Tweed Super 5E4A tweaking

[SleepLess]

Hi there!
It's been a while. I finished this build two weeks ago, my first 5E4A and there are a few things that bug me: I find the amp to lack in clean sound, it crunches a bit too fast. Second, the output on the amp is at full tilt when the guitar volume is on 10, but if I lower the guitar volume knob to 7 or 6 it's like I'm playing an acoustic, all the sound is gone. Something's wrong there... Even with a lower wattage tweed like a Princeton I still have an amplified sound on 3 on the guitar...

Here are a few of the things that may need to be corrected. The schematic says B+ should be around 380V. I have almost 100V more at 475V... The 6V6s are biased a bit hot at 23mA. 70% of max plate disspation recommends 20mA. I'm not too far off.

The amp is wired for 230V. I have 233V at the wall. I have another wiring option at 240V. Should I use instead it to lower the B+ a little?
Any recommendations are welcome. I have built many amps and am a bit disappointed in this one's sound. It should sound a lot better. It lacks the roundness of tweeds. It's too harsh for a tweed.

There was no room to install the bias pot because of the MM PT which is BIG. I have the three-legged bias pots one can see in Marshall builds but don't know how to install them (anyone?)... So I chose a 10K  resistor to bias the amp instead... Thats the kind of pot I have in stock but don't know how I can solder the three legs...



A few pics, if you need more precise ones, I'll take more: thanks a lot for your help!

[PRR]

The build is beautiful!

The voltages are frightening.

Link to schematic you used.

Link to specs for the power transformer and output transformer you used.

Measure voltage all down the B+ filter chain and at each tube's Plate. In Fendery amps, you expect most plates to sit at around 70% of the B+ node they feed from. (B+ node at 300V, plate near 210V.)

With power off, put a cord in the input jack. Clip your ohm-meter to the cord's tip and shell. Is the resistance 1Meg (correct) or 34K (mis-wired jack)?

[SleepLess]

Hi PRR and thank you!

I followed the Ampwares.com schematic and layout. Plus a 5F4 layout I had from Mojotone to A/B both builds. I am surprised to say that there' no difference between the two builds. I thought that there would be a few caps or resistors changes because of the 6L6/6V6 difference between both amps but there was none...

Here are the ampwares files:

http://ampwares.com/schematics/super_5e4a.pdf

I used the FTS-PM/240 Mercury PT:


All the B+ filter chain is higher than it should be. I do have the 1M resistance at the input's tip and shell...

[HotBluePlates]

Are you sure the PT is an FTS-PM/240?

I see no such transformer on Mercury Magentics' webapge, but I do see a FTS-P/240, which looks as though the secondary is 380-0-380vac and would possibly align with the B+ you're seeing.

Try measuring the a.c. volts across half the secondary to confirm.

... I am surprised to say that there' no difference between the two builds. I thought that there would be a few caps or resistors changes because of the 6L6/6V6 difference between both amps but there was none...

The bias supply is different (6.8kΩ series resistor instead of 18kΩ series resistor) to send more negative voltage to the 6L6's control grids.  Also, the power and output transformers are different (and the resulting B+ is then a bit higher in the 5F4).

There was really no need to change the preamp at all, and the 5E4-A phase inverter had enough output capability to drive the 6L6's as biased.  That means the only change other than bias was to change the load the output tubes worked into, and the PT to supply the additional current at that load for the extra power output.

[HotBluePlates]

... I have the three-legged bias pots one can see in Marshall builds but don't know how to install them (anyone?)... Thats the kind of pot I have in stock but don't know how I can solder the three legs...

I have a 5F4 Super copy, and I added a small multi-turn pot (like the picture below) to the fiberboard to adjust bias.

I put a drop of superglue on the short face of the pot opposite the screwdrive adjustment, and stuck that to the board.  The screwdrive adjust point then pointed "up" to the open side of the chassis.  I then soldered to one of the static lugs and to the wiper to get an adjustable resistance (I don't recall the exact pinout at the moment; you'd need to look at your particular pot, as they aren't always the pins you'd expect).

I think I may have made the connections to the pot itself with the actual leg of the bias supply diode and resistor-to-ground, after lifting that leg out of the eyelet.

