JCM800 2204 Phase Inverter issue and other

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello everybody.
My first post on this websites. I'm a guitarist and newbie in working tube amps (mostly a user), a small lab with basic instruments (a cuple aof DMM - one Fluke, one scope, one signal generator, a variac, a diy dummy load, and a lot of courage).
After small mods to taste new tones I decide to work on a JCM800 2204 clone I bought few years ago, heavily modified by the previous user, to bring it to the original state (I used as model the Mark Huss schematic - attached).
I substituted the EL34 and the great part of coupling capacitors.
Also a new 50+50uF electrolytic (the first in the chain) was introduced because a different value was present.
All the voltages are similar to the schematic: only 5% less, but starting from B+.
The sound tone is quite good, but it seems that the amp doesn't reach the full power.
I've a few questions about some discrepances I found on the idle values:
1) I substituted the 22n coupling capacitor of V1a (pin 1) because I measured around 700mVDC on the output part (should be near 0, if I'm not wrong). But also with the new cap the value doesn't change. The other coupling capacitors have generally not more than 10-15mVDC.
2) On the schematic and also digging on the web the voltages on pin 1 & 6 of
the 12AX7 phase inverter have to be similar. In my 2204 I've 215VDC on pin 1 and 255VDC on pin 6. I swapped the tubes, changed with Others, tested all the resistors, resoldered the Presence part bot with no results. A difference with the attached schematic is also the voltage after the 100K NFB resistor. I have 14.9Vdc (expected 13.5Vdc, consider that all my voltages are -5% starting from the PT) and -2Vdc before the resistor. I don-t know hot to balance the PI and if a PI unbalanced can generate the loss of power.
Thanks in advance for your comments and helps.
 

Attachments

  • MH2 - JCM800_2204.gif
    MH2 - JCM800_2204.gif
    196.1 KB · Views: 413
Moderators may move this thread to the Instruments and Amps forum - but don't worry about that - it happens all the time.

Check the DC voltage at the grid of V2A - a bad connection somewhere there could leave that grid 'floating'.

Similarly the voltage of 215V on one of the anodes of the LTP PI suggests that triode is conducting more than expected. Again, check the grid voltage of that triode to see if it is not getting referenced properly via its 1M grid leak resistor. I think the higher than expected current here could also explain the high voltage at the top of the presence pot.
 
Last edited:
Malcolm, Koonw, thanks for your suggestion.
About the pre V1/V2.
The 22n on pin1 was the first cap I replaced, supposing a leakage, but nothing changed. Now the voltage after the 22n coupling cap of pin 1/V1a is around 1V - 750mV at pin 2 (grid) of V2a. Other values:
V1a/pin1 = 277V
V1a/pin3 = 2,85V
V2a/pin1 = 160V
V2a/pin3 = 1.37V

About the PI I'm sorry but I gave a wrong indication. The voltage before the NFB resisotor is not -2V but -2mV (pratically OV, the DMM was on auto-scale and I don't be aware of the mV light). Anyway I confirm 14.7V after the NFB resistor, 45V between the 1M leak resistors, aroung 29V on pin 2 and 30V on pin 7, and around 40V of difference between plates.
The grounding of output jacks and of the OT was checked and is good, too.
Both the coupling caps to power tubes were replaced, too.
 
Disabled Account
Joined 2013
The 22n on pin1 was the first cap I replaced, supposing a leakage, but nothing changed.

Because it is such small leakage, it's easy for another cap to have the same leakage. So the better way is to disconnect 22n on pin-1 of V1a or unplug V1 so doing so that will remove any positive voltage asserted on the cap. Check again when done, there shouldn't be any positive on grid of V2a since you have remove the only (or one of the) positive voltage source.

Here is the simulation files for LTSpice users:
 

Attachments

  • JCM800.asc
    6.2 KB · Views: 100
Disabled Account
Joined 2013
Download all the tube model from here. Unzip to "sub" directory of LTspice. Or paste the following code of one tube model directly into using directive "s" or save the code as "12ax7.inc" in "sub". Remove .txt extension of the attached file and place in "sub" directory.

