Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: uki on February 19, 2017, 11:24:48 pm
-
Hey guys !
Gonna share some experience here for new people, not like I'm that smart about amps but we all learn with our mistakes.
When I build my 1st amp I had to find out the specification for the PT, in the schematic for the amp there is a number, 67233, so I went searching the internet for that mysterious PT, and found a schematic somewhere with that number, and I thought that I had found the right one, but only for the number printed on it, the PT wasn't enough for the amp, it were well custom made and although it worked for a year and in a VERY VERY HOT way before it finally shorted :BangHead: , gladly it didn't damaged the amp circuit, and only had the heater wires working in low voltage. After some thinking/reading about it, I realized that the PT only should feed about 5-6 tubes tops, 2x 6L6 and 3-4 12ax7s or something of the sort of amp,. NOT 10 tubes, 4x 6L6s and 6x 12ax7s.....
So now I got a "new" PT the old one were remade stronger, but it is slightly different and now have separated bias wires. The guy that made the PT was very kind suggesting separated bias wiring so the PT would work much cooler.
After some research I realized that I couldn't just plug and play it... and thanks to Sluckey and his scrapbook I found out the right way to use the separated bias taps, but I have no clue whatsoever what are the values for the components in the in that circuit,
could someone please help me with it? :help: The amp is the Fender 6G8, here is the schematic fot it, the diagram for new PT and the bias circuit for separated bias wires.
Thanks in advance !
-
You don't need that circuit with the full wave bridge.
Just connect one of the bias winding wires to chassis at the same point your red/yellow center tap is connected. Then connect the remaining bias winding wire to the bias diode (like the red/blue wire on the layout you posted).
-
> one of the bias winding wires to chassis
> {other} bias winding wire to the bias diode
-
Hey guys thanks!
Oh.... I thought that couldn't be done, that simple ?
How to find out which wire is zero and 50? I measure both wires with the PT chassis and got nothing, only when measuring across both wires it read 50v. Does it matter ?
-
How to find out which wire is zero and 50?
This can be tricky, especially if both wires are the same color. Close your eyes and choose one. Call it zero and connect it to chassis. The other one will be called 50 and will connect to the diode. I hope you get it right. :wink:
-
Thanks Sluckey !
-
Just for the sake of understanding, this is one big question, could someone explain please.
Wouldn't I be putting 50VAC potential in the chassis GND?
-
Wouldn't I be putting 50VAC potential in the chassis GND?
No. Remember, you will call one wire zero and connect that wire to chassis.
-
Wouldn't I be putting 50VAC potential in the chassis GND?
No. Remember, you will call one wire zero and connect that wire to chassis.
Exactly, if you put BOTH wires to chassis, then you're right, the circuit closes out with a ground in AC it's like the 'neutral' vs 'hot' idea here, but the circuit just needs some way to 'return' the current back to where it started, closing the loop, the chassis and ground becomes that for pretty much everything in the amp.
Then the 'earth' comes from the three pronged cord's green wire connected to the chassis as well for safety in case somehow both wires of any voltage end up directly shorted to ground, that shoots to earth instead of shocking the crap outta the user.
~Phil
-
Alright many thanks guys!!
In what situation I would need that full bridge rectifier/bias circuit?
I'm also taking it as an opportunity and I remade the rectifier board and added a fuse between the diodes and 1st filter cap. 250mA fuse will do ?
-
The FWB will give about 70vdc voltage and it will have a 120Hz ripple (assuming 60Hz line frequency). Easier to filter. That's enough voltage for 6550s.
But you don't need that. The bias voltage is a critical circuit. A lot of expensive components depend on a reliable bias circuit. Keep it simple.
-
PT is connected as you instructed Sluckey/PRR Thanks !!
About this FWB bias circuit, supposition if it were going to be used in my amp, It would then need only 30 volts coming out of the PT? Is that correct ? Or there are ways to adjust the voltage with 20 volts higher ?
-
You would juggle the resistor values to give the voltage you want.
-
I got the amp wires all soldered back, my readings are, without tubes, power on, and standby off,
B+ 438-440 vcd
bias -54 vdc
All pin 5 of power tubes are reading -44vdc is that the expected voltage ?
I got a mod on power tubes, got resistors on each tube of 1500R . look the pic.
Anything else to look at ?
-
All good to go ?!
-
All pin 5 of power tubes are reading -44vdc is that the expected voltage ?
That may bias your output tube on the hot side. The original schematic shows -55v.
Try it and see what you think. Watch for possible redplating.
-
All pin 5 of power tubes are reading -44vdc is that the expected voltage ?
That may bias your output tube on the hot side. The original schematic shows -55v.
Try it and see what you think. Watch for possible redplating.
I did turn on the amp and played for about 5-10 mins tops, watching the power tubes, I didn't see any redplating, but the power tubes are working way hotter than before, I would say something around 20-30% hotter, amp is breaking up sooner at volume 6-7, before it was only breaking up at 8 and half, the power seen to be little smaller as well. What to do?