[SleepLess]

It is the FTS-PM/240, so says the label on it. How do I measure the ac volts across half the secondary? I may have a wrongly labelled PT, who knows?

I ended up using a 10k resistor for the bias. I have a -47V voltage with it and 23mA. Here is the only thing I changed. Maybe I should have changed something else?
Thanks!

[HotBluePlates]

It is the FTS-PM/240, so says the label on it. How do I measure the ac volts across half the secondary? I may have a wrongly labelled PT, who knows?

Assuming MM used the same colors as the original transformer, your meter leads go on 1 red and 1 red-yellow wire.  Highest range for a.c. volts you have available.

You're mainly looking to see if the PT supplies closer to 300vac or 400vac.  380vac or higher suggests it's the FTS-P/240.  335-340vac would look like it's the FTS-PM with your multi-volt primary.

And yes, if it were my amp (regardless of which transformer), I'd use the 240vac tap in your case.  Mercury winds accurate transformers, which often have accurately-high output voltages with today's wall voltages (the vintage amps have the same issue).

[SleepLess]

Measured this value and I have 367VAC... B+ is 456VDC... I will wire it for 240V and see what that changes.

[sluckey]

Remember your 5F10 build? Look at reply #62 in that thread. That suggestion will work for this amp too.

     http://el34world.com/Forum/index.php?topic=17527.msg176719#msg176719

[SleepLess]

Hi Sluckey! Thanks a lot!
Do you think this will solve the high B+ issue I have as well? I will solder the pot tomorrow!

[sluckey]

Do you think this will solve the high B+ issue I have as well? I will solder the pot tomorrow!

Not likely. Use the PT 240V tap. Put a NOS 5Y3 in the rectifier socket. Get a 300-0-300 PT.

[SleepLess]

Not likely. Use the PT 240V tap. Put a NOS 5Y3 in the rectifier socket. Get a 300-0-300 PT.

I paid more than $200 for this PT which clearly says "for 5E4A dual 6V6 version"... 

[sluckey]

Yeah. How's that working for ya?   

[Willabe]

I paid more than $200 for this PT which clearly says "for 5E4A dual 6V6 version"... 

Call Merc. Mag and tell them what your getting for ACV on the PT high voltage secondary and see what they say.

Seems like they made a mistake/mislabeled that PT, it can happen. 

[HotBluePlates]

Measured this value and I have 367VAC... B+ is 456VDC... I will wire it for 240V and see what that changes.

I think this will help some.  And Sluckey's recommendation to try a (non-Russian) 5Y3GT.

Hopefully that will bring the supply voltage down some.  At that point, increasing the value of the 10kΩ resistor you have now (or adding the bias trimmer) will increase the current draw of the output tube, which should drag the supply voltage down some more.

It's probably safe in any case to idle the output tubes at ~30mA (100% dissipation for 14w at 467vdc).  As you start dropping the bias voltage to inch up to that 30mA, the supply voltage will probably creep downward, enabling you to bias hotter for more current.  You're broadly in the ballpark between 21-36mA, assuming you can get the plate voltage of the output tubes to come down near 385vdc.  Then you can tweak to taste.

[SleepLess]

I changed the 230V tap for the 240V tap and it brought the voltage 10v down. It's now 445V.
This amp doesn't use a 5Y3GT but a 5U4GA. I put a 5Y3GT in there instead and the voltage went up 10V to 455V. I put the 5U4GA back in.

Now I'm going to install the bias pot. It's a 25K. Should the 56K resistor on top be soldered to both legs as per your pic Sluckey or just one? I got that the other leg gets soldered to the - side of the diode.
Thanks!

[HotBluePlates]

... This amp doesn't use a 5Y3GT but a 5U4GA. ...

That doesn't matter.  This amp only uses a pair of 6V6's, and the 5Y3 can easily pass the current required by those tubes.  But a real 5Y3 drops more voltage than a 5U4, which is like a V8 to the 5Y3's 4-banger.

... I put a 5Y3GT in there instead and the voltage went up 10V to 455V. I put the 5U4GA back in. ...

If you used a modern, Russian "5Y3" it's not really a 5Y3 and doesn't drop nearly as much voltage.  Your test appears to show the tube you used wasn't a genuine 5Y3.

... Should the 56K resistor on top be soldered to both legs as per your pic Sluckey or just one? ...

