EVjr head mods

pchilson · May 19th, 2008, 10:00 PM

Hi,

I guess this thread is more informational than anything. Just thinking if any of you have this amp and want to share or contribute what you like/dislike or mods you've done we can do it here.

I'll start.

This past weekend I picked up a used ver 3 Valve Jr head. I've been doing a lot of reading up on modding these. The first thing to take care of is the biasing of the EL84 output tube, its said they are hot from the factory and you don't want more than 12 watts plate dissapation on this tube. With the help of Vintage Jon of Tonecraft Amp Repair in Austin, Texas I have done the measurments, calculations and found I had an IPD of 15.42 watts. HOT, HOT, HOT

I have changed out the R14 resistor from the stock 2w 220 ohms to a 5w 330 and the measurements and calculations now give me 12.17 watts. This is good I guess, at least better than the tube eating 15.42 from before.
I still have to get info back from Jon as to what more I need to check or do before I can move on to some voicing mods. I want to try the Marshall or Voxy mods on this. This is a great little inexpensive test and trial mod platform.

I am running the head through a concocted speaker cab with a 16 ohm Celestion G12T-75. Before the re-bias the amp was rather dark sounding, after the re-bias it has brightened up considerably. I also have an 8 ohm G12-35 MG that I can try running this through.

Ok, whos next...

Pic 1: Stock R14 @ 220 ohms
Pic 2: Mod R14 @ 330 ohms
Click to enlarge

jefrs · May 19th, 2008, 10:43 PM

The Mullard book of 1969 says EL84 pa max 12W. But that's continuous dc. As a rule, the valve does not conduct continuously, if it does it goes red plate. So that's why they fit 220 ohm. The R14 cathode bias mod is normal.
Attachment 11624
Loads and loads of EVJ stuff on fora over at sewatt.com (also wattkins.com and 18watt.com).

What impedance is your o/p transformer? My new EVJH has the 7.5k (it's printed on the label). Did you take a pic of the entire board? I fancy the head uses a different circuit to the combo.

pchilson · May 19th, 2008, 11:18 PM

My ver3 head has the 5k impedance transformer in it.
I am doing the biasing per Vintage Jon and of the reading I've done over at sewatt.com
They all are suggesting to lower the plate dissapation watts to, some say 90% of 12watts but no greater than that. I do this for fun and I'm going by what experienced amp techs are saying. I'm all ears though.

jefrs · May 20th, 2008, 08:34 AM

Thanks for the pics. That board has the identical version nymber to mine. But it only has one set of diodes and appears to use ac heaters like the earlier version 1.
Sorry, I posted the wrong schematic. The one I wanted to load was StockVJ_1-3.pdf but it is too big for this site - so I've converted to jpg...
Attachment 11633
Attachment 11634
Attachment 11635
And one of the EVJhead
Attachment 11632

Regarding R14, its purpose is to put the desired bias voltage on the cathode, not limit the current (which it does too). 220 is a bit low, the value depends on the HT. Our EU supply transformer is set to 240Vac, it should be set to 250Vac as our supply is 220-240V /minimum/ - EU is ok but we are going to get too much HT here (UK and Ireland) from 248Vac. The valve only passes current when it's conducting, and you will only get 5 watts clean through a class A EL84. So unless we manage to switch the valve permanently full on, we are not going to get 12 watts, let alone 15 watts dissipated through it - as heat in the valve and the o/p primary. And yes, I've seen it done - not pretty but the transformer survived (they can take full mains current).

Mine came with Sovtek 7025/12AX7WA and Sovtek EL84/6BQ5. I swapped out the 7025 for:- (worst to best)
Brimar ECC83 (well used rocking horse poo) - quiet, full tone but way too much gain
TAD 7025/E83CC Highgrade ($$$) - not as quiet, tone ok but too much gain
Mazda ECC83 (used) - poor tone
TAD 12AX7A-C Selected - tone lacking oomph
RFT ECC83 Balanced - medium gain, good tone - almost there
The first three lack something, the last two were acceptable, which is surpising considering the pedigree of the first ones.
And the winner was... the original Sovtek.

I also added the simple 'Tweed' tone control but this was a tone sucker and got ripped out. I think the R6/R7 voltage divider has to be replaced as in the GA5 for this to work. GA5 does not always have R6/R7 at all.

Pvee · May 20th, 2008, 11:20 AM

I made several mods to mine. Added some extra filtering for the B+ voltages and lowered B1 about 20 volts.

It's a winter time project for me.

I think there is a signal level issue in mine, maybe all of them. This was before the mods also.

I can over drive or cause it to distort with it around 1/2 volume and guitar output high..

