Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: Mike_J on March 21, 2021, 10:31:00 am
-
The Hoffman Amplifier Common Hookups uses five components whereas the 5f6a Bassman layout requires seven components and an additional column on the turret board. A column of turret space can be some valuable real estate. Who knows what could be put there.
As for my question for future purposes are either bias circuits equally useable and is there an advantage to using seven components over five?
Thanks
Mike
-
If referring to an internet page, it’s handy to provide a link.
A quick search doesn’t bring it up.
I guess that the 5F6A has an extra stage of filtering for the bias supply? If so, it will have less ripple. Dunno that I’ve ever noticed a tangible benefit from that in regard of amp performance.
But it’s nice not to feed ripple into power valve grids.
For push pull amps, the ripple is common mode, so mostly gets cancelled out in the class A region of operation. But if / when signal level pushes operation into class B, that cancellation ceases.
Beware of excessive filtering on the bias supply, as it would be bad to slow its response such that it was still coming up when the valves have warmed up and are fully conducting.
-
Beware of excessive filtering on the bias supply, as it would be bad to slow its response such that it was still coming up when the valves have warmed up and are fully conducting.
As it stands right now the bias supply has its own filtering which looks to be 20uF/100v. It is grounded along with the power tube cathodes and PT CT to the B+ filter supply ground. It is also connected to the ground on the filter cap for the screen supply. The Fender reissue amps use 11uF/126v plus a resistor then another 10uF/100v cap in parallel it looks like. The original schematic shows what looks like 16uF/150v. Can't see anything wrong with the values the amp is using now can you?
Thanks
Mike
-
Beware of excessive filtering on the bias supply, as it would be bad to slow its response such that it was still coming up when the valves have warmed up and are fully conducting.
Put a standby switch in to mitigate this evil then filter to your heart's content.
-
Beware of excessive filtering on the bias supply, as it would be bad to slow its response such that it was still coming up when the valves have warmed up and are fully conducting.
Put a standby switch in to mitigate this evil then filter to your heart's content.
Since this amp has a GZ34 I am going to forget this amp has a standby switch.
-
Since this amp has a GZ34 I am going to forget this amp has a standby switch.
I was only teasing pdf64 when I suggested a standby switch.
-
Since this amp has a GZ34 I am going to forget this amp has a standby switch.
I was only teasing pdf64 when I suggested a standby switch.
I’ve had a cup of tea and calmed down :laugh:
-
...Can't see anything wrong with the values the amp is using now can you?
No, that issue only starts to arise with much longer time constants, eg caps >10x higher.
-
If referring to an internet page, it’s handy to provide a link.
Hear hear.
-
If referring to an internet page, it’s handy to provide a link.
A quick search doesn’t bring it up.
I guess that the 5F6A has an extra stage of filtering for the bias supply? If so, it will have less ripple. Dunno that I’ve ever noticed a tangible benefit from that in regard of amp performance.
But it’s nice not to feed ripple into power valve grids.
For push pull amps, the ripple is common mode, so mostly gets cancelled out in the class A region of operation. But if / when signal level pushes operation into class B, that cancellation ceases.
Beware of excessive filtering on the bias supply, as it would be bad to slow its response such that it was still coming up when the valves have warmed up and are fully conducting.
Sorry I didn't refer to an internet page. If I had any idea how to do that I certainly would have. Any advice on how to do same would be greatly appreciated. This bias supply has two 10uF/500v caps in parallel. Hoffman's five piece bias supply has a single 47uF/100v. So as I see it the five piece bias supply has better filtering. Pretty sure the reason used the seven piece in the 5f6a is because he was closely following the bias supply the 5f6a schematic showed with the addition of a bias pot. The five piece bias supply closely follows the AB763 bias supply. I am sure either one would work in either amp although I would lean toward the five piece in the future due to the increase in filtering, extra column of board space and maybe two. Plus I have a lot of 47uF/100v radial caps so parts are on hand.
In the meantime, have a new shiny thing to dwell on that I hoped you or at least somebody would be interested in. As I may have mentioned many times wanted to place ecaps on the main board which I did. Required that some components be moved to accommodate doing so. Normally the power tube cathode grounds and the bias supply grounds would be all on the same row. However, not now.