Yes.  You're using the trimmer as a variable resistance, so the wiper gets tied to the outer leg as shown in Sluckey's picture.

However you're still getting more ac volts from the PT secondary than you should, especially since you now have ~236vac applied to the 240vac tap.  If MM had the transformer right, you should have less output voltage at the secondary than even Fender's schematic shows (or at least closer to the 390vdc Fender shows).

Seems like it's time to contact MM or the vendor where you got the PT.

[SleepLess]

OK, thanks. I'll get in touch with MM and let you know what they say. In the meantime I'll install a proper NOS 5Y3 (you're right I used a Sovtek for the test above) and the bias trimpot.

[sluckey]

Do you have the OT wired like the Fender layout, that is, does the OT primary center tap connect to pin 4 of the 6V6s as shown?

About the bias pot... The whole idea was to allow you to adjust the bias hotter which would bring the B+ down a bit more. If you keep that 56K resistor you will be able to adjust cooler but you wont be able to bias any hotter than you are right now. I chose the values of 50K pot and 24K resistor so you could adjust cooler and hotter than you could with just the 56K resistor. If you use your 25K pot and 56K resistor you will still need to juggle the value of the 6.8K to get to a hotter bias. You have 10K now, try a 12K or 15K, or 18K.

When you speak with MM, ask them how they determined the secondary voltage for that PT. I can't find any Fender documents that list the secondary voltages or even a PT part number. That PT is never gonna give you voltages even close to those shown on the original Fender schematic.

[SleepLess]

Do you have the OT wired like the Fender layout, that is, does the OT primary center tap connect to pin 4 of the 6V6s as shown?

Yes I do Sluckey. OT primary CT goes to pin 4 of the 6V6s.

About the bias pot... The whole idea was to allow you to adjust the bias hotter which would bring the B+ down a bit more. If you keep that 56K resistor you will be able to adjust cooler but you wont be able to bias any hotter than you are right now. I chose the values of 50K pot and 24K resistor so you could adjust cooler and hotter than you could with just the 56K resistor. If you use your 25K pot and 56K resistor you will still need to juggle the value of the 6.8K to get to a hotter bias. You have 10K now, try a 12K or 15K, or 18K.

OK, got it. I can change the 56K resistor too if that gives me more range than keeping it and avoids juggling with the 10K resistor. What value would you recommend then for the 56K? 24K?

When you speak with MM, ask them how they determined the secondary voltage for that PT. I can't find any Fender documents that list the secondary voltages or even a PT part number. That PT is never gonna give you voltages even close to those shown on the original Fender schematic.

OK... I'm waiting on their response and will ask that when I get it.

[SleepLess]

Here's their response: oh well...

"You have to keep in mind and you will see clearly stated on all Fender Schematics, “All Voltages Plus or minus 20%” And that is being kind. There was no consistency back then and all of the Fender B+ variations you see us offer are exact clones that came from an original amp as is the case here.
Nothing wrong with the transformer. And disconnected from the amp, the B+ will read 335 0 335 approximately each leg to center tap. Unloaded AC open circuit.
 Thanks."

[HotBluePlates]

"... Nothing wrong with the transformer. And disconnected from the amp, the B+ will read 335 0 335 approximately each leg to center tap. Unloaded AC open circuit."

Except you (hopefully) told them the PT is measuring 367vac loaded.  If they sent that email to me, I'd probably never buy MM again.  But that's just me.

... I ... am a bit disappointed in this one's sound. It should sound a lot better. It lacks the roundness of tweeds. It's too harsh for a tweed. ...

Well, let's get back to this. Make sure you're not plugged into the channel with the bright cap on the volume control (with that channel volume set low), or it will naturally sound harsh.  I cut the bright cap off my Super.

[SleepLess]

Here's what I had written before their answer:

"I'm writing because I finished this tweed Super 5E4A build two weeks ago and am disappointed in the amp. I noted a few voltage and current values and noticed that the FTS-PM/240 PT I have actually gives out more current than it actually should. I measured the a.c. volts across half the secondary and got a 367VAC value with a B+ of 456VDC (with a NOS GE 5U4GB installed, not a crappy Russian rectifier tube). This is way higher than what the Fender schematic recommends (380VDC). Something's wrong with that PT. I think it's the reason why the amp sounds harsh and lacks in clean volume. It doesn't have the roundness tweeds are known for. I may add that I build amps as a professional, I'm not an amateur. I have built dozens of tweeds so far. What do you say?"