With the volume control all the way up, it seemes to be a cleaner tone and that's what I am after.
Not really a big probem or anything but I want to get it so that t is pure and clean at all volume pot settings.

I can get a good big fat sound out of it with the 10" speaker cabinet I built.

Over all I am satisfied with it. It's just another project to play with.

pchilson · May 20th, 2008, 04:23 PM

Hey Pvee, what did you do to filter B+ and what was your B1 before and after the "lowered 20 volts"?
What version board is yours?

After re-biasing mine with the 330ohm R14 resistor the plate dissapation lowered to 12.17, the voltages went up and the gain increased, almost unuseable at upper volumes. I'm told this should have gotten cleaner and rounder sounding instead. I don't know but I think I need to do what you did with B+ and B1.
I mean, doesn't 330 ohm in R14 seem like an awful high value and still be over 12 watts? I think the voltages are to high and if I can lower them then maybe 270 ohm or so in R14 might be more in line...What do you think?

Pvee · May 20th, 2008, 06:12 PM

I changed R10 to 1 K ohm. it was 220 ohms. That affects all the high voltages but does not seem to be a problem.
I think B1 went from around 340 to 320 volts.

I added another 1000uf cap parallel with R 11.

Changed C5 to 1000uf. Changed R 15 to 3.3k.

Changed C3 to .047uf.

This is all just experimental. It works pretty well right now. I just got through playing through it. I'll get back to it again in the fall.
I wish I still had my scope and signal generator.!!

I think it needs some minor signal level adjustment in the volume control and
R6 / R7 area.

Some guys change the R1 and R2 resistors but i don't think that is needed.

I also changed the output transformer to the Hammond 125cse and did not see much of a change there so I think the stock one was pretty good in my case.

If I put an EQ pedal and delay pedal in front of it, it really sounds good.

Quote:

Originally Posted by pchilson

Hey Pvee, what did you do to filter B+ and what was your B1 before and after the "lowered 20 volts"?
What version board is yours?

After re-biasing mine with the 330ohm R14 resistor the plate dissapation lowered to 12.17, the voltages went up and the gain increased, almost unuseable at upper volumes. I'm told this should have gotten cleaner and rounder sounding instead. I don't know but I think I need to do what you did with B+ and B1.
I mean, doesn't 330 ohm in R14 seem like an awful high value and still be over 12 watts? I think the voltages are to high and if I can lower them then maybe 270 ohm or so in R14 might be more in line...What do you think?

jefrs · May 20th, 2008, 09:21 PM

Is the rectangular object next to the orange wires a bridge rectifier? If so then VER051122 EVJH board has dc heaters, doh. But why have I got a 7.5k o/p transformer? What are your R1, R8 and R9? Is this version 2 or 3 or something in between?

Now if we have actually got 15 watts dissipated then that current is passing through R14 220ohm 2W resistor - it should be awfully hot by now. What are your measurements? I have not delved into my board with a meter yet.

Mullard EL84 data tells us Va=250V (anode to cathode) and Ia=48mA (12W) and Rk=135ohm with Ra=4.5kohm. What voltage do you have at anode and cathode wrt earth?

Scott S · May 20th, 2008, 10:30 PM

Um, if I'm not mistaken, this is a single-ended class A amplifier, and the output tube SHOULD be conducting at all times. That's why the idle dissipation should be 12W or less in this amp if tube life is a concern.

- Scott

jefrs · May 21st, 2008, 10:48 PM

Correct, it's single-ended so it has to be Class A = the anode current is set at a level such that even with the maximum allowable input signal, the anode current never falls to zero, i.e. the valve never switches off.

But if it were pulling 12 to 15 watts then it should be running much hotter, literally as heat, than it is. The Laney L5T, single EL84 uses a 150ohm cathode bias, but I dont know what the supply rail is.

Which is why I was interested in the actual voltage-to-earth measurements in the chain:- supply rail; anode; cathode. From these we can calculate the power dissipation in the o/p transformer, the EL84 and the R14 bias resistor.

pchilson · May 22nd, 2008, 08:57 AM

jefrs,

Sorry, haven't even looked at the thread to see your question.
I'm too much of a newb at this to know what measurements you are looking for.
Tell me what to measure and I'll be happy to measure them.
I'm going in tonite to change R14 back to the 220 ohm from the 330 ohm currently in there. I've been told it doesn't make any sense but with R14 at 330 the gain has increased so much as to make the amp almost unuseable. I need to put it back stock and perhaps look at lowering some of the B voltages first before changing R14 again.
I'm just trying to learn this stuff so you need to be specific if you need values from me...