Oh I only have V1 and V6 plugged for the preamp tubes, all power tubes plugged.
-
Decrease the value of that 10K resistor on the rectifier/bias board until you have about -50 to -55v on pin 5 of every output tube. I would not go any lower than about 2K.
-
How about add to the bias circuit a pot to adjust it, would that be a good move ?
-
Sure. But first you should get it working right.
-
Sure. But first you should get it working right.
What type is the biasing in this amp? Grid ?
Reducing the bias resistor will decrease the voltage to compensate the drop after the resistors before the tubes? How many volts I will need, -65, to get -55 at pin 5(getting -10 drop) ? Something like that yes? I'm guessing between 7-9k resistor will do.
-
JUST DO IT
-
JUST DO IT
Like a dealer pushing his stuff on you :)
DO EET :)
yeah Sluckey's right, just try making sure the 'expected' behavior is working, then try modding it to a new possibly 'not working' method, so you can ensure 'I know the bias worked before, what did I break with this change' instead of 'I've changed two things, sheesh which is the 'wrong' one'
-
All pin 5 of power tubes are reading -44vdc is that the expected voltage ?
I am curious about how you got -44Vdc from a 50Vac winding that is rated at 100mA. The turns ratio is correct and the components/wiring looks correct in the pictures. The first thing that comes to mind is that the winding is 40Vac instead of 50Vac. If that is the case, you may only get -53Vdc half-wave with a single RC filter.
Another thought that comes to mind is that the winding is 50Vac, but the first filter after the diode is not functioning properly. It takes a couple of uF's at that position to get the full potential of this circuit type. Lessor value capacitors will result in lower voltages (absolute values) at the bias point.
It could be helpful if you could post the measured ac voltage of that bias winding.
-
The bias supply goes through a voltage divider resistor to get it to the 'right range' for the tubes. Some also include a variable potentiometer to allow for tweaking it. You could do the math pretty quickly to figure out the resistors used, but I think they were listed anyway.
-
JUST DO IT
Yes Sir !! :icon_biggrin:
I did read the bias winding suggested above by 2deaf and it is reading 47vac ...
Ok I did some tests, and here are the results
with 8.2k resistor the result is -57vdc at the bias resistor and at pin5 -46.6vdc
then 5.6k -60vdc -49.2vdc
then 4.15k -60.4vdc -49.3vdc
I don't see it getting any better, the amp now sounds awful, there is a bad overdrive at volume 4-5...
Too bad I don't have those reading before the problem with the PT to compare.
:help:
-
I don't see it getting any better, the amp now sounds awful, there is a bad overdrive at volume 4-5...
Well, put it back like it was and live with it.
-
with 8.2k resistor the result is -57vac at the bias resistor and at pin5 -46.6vac
Is the voltage at pin 5 the same as the voltage at the junction of the 8.2K and 56K resistors?
-
The tubes are going way too hot.
What really change in the way the PT is now in comparison to the previous one? Could be the amperes in the new winding ?
Why the new PT change so much the way the amp work ?
with 8.2k resistor the result is -57vac at the bias resistor and at pin5 -46.6vac
Is the voltage at pin 5 the same as the voltage at the junction of the 8.2K and 56K resistors?
When I read from the that junction(orange dot in the picture) to ground I get a different reading than pin5(green dot in the picture) to ground.
-
Should be the same. Time to find out why they are not.
Pull the power tubes and see if the voltages are the same.
-
Without the tubes the readings still different, at those same points.
-
Without the tubes the readings still different, at those same points.
So that means that those coupling caps are leaking positive voltage from the PI plates ***OR*** something is not wired correctly. You did a mod to add grid stopper resistors. Check that for accuracy.
-
Without the tubes the readings still different, at those same points.
So that means that those coupling caps are leaking positive voltage from the PI plates ***OR*** something is not wired correctly. You did a mod to add grid stopper resistors. Check that for accuracy.
The grid stoppers check good and connections are correct, those have been in the amp since the begin.
There are two things that changed is the PT, and the rectifier board, I've remake the board to better fit inside the chassis.
How do I check those coupling caps ?!
-
How do I check those coupling caps ?!
Just disconnect one end of BOTH caps. Now is the voltage on pin 5 of each tube the same as at the orange dot?
-
What about some wheels? ....................I'm sorry! But I had to!. Hope it at least made you smile.
-
What about some wheels? ....................I'm sorry! But I had to!. Hope it at least made you smile.
Heh !! :smiley:
How do I check those coupling caps ?!
Just disconnect one end of BOTH caps. Now is the voltage on pin 5 of each tube the same as at the orange dot?
The readings still the same no changes, the bias resistor in place is the last one tested with same results as before 4.15k -60.4vac -49.3vac. It isn't the caps then. What else to check ?
