# Harvard in a 5F2A Chassis Build

Discussion in 'Shock Brother's DIY Amps' started by James Knox, Nov 22, 2020.

1. ### James KnoxTele-HolicSilver Supporter

Posts:
925
Joined:
Aug 13, 2003
Location:
Phoenix
Thx man! I had just a minute tonight to solder them in. I’ll play around with the bias mañana.

jsnwhite619 likes this.
2. ### James KnoxTele-HolicSilver Supporter

Posts:
925
Joined:
Aug 13, 2003
Location:
Phoenix
Ok, I think she (the build) is “a wrap”. Amp sounds amazing! Full power or 1/10th Power. Great woolly, edge of break up sound. Here are the final notes...

Got the 1 ohm Resistors in...

Clipped up so I could have hands free to adjust the bias pot and take a pic, lol...

My notes as I worked through the process...

The RobRob Bias Calc Bottom Line. Special shout out to @robrob for providing an amazing resource and to @Lowerleftcoast and @King Fan for teaching me how to use it!

jsnwhite619 likes this.
3. ### LowerleftcoastFriend of Leo's

Posts:
3,205
Joined:
Dec 30, 2019
Location:
california

If you are using the 1% resistor method, continue on with Rob's Bias Calculator. The calculator subtracts a estimate of 5.5% screen current.
Simply enter your numbers into the section *Tube Dissipation using Cathode Resistor Voltage Drop*.
Enter Number of Tubes that share the 1ohm cathode resistor = 1
Enter voltage across cathode resistor = .025
Enter cathode resistor ohms = 1

The calculator will fill out the rest of the info. So, subtracting an estimated 5.5% screen voltage shows a 60.8% PD.

Stop reading here if you perfectly happy with the estimated result.
------------------------------------------------------------------------------------

Now if we are looking for that "gnat's ass of precision" mentioned in post #132.

@James Knox measured the Plate current for tube V4 as 20.41mA.
Plate current .02041 x Plate voltage 309.4 = Plate Dissipation 6.3watts.
Plate Dissipation 6.3watts / 12watt tube rating = Plate dissipation 52.6%

The Screen Current can be calculated:
(Plate and Screen Current) 25mA - (Plate Current) 20.41mA = (Screen Current) 4.59mA.

Rob's calculator estimated the screen current at 5.5%.
The calculated screen current in this case was just over 18%

disclaimer: Do not use the ~18% figure to estimate Plate Current.
Screen current will vary from tube to tube and from amp to amp due to several factors. Rob's Calculator has an estimated 5.5% value to cover many tubes and scenarios.

LLC

robrob likes this.
4. ### James KnoxTele-HolicSilver Supporter

Posts:
925
Joined:
Aug 13, 2003
Location:
Phoenix
I am happy with the result, but I’m want to see more of the Math Magic you are sussing out of the Bias Calc. Good stuff Bro!

And, why give up a perfectly great opportunity to pass up using the phrase, “gnats ass of precision”, lol.

Lowerleftcoast likes this.

Age:
47
Posts:
674
Joined:
Jan 4, 2019
Location:
Australia
Nice job, mate. Great build! So now ya have to start building the cab

Huddy and James Knox like this.
6. ### HuddyTele-Holic

Age:
37
Posts:
703
Joined:
Nov 5, 2016
Location:
Newport News, VA
I'm calling shenanigans unless I get a sound demo

Just kidding! Looks great!

gabasa and James Knox like this.
7. ### robrobPoster ExtraordinaireAd Free Member

Posts:
8,301
Joined:
Dec 29, 2012
Location:
United States
All the formulas are listed on my bias calculator so you can calculate them manually if you'd like.

Lowerleftcoast and James Knox like this.
8. ### 2L manTele-Meister

Age:
61
Posts:
292
Joined:
Nov 23, 2020
Location:
Finland
Concratulations James, very nice build which proofs you can use your hands and that you can learn electrics as well!

I think measuring current over 1 ohm resistor on each tube cathodes is good measure of power amp health when also G2 current runs thru it and both anode and G2 currents heat the tube. Using OPT coil resistance is also good method after they are measured but sometimes they are not the same which needs to be noted. Also there is bit more danger to measure high voltages than only millivoltages over resistors.