jefrs · May 22nd, 2008, 04:03 PM

PChilson,
Right, R14 is used to bias the cathode with respect to earth so the grid is negative in respect of it. This is cathode bias. Some amps supply the grid with a negative voltage e.g. PV Classic, this is grid bias. This bias defines the gain. Changing the bias resistor to limit current does not make sense. If it is running hot, because it is dissipating too much power, then anode has too many volts on it - so lower its supply rail voltage. Lower the voltage too much and we get starved plate (anode), in a hi-fi amp this is anethema but may be desireable for guitar. I may measure some volts on my un-modded amp tonight. I am disinclined to mod the bias because it sounds just right.

pchilson · May 22nd, 2008, 06:46 PM

Ok, here is a screenie from the vjr__bias_calculator.xls

None of the values seem to be correct according to the schematic. The idle plate dissapation is however on the mark.
My problem is I don't know how to bring the other values inline and at the same time keep the plate dissapation where it should be. Every value you change seems to change every measurement.
Who has some direction here?

Thanks

Click to enlarge

jefrs · May 22nd, 2008, 08:35 PM

Ok, here are my measurements FYI
AC Mains 242VAC (UK)
B+ 335Vdc
B1 346Vdc
B2 306Vdc (screen grid2 pin 9)
B3 291Vdc
Anode plate pin 7 316Vdc
Cathode pin 3 9.76Vdc
Bias resistor R14 220 ohm
Calculated cathode current 9.76/220 = 44mA
Total dissipation in valve 306 x 0.044 = 13.46W
Assume screen dissipation 1.32W
Net anode plate dissipation 13.46 - 1.32 = 12.14W
But screen dissipation could be from 1mA (0.3W) to 5.5mA (1.6W)
If screen dissipation >1.5W then Anode dissipation <12W...
I shall have an expert I know check my findings.

So my EL84 is running at full chat, but not hot enough to warrant changing the bias. B+ was measured as d.c. but it is not fully smoothed, so reads lower than B1 which is. The Mullard data states values for when Va=250V, but this is not a maximum, I have got 306Vdc across A-C, this is hot but not unreasonable. It sounds good and Sovtek EL84 are not expensive.

Note my EVJH was bought recently 2008, and has board 'VER 051122':-
Attachment 11695
It has R1 = 68k, R8 = 2.2k, R9= 2.2k, and 7.5k o/p transformer, this is Version 2.
Why is this Version 2 in a new amp?
Is that VER a 2005 date?
Are all EVJ Heads version 2?

I have re-mounted the EVJH in an old radio case, with an 8-in Elac speaker (£2/12/6). To do this I had to attach a new front panel and extend the jack and volume through to it. The mains switch has been moved to the back of the case. The original jack can be seen lower left with the cable looped out of the way

pchilson · May 23rd, 2008, 08:27 PM

I'm certainly no expert on this but that green pcb is indicative of a ver 1 and ver 2 boards. The ver 3 boards are dark brown. So I would say you got NOS amp, that would explain the 7.5k OT. When you look at the schematics the ver 2 is much closer to the ver 3 than the ver 1.

I think my next move is to bring B1 down to 300VDC and then work from there. Just have to go to the parts store and get a few to play with. I'm not at all sure this is something that is needed if the plate dissapation is under 12 watt but I see lots of reference to it so I will do it.

I also have a high reading on the heater measurements 6.3v by spec and mine is running 7.4. I do know one thing, when I bumped R14 to the 330 ohm resistor all my voltage readings went up, so I don't know at this point if the heater voltage went up then as well. I may need to drop the stock 220 ohm resistor back in R14 just so I can measure all my baseline stock values.

Remember, I'm learning as I go here...

pchilson · May 23rd, 2008, 10:27 PM

Ok, brought the amp back to bone stock and took measurements again to establish a baseline.
Heater voltage is too high. How to bring it down to 6.3v?
Plate dissapation is too high. Lower this with R14 but as I said above I want to bring B1 down to 300V first. Will try some larger values in R10 this weekend maybe and go from there.

pchilson · May 24th, 2008, 07:27 PM

Today I bought some caps and resistors to play with.
Here is what I ended up with after several trial and error attempts.

R10 = 1.18k (from the Basic Mod)
R14 = 216 ohm (Bias resistor)
C4 & C5 = 2.2uf/50V (from the Marshall Mod)
R1 = removed from pcb and placed across input jack (from the Basic Mod)

IPD is in good shape now but the heaters are still over spec at 7.1V. I don't know how to reduce this. Is this a result of the wall voltage being 123VAC and the PT being rated at 115VAC? Would leaving it as is be destructive to the tube? Does this heater value introduce anything to the sound?

Thanks

jefrs · May 24th, 2008, 09:17 PM

pchilson, you board pic is green, it's ver2 also?