-
Turn the power off and measure resistance from pin 5 of each power tube to chassis. Should be about 270KΩ. What do you measure for each tube?
-
V7 294k
V8 294k
V9 283k
V10 283k
A bit too high ?
-
Resistances are fine.
-60.4vac -49.3vac
You keep saying your bias voltage is vac. Bias voltage is negative DC. Be sure your meter is set to measure DC voltage, not AC voltage.
-
Sorry about that , that is correct it is indeed DC , that was a mistypo...
I put back the 10k resistor and gonna check how it sound again and try to get some feeling about how hot the tubes go... I had only V1 and V6 preamp tubes on when tried 1st time, Having all 10 tubes on makes difference?
ok what next ?
-
I don't think you should put the power tubes back in until the bias is functioning properly.
-
When you put the 10K back in the tubes will run hot. I would put a 470Ω in and measure the voltage on pin 5 with no tubes plugged in. Once you get that voltage up to -55vdc or higher, we can tackle making it adjustable.
But I'm still concerned about the difference in voltage between the junction of those resistors and pin 5 of the tube. This difference may be caused by your meter. What is the name and model of your meter?
-
I don't think you should put the power tubes back in until the bias is functioning properly.
Understood !
When you put the 10K back in the tubes will run hot. I would put a 470Ω in and measure the voltage on pin 5 with no tubes plugged in. Once you get that voltage up to -55vdc or higher, we can tackle making it adjustable.
But I'm still concerned about the difference in voltage between the junction of those resistors and pin 5 of the tube. This difference may be caused by your meter. What is the name and model of your meter?
Power tubes been off of the amp all this time.
My meter is china cheap stuff model is DT830B brand TBlack here is a pic:
-
That meter only has a 1MΩ input impedance. That explains the difference in voltage readings. So, forget about that and replace the 10K with a 1K. Then install a 1Ω resistor between pin 8 and chassis. Do this for all 4 tubes.
Now read this about biasing an amp.
http://www.duncanamps.com/technical/lvbias.html (http://www.duncanamps.com/technical/lvbias.html)
-
LOL I emailed Uki that exact link last friday.
-
Hey, where you been hiding Ed?
-
When you put the 10K back in the tubes will run hot. I would put a 470Ω in and measure the voltage on pin 5 with no tubes plugged in.
Do you mean in place of the 10k ?
Now read this about biasing an amp.
http://www.duncanamps.com/technical/lvbias.html (http://www.duncanamps.com/technical/lvbias.html)
I been reading that, did most of it, still need to complete the reading and read all over again to keep it in mind. I got most of the idea, and I see you're guiding me that way, the doc does help !!
That meter only has a 1MΩ input impedance. That explains the difference in voltage readings. So, forget about that and replace the 10K with a 1K. Then install a 1Ω resistor between pin 8 and chassis. Do this for all 4 tubes.
I need to get some 1Ω resistors to do that I don't think I have enough, I know I have some 1Ω-5W is that too much ?
About the meter could you please educate me, 1MΩ input impedance, what that means?
What kind of meter is recommended?
Thanks sir !!!
-
When checking a dc voltage with a meter with a 1M ohm input impedance, it is the same as inserting a 1M resistor into the circuit at the points contacted by the probes. The attached diagram shows what happens when your meter is used for voltage.
The meters that I use have 10M input impedances. Even with that, they sometimes alter high impedance circuits enough to throw me.
If you measure the dc voltage between point A and point B in the top diagram with your meter (or any meter) you should get 0.00V. This would have been the better test to suggest.
-
When checking a dc voltage with a meter with a 1M ohm input impedance, it is the same as inserting a 1M resistor into the circuit at the points contacted by the probes. The attached diagram shows what happens when your meter is used for voltage.
The meters that I use have 10M input impedances. Even with that, they sometimes alter high impedance circuits enough to throw me.
If you measure the dc voltage between point A and point B in the top diagram with your meter (or any meter) you should get 0.00V. This would have been the better test to suggest.
Oh I see, the 1M in the meter is doing that 10v drop ! The picture show it very clear! Thanks!!
Does the 1R(found some) resistors have to stay, or is it just for this reading ?
-
Does the 1R(found some) resistors have to stay, or is it just for this reading ?
I would install them in the amp permanently. They are too small to affect the performance of the amplifier and the occasion for their use is going to come around again.
-
What is the minimal and maximum watts for this resistors ? I got some that seen to be or 1/2 or 1/4 watts. is that good enough? They seen to be metal film 5% , grey body.
-
I know you will want 1% tolerance not sure minimum size though
-
Those will work. High wattage is not needed. 1% tolerance would be more accurate, especially with your $6 meter. Hoffman sells them for $0.15.
-
Uki once you install them there is no reason to remove them. You may need them to check your bias later if you change to a different tube or even a different brand tube.
-
Uki once you install them there is no reason to remove them. You may need them to check your bias later if you change to a different tube or even a different brand tube.
Got it , thanks Ed !!