Code:
*
* Generic triode model: 12AX7
* Copyright 2003--2008 by Ayumi Nakabayashi, All rights reserved.
* Version 3.10, Generated on Sat Mar  8 22:41:09 2008
*             Plate
*             | Grid
*             | | Cathode
*             | | |
.SUBCKT 12AX7 A G K
BGG   GG   0 V=V(G,K)+0.59836683
BM1   M1   0 V=(0.0017172334*(URAMP(V(A,K))+1e-10))**-0.2685074
BM2   M2   0 V=(0.84817287*(URAMP(V(GG)+URAMP(V(A,K))/88.413802)+1e-10))**1.7685074
BP    P    0 V=0.001130216*(URAMP(V(GG)+URAMP(V(A,K))/104.24031)+1e-10)**1.5
BIK   IK   0 V=U(V(GG))*V(P)+(1-U(V(GG)))*0.00071211506*V(M1)*V(M2)
BIG   IG   0 V=0.000565108*URAMP(V(G,K))**1.5*(URAMP(V(G,K))/(URAMP(V(A,K))+URAMP(V(G,K)))*1.2+0.4)
BIAK  A    K I=URAMP(V(IK,IG)-URAMP(V(IK,IG)-(0.00058141055*URAMP(V(A,K))**1.5)))+1e-10*V(A,K)
BIGK  G    K I=V(IG)
* CAPS
CGA   G    A 1.7p
CGK   G    K 1.6p
CAK   A    K 0.5p
.ENDS
 

Attachments

  • potentiometer_standard.lib.txt
    4.9 KB · Views: 75
Last edited:
Thank's for the help. I am almost there, I still get a message, Couldn't find symbol(s): potentiometer_standard. I thought I placed the file in the folder right, did not know how to get rid of the txt extension but I read to put " " around what you want and save it. It looks right (like I really know) but still does not seem to be playing happily with the asc file.


RCLxclG.png
 
Disabled Account
Joined 2013
Removed .txt extension from attached file and save in "sym" directory.

LTspice use .asc for simulation file, .inc or .lib for model file, and .asy for symbol. These files are not native but created by third parties for sharing.
 

Attachments

  • potentiometer_standard.asy.txt
    652 bytes · Views: 52
Other test and changes - with attachments

Hello guys. First of all thanks for your patience and your suggestions. In the following the results of the new check and changes.

V1a/pin1 22n coupling cap (without signal applied, all pots on 0).
I test the voltage after the cap removing V1 or V2 and it was Always around 1Vdc. I replace the cap with a brand new one but I still I've 0,55Vdc. What do you think about? Attached a couple of picture.

V3 Phase Inverter output (signal applied)
I attached the images of the scope dual trace. The signal are referred at a 400Hz signal on low input, both preamp volume and master volume on 12 o'clock. As you can see one signal is lower than the other, around 10% less. The same at all volumes changes.

Power tubes (without signal applied, all pots on 0)
The Bias current is now around 27.2 on one tube and 28.7 on the other. The tubes are a couple of new JJ. I replaced them as first step. the load is a dummy load 16Ohm, pure resistive.

Hope this can help. Have a nice evening.
 

Attachments

  • 20180918_200338.jpg
    20180918_200338.jpg
    252.9 KB · Views: 64
  • 20180918_200326.jpg
    20180918_200326.jpg
    412.5 KB · Views: 61
  • 20180918_192655.jpg
    20180918_192655.jpg
    277 KB · Views: 135
  • 20180918_191951.jpg
    20180918_191951.jpg
    254.3 KB · Views: 138
  • 20180918_191913.jpg
    20180918_191913.jpg
    262.5 KB · Views: 125
Disabled Account
Joined 2013
That makes sense. I'm still not feeling the love though.
Click on "Component" and then type "pot" to locate it, insert manually. If not found, restart the program.

I test the voltage after the cap removing V1 or V2 and it was Always around 1Vdc. I replace the cap with a brand new one but I still I've 0,55Vdc. What do you think about?


When V1 is removed you still got 1V, because the HT is still there via 100k. Any voltage (on top of 470k, where it was soldered) when 22n is removed? Try this first.