I'm no expert either, it's a hobby for me, but has been since I was 8 or so, and it is related to my work. However I am fortunate to know several chaps who are genuine experts, and whom I pester mercilessly for the public good whenever the opportunity arises. I'm still not too old to learn.

I don't know what 'IPD' is, we appear to be divided by a common langauge.

I suggest the the heater voltage is unsmoothed dc, your 7.4V is peak, switch you meter to ac and re-measure, I guess average will be around 6.3V and ok. The heaters are taken from a separate secondary winding and circuit, thus independent of R14.

Tricky thing bias. The grid must be negative in respect to cathode. If we change this voltage with R14 then we change the gain curve. On spec sheets we often see a nest of graph curves for different bias on gain vs anode voltage plot. We chose one that will give us a linear gain (or being a guitar amp, chose one that doesn't). Your gain changed because you changed the bias.

If your mains is 123VAC and your transformer tap is 115VAC then you will get more volts out of the secondary, HT supply (278VAC?). The ultimate solution is a regulated and stabilized power supply (plain daft), or just a bigger resistor at R10. Now the RC network is too complicated for me to work out (I know it can be modelled in a spreadsheet) the value of R10 to drop B1 below 340VDC. Try next standard value, 240ohm or 270ohm [E12(E24) series180 (200) 220 (240) 270 (300) 330 (360) 390]. But this will drop all B1 B2 B3 HT voltages. The trouble with R10 is that changing it will be interactive as you found when R14 altered all the voltages. Get B1 right then experiment with R12 and R13 to fix B2 and B3.

Our mains voltage varies a great deal, especially at early evening peak when the voltage drops as all the cookers and tellies in the country go on. They used to regulate to 245VAC and let the frequency dip but that threw all the clocks out, so now they maintain 50Hz and can drop to 220VAC. That's rare though, most of the time it's above 240VAC

Was your R1 not 680k? R1/R2 is a voltage divider. Putting R1 in front of R2 decreases the input impedance and decreases signal to V1 grid. Does it make much difference to the sound?

What about those mini-Marshall capacitors? What effect did they have?

Today I refurbished the radio-case cabinet and fitted a brass ventillation grille in the door where the tuning dial used to be, so far it has not got hot enough to warrant one at the back, so it's a closed-back combo. The Elac sounds real pretty in there, so I'm disinclined to do too many mods to the circuit, but I would still like some variation in tone, or maybe gain.
Attachment 11765 Attachment 11766
The colour's a bit wrong, the walls are a creamy yellow, not pink...

jefrs · May 24th, 2008, 09:36 PM

I misread your question re heaters, yes, increased mains will increase volts from heater windings too. But the schematic shows 120VAC primary, your is labelled 115V? Are there any other mains taps on the transformer label? A small resistor in series with the heaters will drop the volts but would need to be at least 6Watt. Value depends on resistance of heaters, which gets less as they heat up. They are probably ok as they are, at least the valves are cheap. I would only fix it if it blows heaters. It probably has some minor effect because heaters do emit a few electrons and running hotter will effect cathode emission. You may observe the tone changes as amp warms up and after a few hours on.

pchilson · May 25th, 2008, 12:04 AM

Ok, I'll try to get to all your points and questions.

Quote:

pchilson, you board pic is green, it's ver2 also?

Mine is a version 3 board, it is a dark brown not green PCB.

Quote:

I don't know what 'IPD' is, we appear to be divided by a common langauge.

IPD = Idle plate dissipation, terminology from another fellow I'm working with.

Quote:

I suggest the the heater voltage is unsmoothed dc, your 7.4V is peak, switch you meter to ac and re-measure, I guess average will be around 6.3V and ok. The heaters are taken from a separate secondary winding and circuit, thus independent of R14.

I'm measuring heater voltage at pin5 of the EL84 and pin5 of the 12AX7. I'm under the impression this is DC not AC. Are you saying to measure these pins with the meter set to AC and I would supposedly see 6.3VAC?

I've also been given some other info on ways to bring this value down using diodes in parallel with reverse polarity. Waiting for follow up on this.

Quote:

Tricky thing bias. The grid must be negative in respect to cathode. If we change this voltage with R14 then we change the gain curve. On spec sheets we often see a nest of graph curves for different bias on gain vs anode voltage plot. We chose one that will give us a linear gain (or being a guitar amp, chose one that doesn't). Your gain changed because you changed the bias.

I definitely noticed a gain increase with the bump in the value of R14 but now I'm back to the stock 220 ohm in R14.