I got the resistors in place, and power tubes in place, I tried to read across the resistors with standby off, but I got nothing, does standby need to be on ? What next ?
-
Read the link
-
Read the link
Ok I got it !
10k bias resistor in place.
B+ is reading 442vdc
Pin3 on all power tubes are reading 433-434vdc
Across the 1ohm resistors I got the following, meter at 200mV dc
V7 6.4
V8 6.3
V9 6.8 (does this difference bother?)
V10 6.3
All tubes were very hot after about 10 minutes.
433 * 6.3 = 2727.9 so that is about 27watts 2.7watts
433 * 6.4 = 2771.2
433 * 6.8 = 2944.4 wow super hot !! ouch on the limit !!! :blob8:
also I take notes of the grid stopper resistors resistances, if that matter
V7 1607
V8 1632
V9 1636
V10 1569
What else should I look at ? What next ?
-
433 * 6.8 = 2944.4 wow super hot !! ouch on the limit !!!
No. That should be 433 * .0068 = 2.944 watts. That's super cold. You need to increase the value of that 10K.
-
433 * 6.8 = 2944.4 wow super hot !! ouch on the limit !!!
No. That should be 433 * .0068 = 2.944 watts. That's super cold. You need to increase the value of that 10K.
But if that is cold why the tubes are so hot ?
-
6L6s get hot just because of the filaments! You don't see any plates glowing red do you?
If I can trust your measurments then your tubes are running super cold. Keep lowering the value of that 10K resistor until you measure 40 to 50mV across each 1Ω resistor.
-
6L6s get hot just because of the filaments! You don't see any plates glowing red do you?
Hotter than before at least that is what I've noticed, is it because the PT have now more mA ? No red spots on the plates, it would be easy to see yes ?
You need to increase the value of that 10K.
Keep lowering the value of that 10K resistor until you measure 40 to 50mV across each 1Ω resistor.
The meter show 6.3mV where did the other 40 something go?
You got me lost now !!! :laugh: Up or down ?!
-
I was mistaken to say lower. Please forgive me.
The meter show 6.3mV where did the other 40 something go?
When you INCREASE the value of that 10K you will flush it out of hiding.
-
All good Sluckey !! I appreciate very much your help and patience!!!
Been playing for about 20-25 mins watching the power tubes closely, no red spots in the plate.
All tubes are plugged, all hotter than before.
Power tubes are going way hotter than before, due to the new amperage on the PT yes?
The tone is definitely different than the old PT, slightly, bit dry and breaking up sooner than before.
Overall tone is about the same.
How much can the bias resistor go up, what is the safe limit ?
-
> That meter only has a 1MΩ input impedance
Or maybe 2MΩ.
Apparently "DT830B" covers several meters from several factories all sold as the same thing. Different colors and even different jack layouts (so not the same molds). A thorough reviewer on Amazon says his is 2Meg while others are 1meg.
-
When you say "hotter than before" how are you measuring that? The power tubes will be very hot... too hot to touch or hold. This is normal. If biased to about 70% (recommended) or whatever you prefer... and there is no red plating then I wouldn't worry.
When you read 6mV across the 1ohm resistor was the bias pot full CW or CCW?
You should have a decent range from one end to the other with that pot. Maybe 6mV until 30mV+
As you increase the current (measured over the 1ohm resistor) the plate voltage will lower which will change the 70% calculation. So adjust, re-measure, and re calculate as you go.
EDIT: Realize you have just a fixed bias resistor... follow Sluckey's advice and increase it until 40 - 50mV
-
Or maybe 2MΩ.
The results posted in reply #13 and reply #25 firmly support 1M. Not that it matters.
-
When you say "hotter than before" how are you measuring that? The power tubes will be very hot... too hot to touch or hold. This is normal.
I mean temperature, measuring just by the touch, not the "hot" term for bias adjust. Ok maybe I was not so sure about temperature before, because the previous PT was overheating due to low specification. The tubes gets indeed too hot to touch or hold.
How much can the bias resistor go up, what is the safe limit ? 18k ?
-
Take it up until you are satisfied with the static plate power dissipation. I like 70% of max.
Read the link!
-
With the bias resistor, as you increase its resistance you are supplying less and less -Vdc to the tube. Less -Vdc = more current. The max safe amount for that resistor is something that gives you the correct bias. Unsafe would mean you have too much current for the 6L6 and it will red plate and/or not last as long as you might want it.
For real though the link that Sluckey posted on page 1 will absolutely explain everything.
-
With the bias resistor, as you increase its resistance you are supplying less and less -Vdc to the tube. Less -Vdc = more current. The max safe amount for that resistor is something that gives you the correct bias. Unsafe would mean you have too much current for the 6L6 and it will red plate and/or not last as long as you might want it.
For real though the link that Sluckey posted on page 1 will absolutely explain everything.
Oh that is good info !!! Thanks !! I got to check out the tube specs also !!