Saying to bias to 70% is just a starting point for a push pull tube amp and playing and listening and adjusting you can find better sounding bias. Generally higher bias lessens distortion and makes sound "warmer" and therefore HiFi amps are often biases closer to 100%. But sometimes more distortion is found to sound better and bias is adjusted colder. Often it is quite easy to adjust also G2 to tune sound changing resistors in G2 circuit. I have quite cheap and durable 6P6S on my F3E and I biased them to about 95% and I think it sounds better.

One thing I noticed is you connected 1 ohm cathode resistors to chassis? I have connected signal ground to chassis only near input jacks or on Fender amps metal jack(s) do the grounding and to me power tube anode circuit belongs to signal circuit. Also chassis is grounded to mains wire/connector next to it but its function is to burn mains fuse in case operating voltages jump to chassis inside circuit. This way I think i have avoided "ground loop" You also wrote that amp is quiet! Perhaps when anode current runs thru chassis it can not greate too much cround loop? Now I think perhaps I have not done it "right" when no one else did not question it so to you all think it is better?

I understand that old Fenders did use chassis to everything possible to save on wiring, even heaters did use chassis as other current path but I thought it is better is to avoid to use chassis in actual circuit.

9. ### King FanPoster ExtraordinaireAd Free Member

Posts:
6,320
Joined:
Jan 1, 2013
Location:
Salt Lake City
if I understand the question, you wonder if it’s OK to ground those 1-ohm resistors to the chassis? It’s OK for these particular grounds, and convenient, but if you run a ground wire to a node or ground bus instead, that's fine too — in fact theoretically better. Does that help?

James Knox likes this.
10. ### James KnoxTele-HolicSilver Supporter

Posts:
925
Joined:
Aug 13, 2003
Location:
Phoenix
Agree with you about biasing by ear! I like to dial in the correct range for tube dissipation and then fine tune by ear.

I see what you are saying about grounding the 1 ohm to the chassis. I just learned about reading the bias current this way and it is new to me. This is also my first fixed bias build.

Regarding grounding schemes. I ha e been using the dual grounding mode of using a ground buss for preamp section and a different one for power section. Oh, and a separate ground post for the AC power cord.

11. ### 2L manTele-Meister

Age:
61
Posts:
292
Joined:
Nov 23, 2020
Location:
Finland
If you connect 1 ohm resistors to chassis then you must also connect power supply "0" or "center tab" to chassis? If you don't the power tube cathode current has to run thru chassis to imput jack and back thru ground bus to PT CT which should make huge ground loop?

I don't understand why chassis is used for signal grounding around power tubes where run lots of current? I have seen it done quite a lot so can anyone explain why because I see there comes two ground paths and without most current goes thru whole ground bus? Is this "partial star grounding" ?

I connect PT CT to the beginning of the ground bus and finish ground but to chassis at input jack.

On mains input I connect only mains cable ground wire to chassis and if toroidal power supply has ground wire it goes to chassis lug as well. I have used only toroidal PT lately and some have ground wire but obviously it is more for electrical shielding so I perhaps it should be connected to signal ground next to 0 ?

12. ### James KnoxTele-HolicSilver Supporter

Posts:
925
Joined:
Aug 13, 2003
Location:
Phoenix
You have some good questions that I do not know the answer to. I learned of the “dual buss” grounding scheme from studying others builds here on the Shock Bro’s a Forum and from @robrob website. Have you checked out https://robrobinette.com/Amp_Stuff.htm ? You might find some answers there! Maybe @robrob will drop by and help us out here!

13. ### 2L manTele-Meister

Age:
61
Posts:
292
Joined:
Nov 23, 2020
Location:
Finland
Actually after I made grounding "clear for myself" I have not researched it anymore but because I sometimes hear some noise and hum on my amps I have been thinking if there is a better method? In your build it was so obvious that it is different than mine when power tube current runs thru chassis that I wanted to question my way? Also engourage you to look where is your power supply center pin connected? And if your amp is noisy there is another method to build tube amp circuit?