Quote:

If your mains is 123VAC and your transformer tap is 115VAC then you will get more volts out of the secondary, HT supply (278VAC?). The ultimate solution is a regulated and stabilized power supply (plain daft), or just a bigger resistor at R10. Now the RC network is too complicated for me to work out (I know it can be modelled in a spreadsheet) the value of R10 to drop B1 below 340VDC. Try next standard value, 240ohm or 270ohm [E12(E24) series180 (200) 220 (240) 270 (300) 330 (360) 390]. But this will drop all B1 B2 B3 HT voltages. The trouble with R10 is that changing it will be interactive as you found when R14 altered all the voltages. Get B1 right then experiment with R12 and R13 to fix B2 and B3.

I changed R10 from the stock 220 ohm to a 1.2k value and it dropped the voltage considerably. Since that time I've been given reason to believe that R11 is a more appropriate location to do the voltage drop from. I'm not sure of the reasoning as it was stated R11 is parallel and R10 is series. I don't know what this means but I'm waiting for followup to my questioning on this. Reliable source is telling me this.
I'm also told that increasing these values increases the heat as the voltage drop is dissipated through heat and this could be hazardous to the PCB. Perhaps increasing both locations R10 and R11 would net the voltage drop desired but not produce as much heat as you are not dropping as much voltage from each location?

Quote:

Was your R1 not 680k? R1/R2 is a voltage divider. Putting R1 in front of R2 decreases the input impedance and decreases signal to V1 grid. Does it make much difference to the sound?

On the version 3 board R1 is 1M value. I'm led to believe placing R1 in front of R2 will not drop as much signal and brighten the sound. It has done so for me unless this is suggestive hypnotism...hehe

Quote:

What about those mini-Marshall capacitors? What effect did they have?

I put these in C3 and C4 as suggested by the Marshall Mod which states "Lower C4 & C3 to tighten bass. 1uf for humbuckers and 2.2uf for single coils."
My Strat was sounding quite raspy/harsh and quite dark sounding before. My LP sounded heavenly though. After this change the Strat greatly improves and the LP is still heavenly...

Quote:

Today I refurbished the radio-case cabinet and fitted a brass ventillation grille in the door where the tuning dial used to be, so far it has not got hot enough to warrant one at the back, so it's a closed-back combo. The Elac sounds real pretty in there, so I'm disinclined to do too many mods to the circuit, but I would still like some variation in tone, or maybe gain.
Attachment 11765Attachment 11766
The colour's a bit wrong, the walls are a creamy yellow, not pink...

That is quite the enclosure you have there. Very nice and creative.

Quote:

I misread your question re heaters, yes, increased mains will increase volts from heater windings too. But the schematic shows 120VAC primary, your is labelled 115V? Are there any other mains taps on the transformer label? A small resistor in series with the heaters will drop the volts but would need to be at least 6Watt. Value depends on resistance of heaters, which gets less as they heat up. They are probably ok as they are, at least the valves are cheap. I would only fix it if it blows heaters. It probably has some minor effect because heaters do emit a few electrons and running hotter will effect cathode emission. You may observe the tone changes as amp warms up and after a few hours on.

I don't know about other mains taps, I haven't known to look. Next time I have it on the bench I have a look see.
As I said above I've been given a way of dealing with the heater voltage. Either method of diode or resistor is a solution only when the mains are higher than what the PT is rated at. If I use it in another location where the mains are higher or lower, then its wrong again. No real good solution to this huh?

jefrs · May 25th, 2008, 11:57 AM

Quote:

Originally Posted by pchilson

Mine is a version 3 board, it is a dark brown not green PCB.

It looks green on my screen but not as green as mine, yours has extra writing to left of the "VER051122"

Quote:

Originally Posted by pchilson

I'm measuring heater voltage at pin5 of the EL84 and pin5 of the 12AX7. I'm under the impression this is DC not AC. Are you saying to measure these pins with the meter set to AC and I would supposedly see 6.3VAC?

Yes, measure with meter on AC because the heaters are rectified but not smoothed

Quote:

Originally Posted by pchilson

I've also been given some other info on ways to bring this value down using diodes in parallel with reverse polarity. Waiting for follow up on this.

Sounds good. But you may not need to bother.

Quote:

Originally Posted by pchilson

I definitely noticed a gain increase with the bump in the value of R14 but now I'm back to the stock 220 ohm in R14.

R14 alters V2 gain. I've seen a mod using a gain switch at R8 or R9 on V1.

Quote:

Originally Posted by pchilson

I changed R10 from the stock 220 ohm to a 1.2k value and it dropped the voltage considerably. Since that time I've been given reason to believe that R11 is a more appropriate location to do the voltage drop from.