Take it up until you are satisfied with the static plate power dissipation. I like 70% of max.
Read the link!
I been reading sir !! I'am :BangHead: , I'm not sure what is 70% of max , that I get by measuring the 1ohm resistors ?!
-
If I were you I would either do:
1) Get a trim pot to adjust and find the sweet spot for your liking, measure the resistance over the pot in that position, then hard wire the applicable resistor. (Or keep the pot if you feel you may want to tweak it or change tubes in the future)
2) Use alligator clips to quickly swap resistors until you get the correct readings. This will save you time because going from 6mV to what you need (40 - 50) will change your plate voltage quite a bit. I just biased my amp yesterday after installing a new PT. Original plate voltage was 445ish with bias set at 6mV. I adjusted my trim pot to get 70% dissapation which lowered the plate voltage to around 410. I then had to re-calculate 70% and make another small adjustment.
-
I'm not sure what is 70% of max , that I get by measuring the 1ohm resistors ?!
Maximum static plate power dissipation for a 6L6 is 30 watts. 70% of 30 watts is 21 watts.
Now... READ THE LINK! :violent1:
-
How many watts is it doing now according to my readings ? 3 watts per tube? That can't be right, is it ?
-
433 * 6.8 = 2944.4 wow super hot !! ouch on the limit !!!
No. That should be 433 * .0068 = 2.944 watts. That's super cold. You need to increase the value of that 10K.
Sluckey did the math for you above. With your posted readings you are indeed around 3 watts per tube. Like he said... that is suuuper cold and is exactly why the amp sounds very dry and lifeless.
-
I GIVE UP
-
> How many watts is it doing
Idle dissipation has nothing to do with maximum output (especially on a fixed-bias amp).
There may be something you can read.
-
I'm starting to get this !!
I GIVE UP
Sluckey!! https://www.youtube.com/watch?v=B_rL8vLkxjM (https://www.youtube.com/watch?v=B_rL8vLkxjM)
Ok this, with 16k resistor the voltage still about the same -49vdc. but the mV on 1ohm resistor doubled 12.3
then I put in a 21k resistor voltage dropped a little more -46.3 and the mV went up good to almost 20, 18.4 in one side of the power tubes and 19.4 in the other side. Then I tried it...
NOW THE AMP SOUNDS GOOD !!!! No break up , good loud clean tone until 9 !!!
:m12
> How many watts is it doing
Idle dissipation has nothing to do with maximum output (especially on a fixed-bias amp).
There may be something you can read.
This only change the tone ?!
-
Do you know where to find the maximum plate watt dissipation for whatever tube you are biasing?
From that you can calculate 70% or whatever you want.
-
> How many watts is it doing
Idle dissipation has nothing to do with maximum output (especially on a fixed-bias amp).
There may be something you can read.
That only change the tone ?!
Do you know where to find the maximum plate watt dissipation for whatever tube you are biasing?
From that you can calculate 70% or whatever you want.
I was about to ask that, how to calculate the bias resistor value to get that answer? Is there a formula ?
-
I don't think their is a formula that will tell you the exact size resistor you need to be at 70% plate watt dissipation. The schematic gives you an approximate starting size.
If you check the tube data sheet you will find the maximum plate watt dissipation.
With the voltage measurement across the resistor you now know your current.
With Ohms law use this formula to find power: P= current X voltage
So you want to keep adjusting that resistor size trying to get your plate watt dissipation to 70% or under what the data sheet shows for a 6L6 push pull amp.
Hope I got this correct it's been about 8 years since I bought the tube bias checker and it's been that long since I used the 1 ohm resistors. LOL
So earlier you said you measured 6.3mV?
You have 443v for plate voltage?
So 443*.063= 27.9 watts
I think the max for 6L6 is 30 watts so you should be around 21 if you want 70 %
Someone chime in if this is incorrect, I'm a bit rusty on this.
Oops my mistake it would be 443*.0063=2.79 watts
-
Do you know where to find the maximum plate watt dissipation for whatever tube you are biasing?
From that you can calculate 70% or whatever you want.
I don't know where to find it. Everyone says 70% why exactly have to be that ?
If you check the tube data sheet you will find the maximum plate watt dissipation.
I think this is the maximum negative voltage the tube can take.
Ug1 = -14 V(se) -17,5 V(pp)
> How many watts is it doing
Idle dissipation has nothing to do with maximum output (especially on a fixed-bias amp).
There may be something you can read.
That only sets up the tone ?!
I don't know what some things in the chart means those letters in the left side, like La, Lg2, N, and some others.
-
That doesn't give you much information. Try looking for the 6L6 RCA datasheet. The one I see has 23 watts (I think) for max plate watt dissipation for class AB push pull. Your correct though idle and full out will be different that's why you don't want to bias over 70%
I emailed you the tube datasheet
-
Please guys bear with me, time is in short supply here and I go as I can, I did big progress believe me, I'm kinda slow due to the lack of time, got my hands full.