In order to tie their grounds to the reservoir capacitor or first supply capacitor or B+ capacitor, whatever we are going to call it today, I created a rats nest of ground wires in a very small space. Not only is it unsightly, I am somewhat concerned it may cause a problem with the noise level in the amp. What I think I will do is tie the bias grounds to the same place the PT CT is terminated. After all they are related by being part of the PT. There is a wire that goes from that point to the negative side of the B+ ecap. Will run the power tube cathode grounds directly to the same point on the B+ ecap. Will clean things up nicely. Probably not the least bit necessary but not the first or last time I will have done that.
Thanks
Mike
-
If referring to an internet page, it’s handy to provide a link.
Hear hear.
I just learned how to attach a picture to my post. Would be glad to attach a reference to an internet page if I had a little tutorial or at least some advice on how to do it.
Thanks
Mike
-
If referring to an internet page, it’s handy to provide a link.
Hear hear.
As an extension of the above, I would really like to know how to add a layout or schematic and draw the nice little boxes around components like some people know how to do. If I could do that I would really be in tall cotton.
Thanks
Mike
-
I’m crap at cad stuff, but even I find MS Paint usable to make basic changes.
To link to an internet page, highlight the address, copy and paste it into your post.
Sorry I didn’t appreciate you might be stuck on that.
If the above doesn’t make sense, ask again :icon_biggrin:
-
I’m crap at cad stuff, but even I find MS Paint usable to make basic changes.
To link to an internet page, highlight the address, copy and paste it into your post.
Sorry I didn’t appreciate you might be stuck on that.
If the above doesn’t make sense, ask again :icon_biggrin:
Link to the internet page by highlighting the address and copying and pasting it into the post makes good sense. Even thought I have autocad 2020 on this computer I have no idea how to use it. Maybe MS Paint is on there as well.
Thanks
Mike
-
Don't see MS Paint but do see a Paint 3D program. I wonder if that would be helpful?
-
You may do better than me, but I don’t get on with the 3D paint. I get the impression I’m not alone in that.
Just type ‘paint’ into the search box at the start menu.
Then copy the app into your start menu or the task bar, so as to find it again more easily.
-
You may do better than me, but I don’t get on with the 3D paint. I get the impression I’m not alone in that.
Just type ‘paint’ into the search box at the start menu.
Then copy the app into your start menu or the task bar, so as to find it again more easily.
After considerable mental anguish I figured out what you were saying and got Paint to come up. Now I have no idea what to do next. I would like to pull up the 5f6a schematic and draw a circle around the bias components.
Thanks
Mike
-
... it’s nice not to feed ripple into power valve grids.
For push pull amps, the ripple is common mode, so mostly gets cancelled out in the class A region of operation. But if / when signal level pushes operation into class B, that cancellation ceases. ...
I fixed a Princeton Reverb once that had noticeable 60Hz hum. My first step was pulling the phase inverter tube, but I still heard 60Hz hum. The bias filter cap had failed, and a new one killed the hum.
Yes, bias-ripple is common-mode, but we almost never have perfect balance in the output section.
That lead to the hum being "obvious and annoying" but not "15 watts of solid hum" as would happen if the output section were receiving a push-pull signal rather than this common-mode signal.
-
...
I fixed a Princeton Reverb once that had noticeable 60Hz hum. My first step was pulling the phase inverter tube, but I still heard 60Hz hum. The bias filter cap had failed, and a new one killed the hum...
I guess that the failed reservoir cap would have also reduced the bias voltage magnitude, as the VDC would normally charge up to the peak VAC, but would now only reach the average (or rms?) level?
So the 6V6 may have been idling a fair bit hotter than normal, which might increase HT ripple and so increase hum related to that (though it would be 120Hz, rather than 60)?
-
You may do better than me, but I don’t get on with the 3D paint. I get the impression I’m not alone in that.
Just type ‘paint’ into the search box at the start menu.
Then copy the app into your start menu or the task bar, so as to find it again more easily.
After considerable mental anguish I figured out what you were saying and got Paint to come up. Now I have no idea what to do next. I would like to pull up the 5f6a schematic and draw a circle around the bias components.
Thanks
Mike
Here is my first attempt to use MS Paint. Haven't figured out how to pull up a schematic or layout and put lines or a circle around things yet but hopefully that will happen in the near future. What I put red lines is a Hoffman bias circuit for the 5f6a Bassman, slightly adjusted by me.