(Since the voltage after each cap is not the same, probably due to inconsistent leakage as shown by voltages behind it.)

V3 Phase Inverter output (signal applied)
I attached the images of the scope dual trace. The signal are referred at a 400Hz signal on low input, both preamp volume and master volume on 12 o'clock. As you can see one signal is lower than the other, around 10% less. The same at all volumes changes.

Try to trace again when NFB is disconnected or unplug the power tubes so there is no more feedback and see if you got symmetry waveform.

(If waveform is symmetry then the output stage is not. If so try to bias the output tube a little or change to another tube, because there is mismatch of the tube or something else.)
 
When V1 is removed you still got 1V, because the HT is still there via 100k. Any voltage (on top of 470k, where it was soldered) when 22n is removed? Try this first.

(Since the voltage after each cap is not the same, probably due to inconsistent leakage as shown by voltages behind it.)

I unsoldered the 22n cap from V1a/pin1 and the voltage on the other side of the cap is still 1V. So I take all the measures of V2 pins:
1 --> 165V
2 --> 750mV
3 --> 1.37V
6 --> 328V
7 --> 165V (connected to 1)
8 --> 165V
It seems that cap doesn't leak. Where is this 1V coming from?
I'm confused!

Try to trace again when NFB is disconnected or unplug the power tubes so there is no more feedback and see if you got symmetry waveform.

(If waveform is symmetry then the output stage is not. If so try to bias the output tube a little or change to another tube, because there is mismatch of the tube or something else.)

In attachment the scope waveforms before the grid resistors, with the NFB cable detached. As you can see at medium volume (pre and master at 12 o'clock) the two waves are still different, but raising both the volume to the max the difference seams to disappear. I reconnect the NFB and the behavior is the same.

What is the exact output voltage into what load? I think you want about 28v peak into an 8 ohm load, or 20v into a 4 ohm load. Also make sure you are using the correct secondary tap.

I confirm the my load is 16.2 Ohm, pure resistive.
I make two measurments (you can see in the attachment):
- low input --> sinusoidal signal, 400Hz, 200mVpp
- output (with pre and master at max) --> 10Vpp
 

Attachments

  • 20180919_190036.jpg
    20180919_190036.jpg
    278.3 KB · Views: 84
  • 20180919_190155.jpg
    20180919_190155.jpg
    279 KB · Views: 84
  • 20180919_192640.jpg
    20180919_192640.jpg
    276.4 KB · Views: 93
Disabled Account
Joined 2013
I unsoldered the 22n cap from V1a/pin1 and the voltage on the other side of the cap is still 1V. So I take all the measures of V2 pins:
1 --> 165V
2 --> 750mV
3 --> 1.37V
6 --> 328V
7 --> 165V (connected to 1)
8 --> 165V
It seems that cap doesn't leak. Where is this 1V coming from?
I'm confused!

Please measure again where V2 is also removed (while leave 22n unsoldered and V1 removed).

(If 1V disappeared after V2, V1 and 22n are removed, then V2 might have some grid abnormal current leak). With 1V on top 470k, it's 323mV on grid of V2, so this 1V must come from top of 470k. in other word if 323mV come from grid of V2, you couldn't have measured 1V on top of 470k.

In attachment the scope waveforms before the grid resistors, with the NFB cable detached. As you can see at medium volume (pre and master at 12 o'clock) the two waves are still different, but raising both the volume to the max the difference seams to disappear. I reconnect the NFB and the behavior is the same.

The difference is normal as the plate resistors are not identical (82k/100k), so output at 100k would be slightly higher. The outputs of phase splitter clip after they hit the grid swing limit, and both can not swing any higher, hence the difference is narrow down. (NFB suppose to correct the difference)

I confirm the my load is 16.2 Ohm, pure resistive.
I make two measurments (you can see in the attachment):
- low input --> sinusoidal signal, 400Hz, 200mVpp
- output (with pre and master at max) --> 10Vpp

The crossover point is asymmetrical, the top (or left) tube contributed less than the bottom (or right) tube, it appears weaker by comparison. So might be why NFB has little effect as it is beyond the scope of correction.

(There might be other reasons, such as shorted turns in transformer winding or bias is out?)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.