Here:
How about add to the bias circuit a pot to adjust it, would that be a good move ?
...you can tweak the supply to get the mA's that you want at around the midpoint of the pot.
Sure. But first you should get it working right.
I see the need of the value for the bias resistor must be found 1st, then it is possible to tweak the adjustable bias circuit, did I get it right ? The mV should be between 40-50mV is yes?
Refer to Hoffman's schematic. Change R33 from 15K to 10K. Then experiment with the value of R35 (up and down). Can you now get a wider range? Maybe something close to 40 to 60 volts?
Would this work here? Can I follow that schematic?
-
Uki, something just popped into my mind about your amp voltage measurements.
You said your measurement across the 1 ohm resistor was 6.3mV? I was thinking about that voltage measurement.
The filament voltage is 6.3v. Are you sure your 1 ohm resistor is on the correct pin, and also that your taking the voltage reading across that resistor, and not getting your filament voltage some how?
i guess if the mV measurement is changing as you change your bias resistor, then it was just a coincidence.
-
Uki, something just popped into my mind about your amp voltage measurements.
You said your measurement across the 1 ohm resistor was 6.3mV? I was thinking about that voltage measurement.
The filament voltage is 6.3v. Are you sure your 1 ohm resistor is on the correct pin, and also that your taking the voltage reading across that resistor, and not getting your filament voltage some how?
i guess if the mV measurement is changing as you change your bias resistor, then it was just a coincidence.
That 6.3mV was with the 10k resistor, when I put 16k and then 21k the mV went up ! I got new resistors today, with thicker legs, the ones in place are too thin, 1/2 watts and 1% for more accurate readings !!
-
The mV should be between 40-50mV is yes?
Yes.
Would this work here? Can I follow that schematic?
What for? You are getting close with the one you are using. Keep going.
-
Here is a bias calculator. You enter your tube type and the plate to cathode voltage and it calculates approximate.
I think your plate to cathode voltages will change as you change the bias resistor so you will need to check that plate to cathode voltage as you change resistors and recalculate or change the voltage in this bias calculator to get the new mv setting for 70%
https://robrobinette.com/Tube_Bias_Calculator.htm
-
Keep going.
I'am !! as time allow me I keep moving !! In place now a 28k bias resistor, bias voltage at -41, B+ at 400ish, and the mV in the 1ohms resistor is reading 30mV ish. Got to find some 39k or 42k resistors, I think it will be right in the spot !!!
And this is very interesting the tone improved a lot more , it isn't even "there" yet !! Tone is more round and smooth now!!
About B+ is it the expected behavior to it drop as the bias voltage changes the way it is going ?
-
> About B+ is it the expected behavior to it drop as the bias voltage changes the way it is going ?
It is cold here tonight. I turn on all my electric heaters. The lights dim. My 125V drops to 115V.
Say I have a pack horse. I put on the saddle, OK. Then I put a 100W amp in one side and a drum-kit in the other side. The horse sags.
All power sources sag when loaded hard.
You seem to be working-up from 6mV in 1 Ohm (6mA) to now 30mA. Or since you should be measuring one of four similar tubes, 24mA to 120mA. That's 5 times more current than you had before. The B+ voltage will sag some. When played HARD, a 6G8 may pull over 400mA. But as long as the sag now is not huge, it will be fine. (We hope.)
-
When you say the tone is "right there" it is very subjective. What are you after?
-
Many Thanks PRR !!
When you say the tone is "right there" it is very subjective. What are you after?
I was talking about the mV something between 40-50mV , sorry I was unclear.
Well I found a 38k resistor, the results are: bias voltage -36 , B+ 400 ish(405) , mV -46.6 :huh:
But the tubes are super hot now :blob8: even the chassis where the tubes are is getting very hot, and I've noticed as the values been moving up the amp gets a little noisy, some small hum, due to the current increasing maybe?
I did like a lot the result when the 28k bias resistor was in place.
Played for a little over 5 mins with this setting was late and didn't had the chance to really feel the tone, but still it is way too hot. On the good side the PT barely warm up !!
Seen like the range is for the bias resistor is 10k to 39k, Now with all those numbers, what to do next ?!
-
Put the 28k back in (or even better a 5k with a 25k pot in series)
Adjust it until you get around 28-39 mV.
Then do the math or use the calculator I emailed to see what your plates are dissipating.
Then adjust the pot until you like the way it sounds but if you don't want to have to replace the tubes often
You should try to stay under 70%
The cooler they are biased the longer they should last but also if you bias too cold the amp does not sound like we want a guitar amp to sound.
-
Are the bias voltages posted in reply #76 and #90 taken at the 56K resistor?
-
Are the bias voltages posted in reply #77 and #91 taken at the 56K resistor?
The one in the junction(orange dot) with cap 8µF, 56k resistor and the other one that I've been swapping, initially 10k, then 16k, 21k, 28k, 38k.