Thanks about the bright cap thing. You actually got rid of it, is that it? I got a 100pF there...

[drew]

My 2 cents: MM has told you what the thing should measure disconnected from the amp, so disconnect it from the amp and test it.  Make note of your mains voltage at the time of testing, and test both the 230 and 240 volt taps. 

[HotBluePlates]

My 2 cents: MM has told you what the thing should measure disconnected from the amp, so disconnect it from the amp and test it. ...

He would have an identical result by removing the rectifier tube before measuring secondary voltage.

[sluckey]

When you speak with MM, ask them how they determined the secondary voltage for that PT. I can't find any Fender documents that list the secondary voltages or even a PT part number. That PT is never gonna give you voltages even close to those shown on the original Fender schematic.

I have no doubt that their PT voltages match their data sheet which shows 335-0-335. I am curious as to why they choose 335-0-335 for a 5E4-A. I think a much lower secondary voltage will be required to give the 390v B+ voltage listed on the Fender schematic.

However, I do realize that Fender says on that schematic that voltage readings are +20%, so B+ could be as high as 468V. Based on that, you ain't got a leg to stand on with MM. So, if you want lower voltages, your cheapest option is to use a real NOS 5Y3. I recently read that JJ's 5Y3 have a high voltage drop just like the old 5Y3s. I can't confirm that claim.

[PRR]

> like a V8 to the 5Y3's 4-banger.

Well, I once replaced a 135HP V-8 with a 149HP four-cyl.

(It is a good 4-cyl and was a remarkably lame V-8.)

What he really wants is the transmission from that V-8. Real slush-0-matic rubber-band power-robber. Like a true 5Y3. I submit that a 100 or so Ohms may get his B+ down.

[SleepLess]

Hi.
The bias pot is installed and I used an 18K resistor (which actually measures 16.7K). The bias range goes from 28mA to 42mA. I set it on 30mA. B+ has come down to 435V. I don't have any NOS 5Y3 at hand. I need to buy one...
Thanks!

[ac427v]

I like the 5R4GYB too. I used that rectifier to take my alt tweed bassman from 428 volts with Sovtek 5AR4 to 388 volts with Amperex 5R4GYB. It is rated to handle higher voltage than the 5Y3GT but only uses 2 amps of heater current.

[drew]

There's a thread on ampgarage by xtian about a nifty little MOSFET B+ reducer. The board is only about as wide as, and a little longer than, a guitar pick.  Admittedly it would look about as out of place in this amp as a nose ring on Marilyn Monroe, but it's something to consider if a different rectifier tube still doesn't get you where you want to be.

[HotBluePlates]

There's a thread on ampgarage ... about a nifty little MOSFET B+ reducer. ...

If one resorted to artificial B+ reduction, PRR's resistor is as good as any method.

... I submit that a 100 or so Ohms may get his B+ down.

[sluckey]

There's nothing wrong with running at 435V. Remember, those Fender voltages are ±20%.

[SleepLess]

There's nothing wrong with running at 435V. Remember, those Fender voltages are ±20%.

I ordered the NOS 5Y3GT on Saturday, I'll wait until it gets here and try it in and see what happens. If it doesn't sound any better with a lower voltage, I'll put the NOS 5U4GA back in.
Thanks everyone!

[HotBluePlates]

There's nothing wrong with running at 435V. Remember, those Fender voltages are ±20%.

Agreed, but I don't like that he's getting 367vac when Mercury's own paperwork says he should have 335vac.  And he's using the 240v tap while applying closer to 230vac.

Unless this PT has B+ current capacity to support 6L6's (and it might, seeing that the rated 6.3v @ 7A is ridiculous for an amp that only needs 2A or less...)

[SleepLess]

Hi everyone.
I received the NOS 5Y3GT yesterday and tried it in this morning. Here's a value sheet of the amp. If you spot anything strange, please say so!
Thanks again! After writing down all the values I raised the bias up to 27mA and B+ went down to 425V.

[HotBluePlates]

Seems close enough for R&R.

Too bad it's not a 6L6 Super (5F4), as everything would be spot-on.