May 19th, 2008, 10:43 PM	#2 (permalink)
jefrs Tele-Holic Join Date: Nov 2007 Location: Newbury, England Age: 54 Posts: 809	The Mullard book of 1969 says EL84 pa max 12W. But that's continuous dc. As a rule, the valve does not conduct continuously, if it does it goes red plate. So that's why they fit 220 ohm. The R14 cathode bias mod is normal. Attachment 11624 Loads and loads of EVJ stuff on fora over at sewatt.com (also wattkins.com and 18watt.com). What impedance is your o/p transformer? My new EVJH has the 7.5k (it's printed on the label). Did you take a pic of the entire board? I fancy the head uses a different circuit to the combo. Last edited by jefrs; June 9th, 2008 at 08:31 PM.

May 20th, 2008, 08:34 AM	#4 (permalink)
jefrs Tele-Holic Join Date: Nov 2007 Location: Newbury, England Age: 54 Posts: 809	Thanks for the pics. That board has the identical version nymber to mine. But it only has one set of diodes and appears to use ac heaters like the earlier version 1. Sorry, I posted the wrong schematic. The one I wanted to load was StockVJ_1-3.pdf but it is too big for this site - so I've converted to jpg... Attachment 11633 Attachment 11634 Attachment 11635 And one of the EVJhead Attachment 11632 Regarding R14, its purpose is to put the desired bias voltage on the cathode, not limit the current (which it does too). 220 is a bit low, the value depends on the HT. Our EU supply transformer is set to 240Vac, it should be set to 250Vac as our supply is 220-240V /minimum/ - EU is ok but we are going to get too much HT here (UK and Ireland) from 248Vac. The valve only passes current when it's conducting, and you will only get 5 watts clean through a class A EL84. So unless we manage to switch the valve permanently full on, we are not going to get 12 watts, let alone 15 watts dissipated through it - as heat in the valve and the o/p primary. And yes, I've seen it done - not pretty but the transformer survived (they can take full mains current). Mine came with Sovtek 7025/12AX7WA and Sovtek EL84/6BQ5. I swapped out the 7025 for:- (worst to best) Brimar ECC83 (well used rocking horse poo) - quiet, full tone but way too much gain TAD 7025/E83CC Highgrade ($$$) - not as quiet, tone ok but too much gain Mazda ECC83 (used) - poor tone TAD 12AX7A-C Selected - tone lacking oomph RFT ECC83 Balanced - medium gain, good tone - almost there The first three lack something, the last two were acceptable, which is surpising considering the pedigree of the first ones. And the winner was... the original Sovtek. I also added the simple 'Tweed' tone control but this was a tone sucker and got ripped out. I think the R6/R7 voltage divider has to be replaced as in the GA5 for this to work. GA5 does not always have R6/R7 at all. Last edited by jefrs; June 9th, 2008 at 08:31 PM.

May 22nd, 2008, 08:35 PM	#14 (permalink)
jefrs Tele-Holic Join Date: Nov 2007 Location: Newbury, England Age: 54 Posts: 809	Ok, here are my measurements FYI AC Mains 242VAC (UK) B+ 335Vdc B1 346Vdc B2 306Vdc (screen grid2 pin 9) B3 291Vdc Anode plate pin 7 316Vdc Cathode pin 3 9.76Vdc Bias resistor R14 220 ohm Calculated cathode current 9.76/220 = 44mA Total dissipation in valve 306 x 0.044 = 13.46W Assume screen dissipation 1.32W Net anode plate dissipation 13.46 - 1.32 = 12.14W But screen dissipation could be from 1mA (0.3W) to 5.5mA (1.6W) If screen dissipation >1.5W then Anode dissipation <12W... I shall have an expert I know check my findings. So my EL84 is running at full chat, but not hot enough to warrant changing the bias. B+ was measured as d.c. but it is not fully smoothed, so reads lower than B1 which is. The Mullard data states values for when Va=250V, but this is not a maximum, I have got 306Vdc across A-C, this is hot but not unreasonable. It sounds good and Sovtek EL84 are not expensive. Note my EVJH was bought recently 2008, and has board 'VER 051122':- Attachment 11695 It has R1 = 68k, R8 = 2.2k, R9= 2.2k, and 7.5k o/p transformer, this is Version 2. Why is this Version 2 in a new amp? Is that VER a 2005 date? Are all EVJ Heads version 2? I have re-mounted the EVJH in an old radio case, with an 8-in Elac speaker (£2/12/6). To do this I had to attach a new front panel and extend the jack and volume through to it. The mains switch has been moved to the back of the case. The original jack can be seen lower left with the cable looped out of the way Last edited by jefrs; June 9th, 2008 at 08:31 PM.