Got a pic.
-
That's what I figured. Attached are two adjustable bias ideas now that I am reasonably certain that I know what you mean.
-
What is the difference between the two ? Are those similar to the previous examples you posted earlier in the thread, more ripple less ripple ?
-
The one with the 25K pot. does have less ripple, but that was not the main criteria for the designs. I designed them using mostly resistors that you have on hand. 10K is a really common pot. for bias supplies, so I designed a circuit with that and your resistors. If you happen to have a 25K pot., I included a drawing for that using your 21K resistor.
I like the numbers you posted with the bias voltage at -36 and you like the sound at -41V. Both of the variable bias supplies should easily have both values within their ranges.
-
No 25k pots around here, usually 1, 2, 5, 10, 20, 50, 100 and some bigger ones that is usually what the stores have. I do also recycle old electronic equipment like tube screen TVs, or anything that is obsolete but have usable parts, and I use the part to make pedals (http://el34world.com/Forum/index.php?topic=21606.0) !!
I got 3 trimmers, 10k, 50k and 100k, the small ones, and some 10k pots too, I can do the 10k one , I have all resistors and the pot !
-
Been reading the amp voltages while playing on it,
I put back the 21k resistor, readings are:
across 1ohm resistor, peak is about 70mV , it did happen while playing really hard a 6 strings chord, that was about two times in about 5 minutes tops, average mV was around 40-50mV, the amp chassis goes really hot at the power tubes area, and after an hour playing the PT also get hot.
idle plate voltage 413 , and when hitting hard a 6 strings chord the voltage dropped to about 390.
So that is 27 WATTS
390 X .070mA
That seems like possibly that if those measurements are correct the tubes should be redplating at over 23 watts.
Seems like their is too much current swing from just sitting idle to when you play on it pretty hard.
Maybe post again that your bias setting is below the 70% plate dissipation but when you push the amp hard with your guitar it seems to be going well over 25 watts dissipation.
Is that the expected behavior ?
-
Did you say the idle voltage measurement across the 1 ohm was around 30mV?
I think at idle you are around 14 or a bit more watts. Thats around 30%
It seems like a pretty big swing from 30mV to 70mV, almost 140% more, To go from 30% to over maximum for the tube. But if it is only momentary...??
To me it looks like if you tried to bias it at idle around 70% you will be way way over while playing.
Maybe someone more knowledgeable on this bias circuit and current swing will chime in? If it is okay momentarily to go beyond what the tube is designed for, but on average is below the max for the tube?
-
The only time the reading across the 1R resistor is relevant is at idle with no AC signal applied. Readings under other conditions should be done for recreational use only.
-
It seems like a pretty big swing from 30mV to 70mV, almost 140% more, To go from 30% to over maximum for the tube. But if it is only momentary...??
21k bias resistor in place, the idle current is 18-19mV and that jumps to 70mV while playing really hard but only momentary, it did hit that much 2 or 3 times in about 5 minutes.
With the bias resistor, as you increase its resistance you are supplying less and less -Vdc to the tube. Less -Vdc = more current. The max safe amount for that resistor is something that gives you the correct bias. Unsafe would mean you have too much current for the 6L6 and it will red plate and/or not last as long as you might want it.
Tube data says:
Ia = 72 mA(se) 134 mA(pp)
Ig2 = 5 mA 11 mA
which one to look at ?
I think this is the same question as above but a in different way, what is the safe current ?
-
Tube data says:
Ia = 72 mA(se) 134 mA(pp)
Ig2 = 5 mA 11 mA
which one to look at ?
I think this is the same question as above but a in different way, what is the safe current ?
Ia is anode current
Ig2 is a grid current
I think your okay on your bias, any spikes would be just monetary, your average is within limits. You like the sound and you wont be hammering the guitar 100% of the time ;-)
-
Tube data says:
Ia = 72 mA(se) 134 mA(pp)
Ig2 = 5 mA 11 mA
which one to look at ?
Neither. Those are typical characteristics for Class A1 at operating points that aren't even close to yours. They have nothing to do with the maximum current that your tubes can draw. The only number on there that is of any use to you is the 30W rating under limiting values. The only good that number is going to do for you is to determine a bias point that you desire.
The bias point has nothing to do with the maximum current your tubes are going to draw at full tilt. There is no way you are going to measure this current with a 1R resistor and a $6 meter.
-
Tube data says:
Ia = 72 mA(se) 134 mA(pp)
Ig2 = 5 mA 11 mA
which one to look at ?
Neither. Those are typical characteristics for Class A1 at operating points that aren't even close to yours. They have nothing to do with the maximum current that your tubes can draw. The only number on there that is of any use to you is the 30W rating under limiting values. The only good that number is going to do for you is to determine a bias point that you desire.
The bias point has nothing to do with the maximum current your tubes are going to draw at full tilt. There is no way you are going to measure this current with a 1R resistor and a $6 meter.