After "star grounding" there is also "Larry grounding" which I know mostly only name but obviously I uses chassis as only signal ground but I would not use it other than on steel chassis because there are many ground points and soldering them to chassis sould be better than using screwed lugs which is only method with aluminium which is not too easy or perfect either?

14. ### King FanPoster ExtraordinaireAd Free Member

Posts:
6,320
Joined:
Jan 1, 2013
Location:
Salt Lake City
When I started out, I read *all* about ground schemes and got deeply confused. Although there are some general principles (separate safety ground, and yes, ground reference for PT center tap), much of the rest can be done a few different ways. The complex (star/node/Larry etc.) methods are really for big complicated amps or hifi. The huge majority of amps built here work well with a simple ground bus, split or single, if done correctly, with or without a few (minor) local grounds — done correctly.

You’re right. Noisy amps may reflect a ground problem, but noise can also have dozens of other causes. If you have ground/noise questions about a particular amp, you might start a thread here with details.

15. ### 2L manTele-Meister

Age:
61
Posts:
292
Joined:
Nov 23, 2020
Location:
Finland
I couldnot access Robrob web page but here is current thread on Amp Tech about best 5E3 layout and there I saw Robrobs Optimized 5E3 and it has grounding how I do my amps.

James Knox likes this.
16. ### robrobPoster ExtraordinaireAd Free Member

Posts:
8,301
Joined:
Dec 29, 2012
Location:
United States
On my dual bus layouts the power tube cathodes run by wire to the power amp ground (chassis near the PT). I recommend always connecting the power transformer high voltage center tap to the first filter cap negative lead to minimize pulsing return current in the ground bus.

With a split bus all the preamp return current does flow from the preamp ground at the input jack, through the chassis to the power amp ground.

I now normally use a unified bus with the only bus ground at the input jack. On a couple of unified bus amps with excessive 120Hz hum I had to split the bus to get them quiet. I do not know why that happens.

James Knox and King Fan like this.
17. ### LowerleftcoastFriend of Leo's

Posts:
3,205
Joined:
Dec 30, 2019
Location:
california
Several Marshall circuits use resistors to separate ground planes. Essentially most of the noise in the ground is impeded from going the *wrong way*.

In my research I came across:
http://www.valvewizard.co.uk/Grounding.pdf

In figure 15.6, he illustrates the small amount of ground impedance as a dashed line resistor.
He describes with this text:
"Although we may be used to seeing power supply smoothing filters drawn as simple RC (or possibly LC) filters on circuit diagrams, we should really see them as balanced filters made from one capacitor and two series impedances, where one is the intentional dropping resistor or smoothing choke and the other is formed by the impedance of the ground circuit, as shown."
He continues with Figure 15.7.
I find this interesting. In my mind what is described in figure 15.7 is similar to the Marshall circuit's use of resistors to separate ground planes.

I wonder if (a small choke) perhaps only a few coils of the ground bus wire may provide enough impedance to quiet the circuit.

Illustrated on Rob's layout... perhaps coiling the ground bus something like this? Or maybe just near the reservoir cap? Or between the "power" and "preamp" nodes?

Last edited: Dec 18, 2020
18. ### James KnoxTele-HolicSilver Supporter

Posts:
925
Joined:
Aug 13, 2003
Location:
Phoenix
Interesting theory LC. Building a nice 5E3 soon?

19. ### LowerleftcoastFriend of Leo's

Posts:
3,205
Joined:
Dec 30, 2019
Location:
california
Hah! I have a couple 5E3 to unload.

Right now I have been repairing some termite damage at the house. The amps will have to wait 'till the painting is done.

I did pick up a 40mA 660CT vintage PT on the bay so I have a 5E1/Ga5/5F2/Rickenbacker mashup on the horizon. 40mA will keep it from being too adventurous. I already have a choke, 8uF filter caps, and a high k primary OT to keep the stress on that 40mA PT down. I doubt I will do it for this project but... I am toying with the idea of an octal preamp. I will have to see what I have in the junk drawer. This one will be point to point.
...Need to find a chassis yet...

IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.