-
You may do better than me, but I don’t get on with the 3D paint. I get the impression I’m not alone in that.
Just type ‘paint’ into the search box at the start menu.
Then copy the app into your start menu or the task bar, so as to find it again more easily.
Thanks pdf64. I have finally used MS Paint to identify something on a picture. Too bad it is the most terrible picture I ever took but it let's me know I can do it.
Thanks
Mike
-
I think paint can only open a limited range of picture file types, eg .jpg, not pdfs.
Just download a suitable file, this may be ok https://el34world.com/charts/Schematics/files/Fender/Fender_bassman_5f6a_schem.gif and open it!
-
I use paint frequently to modify schematics.
If copying a schematic from Adobe (pdf.), often times, the file will be protected, thus preventing the image from being copied. The solution is to print the document and save the print copy as a pdf. file under a different file name. From the Adobe print button select Microsoft Print to PDF from the printer options. Click on Print, after the file has printed, you will be queried for a file name to save the new pdf file.
Open the new pdf file and select the image that you want to copy. Click on the edit button, select the Take a Snapshot option. Open MS Paint and then click on the paste button. The selected image will be copied into paint. You nay find the need to resize the image. To resize , use the percent option to resize. If you enter the percentage to resize in the horizontal box, the vertical box will automatically adjust to the percentage selected in the horizontal box.
From there, you can use the MS Paint tools to modify the image as desired. Once you're done, use the Save As option to save your Paint pdf file. I like to save the files as PNG picture files. Once saved as a PNG file, the document can be re-opened with MS Paint, if further editing is needed.
-
I use paint frequently to modify schematics.
If copying a schematic from Adobe (pdf.), often times, the file will be protected, thus preventing the image from being copied. The solution is to print the document and save the print copy as a pdf. file under a different file name. From the Adobe print button select Microsoft Print to PDF from the printer options. Click on Print, after the file has printed, you will be queried for a file name to save the new pdf file.
Open the new pdf file and select the image that you want to copy. Click on the edit button, select the Take a Snapshot option. Open MS Paint and then click on the paste button. The selected image will be copied into paint. You nay find the need to resize the image. To resize , use the percent option to resize. If you enter the percentage to resize in the horizontal box, the vertical box will automatically adjust to the percentage selected in the horizontal box.
From there, you can use the MS Paint tools to modify the image as desired. Once you're done, use the Save As option to save your Paint pdf file. I like to save the files as PNG picture files. Once saved as a PNG file, the document can be re-opened with MS Paint, if further editing is needed.
That is a fantastic tutorial.
-
I use paint frequently to modify schematics.
If copying a schematic from Adobe (pdf.), often times, the file will be protected, thus preventing the image from being copied. The solution is to print the document and save the print copy as a pdf. file under a different file name. From the Adobe print button select Microsoft Print to PDF from the printer options. Click on Print, after the file has printed, you will be queried for a file name to save the new pdf file.
Open the new pdf file and select the image that you want to copy. Click on the edit button, select the Take a Snapshot option. Open MS Paint and then click on the paste button. The selected image will be copied into paint. You nay find the need to resize the image. To resize , use the percent option to resize. If you enter the percentage to resize in the horizontal box, the vertical box will automatically adjust to the percentage selected in the horizontal box.
From there, you can use the MS Paint tools to modify the image as desired. Once you're done, use the Save As option to save your Paint pdf file. I like to save the files as PNG picture files. Once saved as a PNG file, the document can be re-opened with MS Paint, if further editing is needed.
I could use your help a little, maybe a lot. Think I got to the select the image you want to copy but can't seem to find an edit button. Sure I am in the wrong place but have no idea where the right place is. Are you in adobe?
-
Yes, in Adobe. The edit button is at the top left hand corner of the screen, just under the header line.
-
Let me see if I got this figured out. My MS Paint had a gif selection. Hoffman had a gif 5f6a file.
-
Yes, in Adobe. The edit button is at the top left hand corner of the screen, just under the header line.
Will have to tackle the Adobe next because most of his schematics were saved in pdf. Thanks much for your help. Without it I would be lost and may still be lost if I don't figure out how to get on Adobe. If I didn't mention it the Paint file accepts gif for some reason but I didn't see pdf.
Thanks
Mike
-
I just use the ‘print screen’ button on the keyboard. Then paste that into paint.