May 24th, 2008, 09:17 PM	#18 (permalink)
jefrs Tele-Holic Join Date: Nov 2007 Location: Newbury, England Age: 54 Posts: 809	pchilson, you board pic is green, it's ver2 also? I'm no expert either, it's a hobby for me, but has been since I was 8 or so, and it is related to my work. However I am fortunate to know several chaps who are genuine experts, and whom I pester mercilessly for the public good whenever the opportunity arises. I'm still not too old to learn. I don't know what 'IPD' is, we appear to be divided by a common langauge. I suggest the the heater voltage is unsmoothed dc, your 7.4V is peak, switch you meter to ac and re-measure, I guess average will be around 6.3V and ok. The heaters are taken from a separate secondary winding and circuit, thus independent of R14. Tricky thing bias. The grid must be negative in respect to cathode. If we change this voltage with R14 then we change the gain curve. On spec sheets we often see a nest of graph curves for different bias on gain vs anode voltage plot. We chose one that will give us a linear gain (or being a guitar amp, chose one that doesn't). Your gain changed because you changed the bias. If your mains is 123VAC and your transformer tap is 115VAC then you will get more volts out of the secondary, HT supply (278VAC?). The ultimate solution is a regulated and stabilized power supply (plain daft), or just a bigger resistor at R10. Now the RC network is too complicated for me to work out (I know it can be modelled in a spreadsheet) the value of R10 to drop B1 below 340VDC. Try next standard value, 240ohm or 270ohm [E12(E24) series180 (200) 220 (240) 270 (300) 330 (360) 390]. But this will drop all B1 B2 B3 HT voltages. The trouble with R10 is that changing it will be interactive as you found when R14 altered all the voltages. Get B1 right then experiment with R12 and R13 to fix B2 and B3. Our mains voltage varies a great deal, especially at early evening peak when the voltage drops as all the cookers and tellies in the country go on. They used to regulate to 245VAC and let the frequency dip but that threw all the clocks out, so now they maintain 50Hz and can drop to 220VAC. That's rare though, most of the time it's above 240VAC Was your R1 not 680k? R1/R2 is a voltage divider. Putting R1 in front of R2 decreases the input impedance and decreases signal to V1 grid. Does it make much difference to the sound? What about those mini-Marshall capacitors? What effect did they have? Today I refurbished the radio-case cabinet and fitted a brass ventillation grille in the door where the tuning dial used to be, so far it has not got hot enough to warrant one at the back, so it's a closed-back combo. The Elac sounds real pretty in there, so I'm disinclined to do too many mods to the circuit, but I would still like some variation in tone, or maybe gain. Attachment 11765 Attachment 11766 The colour's a bit wrong, the walls are a creamy yellow, not pink... Last edited by jefrs; June 9th, 2008 at 08:31 PM.

May 20th, 2008, 11:20 AM	#5 (permalink)
Pvee Tele-Meister Join Date: Mar 2003 Location: Ohio Posts: 269	I made several mods to mine. Added some extra filtering for the B+ voltages and lowered B1 about 20 volts. It's a winter time project for me. I think there is a signal level issue in mine, maybe all of them. This was before the mods also. I can over drive or cause it to distort with it around 1/2 volume and guitar output high.. With the volume control all the way up, it seemes to be a cleaner tone and that's what I am after. Not really a big probem or anything but I want to get it so that t is pure and clean at all volume pot settings. I can get a good big fat sound out of it with the 10" speaker cabinet I built. Over all I am satisfied with it. It's just another project to play with.

May 20th, 2008, 04:23 PM	#6 (permalink)
pchilson Tele-Afflicted Join Date: Aug 2007 Location: Loveland, Colorado Age: 51 Posts: 1,035	Hey Pvee, what did you do to filter B+ and what was your B1 before and after the "lowered 20 volts"? What version board is yours? After re-biasing mine with the 330ohm R14 resistor the plate dissapation lowered to 12.17, the voltages went up and the gain increased, almost unuseable at upper volumes. I'm told this should have gotten cleaner and rounder sounding instead. I don't know but I think I need to do what you did with B+ and B1. I mean, doesn't 330 ohm in R14 seem like an awful high value and still be over 12 watts? I think the voltages are to high and if I can lower them then maybe 270 ohm or so in R14 might be more in line...What do you think?

May 20th, 2008, 09:21 PM	#8 (permalink)
jefrs Tele-Holic Join Date: Nov 2007 Location: Newbury, England Age: 54 Posts: 809	Is the rectangular object next to the orange wires a bridge rectifier? If so then VER051122 EVJH board has dc heaters, doh. But why have I got a 7.5k o/p transformer? What are your R1, R8 and R9? Is this version 2 or 3 or something in between? Now if we have actually got 15 watts dissipated then that current is passing through R14 220ohm 2W resistor - it should be awfully hot by now. What are your measurements? I have not delved into my board with a meter yet. Mullard EL84 data tells us Va=250V (anode to cathode) and Ia=48mA (12W) and Rk=135ohm with Ra=4.5kohm. What voltage do you have at anode and cathode wrt earth?

