to do with the OPT and its primary inductance, that is why OPT's are way much bigger in comparison with pp amps..
so your opt can be 3 x 963ohm or 2.8k, of course you can use 5k since it is a standard...