-
MS Paint does not recognize Adobe (pdf. files). That's why pdf. file images must be copied in Adobe and pasted directly into an open MS Paint document.
-
I just use the ‘print screen’ button on the keyboard. Then paste that into paint.
Thanks will try it.
-
MS Paint does not recognize Adobe (pdf. files). That's why pdf. file images must be copied in Adobe and pasted directly into an open MS Paint document.
Okay. Will try that next.
Thanks
Mike
-
I guess that the failed reservoir cap would have also reduced the bias voltage magnitude ...
Agreed that must have happened to some degree. But there was no red-plating when the amp came in for repair, it just seemed like the average tube amp with some hum.
Consider though ripple at the OT center-tap is canceled, but for the imperfect-balance I noted earlier. Large ripple at the tube screens would be an issue, but that's also a common-mode input.
-
Have attached the Hoffman 5f6a and AB763 bias layouts. The AB763 is definitely superior in my opinion for a number of reasons. First it only takes up four columns on the board instead of five for the 5f6a bias supply. Second the 5f6a has I believe 20uF of bias supply filtering versus 47uF for the AB763 and third and most important, you can get a 10K bias pot that is accessible from outside of the amp. Therefore, you don't have to pull the chassis to re-bias the amp. Just put some red bias tip jacks between the cathode pins on the power tube sockets and the one ohm resistors on the board where the red As are shown and a black bias tip jack connected to the positive side of the 47uF/100v ecap that later gets connected to chassis ground. You can connect your MM to the bias tip jacks. No need to pull the chassis for biasing again unless a different bias range resistor is needed.
Hoffman has the bias tip jacks and the 10K bias pot that can be accessed from outside the amp in his store.
Thanks
Mike
-
Here is a picture of my current bias supply along with a picture of the Hoffman AB763 bias layout. Can take out the 50K trimmer pot and also remove the 10uF ecap next to it. Then would need to drill a hole big enough in the board to get the pot in with the tabs still exposed and rewire with the correct value components. Also would need to add the bias tip jacks. Not a hard job but would make future biasing easier unless a bias range resistor change is needed.
Thanks
Mike
-
I like the 5F6A arrangement; 2nd order filtering is more effective that 1st order, and the trimmer pot is fail safe.
-
Have attached the Hoffman 5f6a and AB763 bias layouts.
It is sooooooooo much easier to analyze a circuit looking at the schematic than looking at the layout drawing. Much harder to see the flow of the circuit in the layout drawing. The layout drawing is for finding where the parts are on the eyelet/turret.
The AB763 is definitely superior in my opinion for a number of reasons. .... Second the 5f6a has I believe 20uF of bias supply filtering versus 47uF for the AB763......
I like the 5F6A arrangement; 2nd order filtering is more effective that 1st order, and the trimmer pot is fail safe.
It's not just the filter caps total value. With 2 caps in series with a resistor in between, that's a 'pi' filter, (because it looks like the 'pi' symbol on the schematic drawing) and that R between the 2 caps increases the filtering. There's math for it.
Read this, on power supply filtering; http://www.valvewizard.co.uk/smoothing.html (http://www.valvewizard.co.uk/smoothing.html)
And this, on bias supplies; http://www.valvewizard.co.uk/bias.html (http://www.valvewizard.co.uk/bias.html)
and third and most important, you can get a 10K bias pot that is accessible from outside of the amp. Therefore, you don't have to pull the chassis to re-bias the amp. Just put some red bias tip jacks between the cathode pins on the power tube sockets and the one ohm resistors on the board where the red As are shown and a black bias tip jack connected to the positive side of the 47uF/100v ecap that later gets connected to chassis ground. You can connect your MM to the bias tip jacks. No need to pull the chassis for biasing again unless a different bias range resistor is needed.
Hoffman has the bias tip jacks and the 10K bias pot that can be accessed from outside the amp in his store.
Not always, you need to check/recheck the power tubes DCV before and after biasing the power tubes. I like to install them, makes it easy to put your meter probes in and not worry about them falling over and shorting out something.
-
Have attached the Hoffman 5f6a and AB763 bias layouts.
It is sooooooooo much easier to analyze a circuit looking at the schematic than looking at the layout drawing. Much harder to see the flow of the circuit in the layout drawing. The layout drawing is for finding where the parts are on the eyelet/turret.