I don't know that I'd 100% say "no way" it's going to be wildly inaccurate, and at best you could massively go below expected 70% dissipation rating, say 60% just to be sure the 'errors' in that cheap of a meter aren't going to cook tubes ;). but I do agree it's not super smart. If you want to be doing this, you should at least try to buy one of th 50$ or so meters out there on amazon or the like that have a moderate rating etc. Or if you don't even want to go full 50$ range, this meter has decent ratings for the price: https://smile.amazon.com/Mastech-MS8268-MS8261-Digital-Multimeter/dp/B000JQ4O2U/ref=sr_1_5?ie=UTF8&qid=1488991635&sr=8-5&keywords=multimeter+autoranging and is only 24$.
That's loads better than a harbor freight 6$ one. (I tried using one at my father in laws because I was out of state and didn't have any of mine, and MAN that thing sucked!)
~Phil
-
That's loads better than a harbor freight 6$ one. (I tried using one at my father in laws because I was out of state and didn't have any of mine, and MAN that thing sucked!)
~Phil
:laugh:
Man thanks for the hint ! In the city downtown there a neighborhood that have several stores that sells electronic stuff, best prices there too, been asking in one of the big stores, the price range for a good meter, starts at around R$100(U$32) and goes up to R$1000(U$320) and beyond ! Check out: http://loja.multcomercial.com.br/catalogsearch/result/index/?p=1&q=multimetro (http://loja.multcomercial.com.br/catalogsearch/result/index/?p=1&q=multimetro)
About the bias circuit, I found a 25.4k resistor and I did like it a lot how the amp is sounding now, it doesn't go too hot like with the 38k one and even the 28.5k doesn't sound as good. Looks like the balance between the tubes aren't too bad.
V7 24.8mV
V8 24.6mV
V9 26.2mV
V10 25.8mV
Gonna leave this way for now..... gig next saturday !!!!
-
I don't know that I'd 100% say "no way" it's going to be wildly inaccurate, and at best you could massively go below expected 70% dissipation rating, say 60% just to be sure the 'errors' in that cheap of a meter aren't going to cook tubes ;).
The current that I was referring to was the maximum current that an output tube draws, not the DC current at idle. It sounded like they were trying to determine maximum current by playing a 6-string chord and measuring DC volts across a 1R resistor.
-
Oh gotcha, you meant specifically that's the 'wrong way' to measure it period. Or do you mean the crap one couldn't measure the mV across a 1ohm on the cathodes?
~Phil
-
> the price range for a good meter
The R$ 31 meter (US$ 10) looks to my eyes like a $10 Chinese meter. These are often fine meters.
I am not sure what you get "more" for the R$ 60 (US$20) meter.
Both are safety "CAT1". This means you can use them inside appliances (including guitar amplifiers), but maybe not in a home fuse-box and certainly not at your electric meter (which you should never do-- your electric supply system has enough problems without you messing with their stuff).
-
Oh gotcha, you meant specifically that's the 'wrong way' to measure it period. Or do you mean the crap one couldn't measure the mV across a 1ohm on the cathodes?
I think that a $6 meter can measure the relatively flat DC voltage at idle across a 1R resistor. But when you get that tube going and the current is wildly fluctuating at a wide array of frequencies, who knows how that meter (or any meter) will interpret it. So I am questioning the methodology. Who cares about the maximum current draw, anyways? It is a function of just about everything EXCEPT the bias point.
-
. . . the price range for a good meter, . . .
I like the Multimetro Osciloscopio at R$ 2,540.00 with three easy payments of $265.14 USD (cheaper than yesterday). Just kidding, don't buy it.
With very little objective reason, I like the Hikari HM-2010.
-
> With very little objective reason, I like the Hikari HM-2010 (http://loja.multcomercial.com.br/multimetro-digital-portatil-hm-2010-hikari.html).
Because it is yellow. But it is not a Fluke.
It does however have " Impedāncia de Entrada: 10MOhms", a useful spec not given on the ten-buck meter?
-
The majority of meters seem to be yellow these days, probably to imitate Fluke and increase their sales to confused dopers.
This is one of the few that lists some specifications, even more than Hikari lists on their website. It has an input impedance of 10M for DC and 1M for AC, but it doesn't list the input impedance for resistance readings.
You could also print out a sheet with all of the choices on it, throw a dart at it and select the one closest to the dart penetration point.
-
> it doesn't list the input impedance for resistance readings.
Probably meaningless as stated. If the meter has resistance it just confuses things.
DMMs usually have a constant-current source ("infinite" resistance) and a volt-meter (not necessarily with the 10Meg used for volt-only measurement). So they try to be "infinite". Which can't happen. But if the ohmeter Z is much-more than the highest resistance it will measure, that is OK.
Yes, the highest (and lowest!) resistance the DMM will read correctly is sometimes important. But >10Meg gets into leakage of cheap plastics, and sub-Ohm requires special technique to control contact errors.