May 20th, 2008, 10:30 PM	#9 (permalink)
Scott S Tele-Afflicted Join Date: Dec 2005 Location: SW CR IA US NA PE Age: 28 Posts: 1,922	Um, if I'm not mistaken, this is a single-ended class A amplifier, and the output tube SHOULD be conducting at all times. That's why the idle dissipation should be 12W or less in this amp if tube life is a concern. - Scott

May 21st, 2008, 10:48 PM	#10 (permalink)
jefrs Tele-Holic Join Date: Nov 2007 Location: Newbury, England Age: 54 Posts: 809	Correct, it's single-ended so it has to be Class A = the anode current is set at a level such that even with the maximum allowable input signal, the anode current never falls to zero, i.e. the valve never switches off. But if it were pulling 12 to 15 watts then it should be running much hotter, literally as heat, than it is. The Laney L5T, single EL84 uses a 150ohm cathode bias, but I dont know what the supply rail is. Which is why I was interested in the actual voltage-to-earth measurements in the chain:- supply rail; anode; cathode. From these we can calculate the power dissipation in the o/p transformer, the EL84 and the R14 bias resistor.

May 22nd, 2008, 08:57 AM	#11 (permalink)
pchilson Tele-Afflicted Join Date: Aug 2007 Location: Loveland, Colorado Age: 51 Posts: 1,035	jefrs, Sorry, haven't even looked at the thread to see your question. I'm too much of a newb at this to know what measurements you are looking for. Tell me what to measure and I'll be happy to measure them. I'm going in tonite to change R14 back to the 220 ohm from the 330 ohm currently in there. I've been told it doesn't make any sense but with R14 at 330 the gain has increased so much as to make the amp almost unuseable. I need to put it back stock and perhaps look at lowering some of the B voltages first before changing R14 again. I'm just trying to learn this stuff so you need to be specific if you need values from me...

May 22nd, 2008, 04:03 PM	#12 (permalink)
jefrs Tele-Holic Join Date: Nov 2007 Location: Newbury, England Age: 54 Posts: 809	PChilson, Right, R14 is used to bias the cathode with respect to earth so the grid is negative in respect of it. This is cathode bias. Some amps supply the grid with a negative voltage e.g. PV Classic, this is grid bias. This bias defines the gain. Changing the bias resistor to limit current does not make sense. If it is running hot, because it is dissipating too much power, then anode has too many volts on it - so lower its supply rail voltage. Lower the voltage too much and we get starved plate (anode), in a hi-fi amp this is anethema but may be desireable for guitar. I may measure some volts on my un-modded amp tonight. I am disinclined to mod the bias because it sounds just right.

May 23rd, 2008, 08:27 PM	#15 (permalink)
pchilson Tele-Afflicted Join Date: Aug 2007 Location: Loveland, Colorado Age: 51 Posts: 1,035	I'm certainly no expert on this but that green pcb is indicative of a ver 1 and ver 2 boards. The ver 3 boards are dark brown. So I would say you got NOS amp, that would explain the 7.5k OT. When you look at the schematics the ver 2 is much closer to the ver 3 than the ver 1. I think my next move is to bring B1 down to 300VDC and then work from there. Just have to go to the parts store and get a few to play with. I'm not at all sure this is something that is needed if the plate dissapation is under 12 watt but I see lots of reference to it so I will do it. I also have a high reading on the heater measurements 6.3v by spec and mine is running 7.4. I do know one thing, when I bumped R14 to the 330 ohm resistor all my voltage readings went up, so I don't know at this point if the heater voltage went up then as well. I may need to drop the stock 220 ohm resistor back in R14 just so I can measure all my baseline stock values. Remember, I'm learning as I go here...

May 24th, 2008, 09:36 PM	#19 (permalink)
jefrs Tele-Holic Join Date: Nov 2007 Location: Newbury, England Age: 54 Posts: 809	I misread your question re heaters, yes, increased mains will increase volts from heater windings too. But the schematic shows 120VAC primary, your is labelled 115V? Are there any other mains taps on the transformer label? A small resistor in series with the heaters will drop the volts but would need to be at least 6Watt. Value depends on resistance of heaters, which gets less as they heat up. They are probably ok as they are, at least the valves are cheap. I would only fix it if it blows heaters. It probably has some minor effect because heaters do emit a few electrons and running hotter will effect cathode emission. You may observe the tone changes as amp warms up and after a few hours on.