The AB763 is definitely superior in my opinion for a number of reasons. .... Second the 5f6a has I believe 20uF of bias supply filtering versus 47uF for the AB763......
I like the 5F6A arrangement; 2nd order filtering is more effective that 1st order, and the trimmer pot is fail safe.
It's not just the filter caps total value. With 2 caps in series with a resistor in between, that's a 'pi' filter, (because it looks like the 'pi' symbol on the schematic drawing) and that R between the 2 caps increases the filtering. There's math for it.
Read this, on power supply filtering; http://www.valvewizard.co.uk/smoothing.html (http://www.valvewizard.co.uk/smoothing.html)
And this, on bias supplies; http://www.valvewizard.co.uk/bias.html (http://www.valvewizard.co.uk/bias.html)
and third and most important, you can get a 10K bias pot that is accessible from outside of the amp. Therefore, you don't have to pull the chassis to re-bias the amp. Just put some red bias tip jacks between the cathode pins on the power tube sockets and the one ohm resistors on the board where the red As are shown and a black bias tip jack connected to the positive side of the 47uF/100v ecap that later gets connected to chassis ground. You can connect your MM to the bias tip jacks. No need to pull the chassis for biasing again unless a different bias range resistor is needed.
Hoffman has the bias tip jacks and the 10K bias pot that can be accessed from outside the amp in his store.
Not always, you need to check/recheck the power tubes DCV before and after biasing the power tubes. I like to install them, makes it easy to put your meter probes in and not worry about them falling over and shorting out something.
Do you have an idea for how we could put a second cap and resistor in the AB763 circuitry. I like the 10KL pot that you can bias without having to take the amp out of the chassis. Really don't I guess with the Bassman because once you remove the back cover it is accessible.
Spent the whole day today converting to the AB763 bias supply and putting bias tip jacks in. Always in favor of making something better though.
Thanks
Mike
-
I like the 5F6A arrangement; 2nd order filtering is more effective that 1st order, and the trimmer pot is fail safe.
Could you explain how the trimmer pot is fail safe but the 10K pot isn't I guess? Or is it the arrangement with the ecaps and resistor that makes it safe?
Have quite a few 47uF/100v ecaps. Can those be doubled up. Nothing to making a board once the values are decided.
Thanks
Mike
-
Do you have an idea for how we could put a second cap and resistor in the AB763 circuitry. I like the 10KL pot that you can bias without having to take the amp out of the chassis. Really don't I guess with the Bassman because once you remove the back cover it is accessible.
Did you look at this?
And this, on bias supplies; http://www.valvewizard.co.uk/bias.html (http://www.valvewizard.co.uk/bias.html)
-
Do you have an idea for how we could put a second cap and resistor in the AB763 circuitry. I like the 10KL pot that you can bias without having to take the amp out of the chassis. Really don't I guess with the Bassman because once you remove the back cover it is accessible.
Did you look at this?
And this, on bias supplies; http://www.valvewizard.co.uk/bias.html (http://www.valvewizard.co.uk/bias.html)
Looked at Merlin on bias supply since I last posted. Good news is I think I can install a bias supply with only three columns of turrets using the Merlin standard method for amps with a center tap on their PT. His method goes right into the diode. Avoids the bias range resistor although I am not sure which resistor he changes if the pot isn't giving the proper range. Probably the 100K from CW to the wiper on the pot. That is how I am saving a column of turrets by the way. Placing the 100K resistor on the pot. It says it is the bias doubler but I edited the doubler circuit out and just included the standard. In this case I am theoretically saving five columns by putting the bias circuit on the side of the chassis and the 10K between the last power tube socket and the rectifier tube socket.
Wish I would have given more thought to doing this. Remembered the AB763 bias circuitry and pot was on the bottom of the chassis. So of course I hacked the bias circuitry out of the board to make room for the bias circuit and pot. After finishing this it became apparent that there would be no way to adjust the bias from outside the board if the adjustment is facing the baffle board. That is how the circuitry got moved to the other side of the PT on the end panel of the chassis and the pot got moved where it is.
Very much appreciate your help Willabe. You have given me great advice in the past and I appreciate it.
Thanks
Mike
-
Not sure if it will suit your build, but Doug has an AB763 bias board -- it's pretty compact -- works great. It is normally used in conjunction with his AB763 "short" board. All the details in his parts/kits section.
-
Very much appreciate your help Willabe. You have given me great advice in the past and I appreciate it.
Your welcome, I have learned a LOT from guys here, try to give back to others. :icon_biggrin:
-
If you want to keep the size down, use modern radial lead ecaps and stand them up. :icon_biggrin:
Have quite a few 47uF/100v ecaps. Can those be doubled up.
Yes, those should work great.
-
If you want to keep the size down, use modern radial lead ecaps and stand them up. :icon_biggrin:
Have quite a few 47uF/100v ecaps. Can those be doubled up.
Yes, those should work great.
Thanks Willabe. The 47uF/100v ecaps are radials. Like to use them when I can. Might use the 10uF/100v caps I took off the Hoffman board since Merlin said we could use anything from 10uF to 100uF, it didn't make a difference. Now if this was truly the voltage doubler circuit then the capacitance would need to be doubled.
Made a layout diagram for those of us that see things that way. For anyone ever thinking of doing something like this let me do my best to dissuade you from such a notion. I spent about one-half day removing the Hoffman bias circuit section out of the board while it was attached. Looks like an army of rats came in and gnawed it off. It is a hideous mess. Additionally, put another hole, this time 3/8" in the chassis. Not a huge problem because I can just cover it with aluminum tape. Problem is this amp already has so much aluminum tape on it if anyone asks me what metal my chassis is made out of I am going to have to say aluminum tape.
A summary of what I am saying is this. Leave the bias section the way the Hoffman layout shows it. I was thinking of all my other amps, with the exception of the 5E3 which is cathode biased, where the opening in the chassis faces either up or down. Doing this makes no sense with a Bassman chassis with the opening of the chassis facing outward.
Thanks
Mike
-
The 100k is the ‘fail safe’ back up, ensures some bias is available even if the pot fails. It should be about 10x the pot track resistance; much higher and the fail safe aspect will be less reliable, much smaller and the range will be reduced, ie magnitude won’t go as low.
If there’s far more bias voltage than might ever be required, a resistor will need to be added, eg before the diode, to reduce it. So it may be best to make provision for that.
Please don’t refer to doublers, especially voltage doublers, in regard of such power supplies, as they are a very different thing.
...you need to check/recheck the power tubes DCV before and after biasing the power tubes...
I think too much is made of that. For a given amp, there will be a fixed relationship between mains voltage, anode current and the HT voltage. Once we know the HT voltage for the max safe anode current at our nominal mains voltage, we only ever need to measure anode current.
Of course mains voltage will vary from nominal. The beauty of a design centre rated valve is that variance is accommodated for within the rating. Whereas with a design max rating, further work is required to derive the max safe anode current.
Whatever, point is there’s no purpose in rechecking the HT voltage every time.
Doing so may be more liable to result in an overly hot bias, if the mains voltage happens to be under its usual level at the time and a target of the highest ’safe’ idle dissipation is being used.
-
The 100k is the ‘fail safe’ back up, ensures some bias is available even if the pot fails. It should be about 10x the pot track resistance; much higher and the fail safe aspect will be less reliable, much smaller and the range will be reduced, ie magnitude won’t go as low.
If there’s far more bias voltage than might ever be required, a resistor will need to be added, eg before the diode, to reduce it. So it may be best to make provision for that.
Please don’t refer to doublers, especially voltage doublers, in regard of such power supplies, as they are a very different thing.
Only reason I mention doublers is the picture I imported into Paint 3D or whatever it is called had both the standard circuitry and the doubler circuitry on it. I removed the doubler circuitry and saved it under a new name. However, the only one I could find was the one they called doubler so I didn't want to be misleading anyone that the circuit is a doubling circuit.
Will include an allowance for the resistor as you suggested. Thank you for the information you provide it is very helpful and I am starting to understand a lot more of what you are saying which is gratifying.
Thanks
Mike
-
Here are the latest pictures. Board is starting to look like Hoffman's with some different values. Board is in dire need of a little clean up. Hard to see without the closeup shots.
Thanks
Mike
-
The bias board looked so bad close up I wanted to do a comparison of how it looked versus how it looked after a little brushing of denatured alcohol on it.
-
There is a re: Power up amplifier first time Procedures post on the forum. On January 1, 2008, Reply 5, by PRR where he made comments which I think are clear enough that they should be mentioned again. Note: earlier he mentioned putting the rectifier tube in which must be done to get any voltage to the 220K resistors. PRR also mentioned the use of a light limiter would be wise before doing this step to be comfortable there is no problem up to this point. I would imagine you would disconnect the light limiter to do what he is suggesting and by which way gave the most bias voltage he meant the most negative bias voltage. Hopefully he will confirm these points.
“CAREFULLY use clip-leads to connect a voltmeter, Black to ground and Red to the point where the 220K resistors come together. Plug-in, turn on. This bias voltage should go to NEGATIVE 40V or 50V, and it should be adjustable with the trimmer. Don't linger, the unloaded voltage may be hard on the filter caps. But you MUST be sure you have LOTS of NEGATIVE bias voltage, before you put tubes in. If you can't get good bias, come back. Note which way gave the most bias voltage, and leave it set that way.”
-
Fired up the amp with the lamp limiter and rectifier tube in only. It started out with little light and went to no visible light. Figured that meant I was okay to go. Disconnected the lamp limiter and measured the range of voltage the 10K has and it went from -32VDC to -66VDC. Figure I am okay to proceed to the next step. Now I need to figure out what the next step is. Unfortunately I know what the next step is. Doing it over and leaving the bias pot at the highest negative voltage level. Got so excited I forgot to do that important step.
Thanks
Mike
-
Leave the LBL in play until reliable operation, eg valves working with signal, has been proven.
When that’s done, take it out of circuit, take true voltages and set bias etc.
-
Leave the LBL in play until reliable operation, eg valves working with signal, has been proven.
When that’s done, take it out of circuit, take true voltages and set bias etc.
Will do. Thanks.
-
Leave the LBL in play until reliable operation, eg valves working with signal, has been proven.
When that’s done, take it out of circuit, take true voltages and set bias etc.
Wherever you said 23 watts of dissipation was the center not the maximum is consistent with what Rob Robinette is using for his 5881 biasing. He is using 26 watts as the maximum so I created a chart with the 26 watts and will consider the 70% dissipation at 26 watts as my max for biasing purposes.
Thanks
Mike
-
Looking at my book and came up with this Humble schematic from 1997 which I have attached. In all of his ecap stages Humble had two caps in series with 470K resistors across all of them. There is also a close up clandestine microchip type markings on the schematic which I circled and took a closeup picture of which I suppose identified values someone was trying to hide. The extent people will go to try to uncover things other people are trying to hide is staggering. Thought it was a little bit of interesting history.
I read something by Merlin, Willabe suggested, http://www.valvewizard.co.uk/smoothing.html commenting on smoothing and filtering the power supply. Specifically dealing with resistors placed across ecaps in series to make sure power is being shared somewhat equally by both caps. “The resistors should be equal to 50/C or less, so two 100uF capacitors would each need a 500,000ohm resistor (470k would be the obvious choice). These also act as bleeders when the amp is switched off.”
I guess he is talking about farads which is why he is multiplying by one million? Also, he says or less. By less, does that mean the Fender value of 220K is fine? Also, if my calculations are correct, I would need a pair of one meg resistors across a pair of 47uF in series. Does this make any sense to anyone?
Another thing Merlin recommends is;
“If space is tight then we might want to use smaller caps for the preamp, say 22uF. For decent hum reduction we should push the cut-off frequency down to a really low frequency, like 1Hz. This will mean the dropping resistors need to be at least:
R = 1 / (2 pi f C) = 1 / (2 pi × 1 × 22×10^-6 ) = 7,234 ohms.
However, since 6mA of preamp current will have to flow in the first resistor, this would drop quite a lot of voltage, maybe too much. We might therefore want to relax the first resistor to 4.7k, say, which would drop only 6mA × 4.7k = 28V and dissipate 0.006^2 × 4,700 = 0.17 watts. Later dropping resistors could perhaps be larger, although 22uF and 4.7k give a cut-off frequency of 1.5Hz which is quite respectable.”
This would mean a 22uF cap for V2 and a 22uF cap for V1 with 4.K dropping resistors to each ecap whereas they currently share one 22uF for both tubes. When both tubes share a 22uF cap there is a 10K dropping resistor. Anyone see an advantage of increasing the voltage to the V2 plate load cap? By the time you get to V1 the plate voltages would be nearly the same. Have 44uF on V1 and V2 now combined as per the Bassman reissue schematic.
Thanks
Mike
EDIT: parse URL