Whats on your amp workbench?

[RectifiedAmps]

Thanks for that RA. Yes, this Jansen 6/20 has 6L6 power valves.

Looking at the 6V6 datasheets, it also specifically states that the grid leak resistor should not exceed 100kOhm for fixed biasing. A very popular vintage amp, let's call it Princeton Reverb, uses 6V6 power valves with 220 kOhm grid leak resistors too. Tremolo circuit very similar and injecting into fixed grid bias circuit. So this design flaw issue wouldn't be limited to Jansen only. (Some would say that Jansen just copied the Fender designs)
I guess these circuits/amps were factory biased relatively cold and very forgiving valves must have been used

This makes me wonder how many 'fixed grid bias wiggle amps' are out there running way too hot

For this amp, I am going to lower the plate voltage a bit first and then try 150 kOhm grid leak resistors. Then set the fixed bias to 50% MPD and hopefully that will still provide a good tremolo function without the latter causing it to go into red plate territory.

6V6s (and EL84s) probably don’t suffer this issue as much since they draw less grid current than a 6L6 (or EL34). I’d be curious to see if Fender used bias wiggle on any of their 6L6 amps and, if they did, how they implemented it… will check some schematics.

Why are you lowering the plate voltage? The max plate voltage is 500v for a 6L6GC so I doubt it’d be contributing to the redplating. It might not be worth the effort.

Forgot about the Tremolux and Vibrolux - both used 6L6GCs with bias wiggle trem. You could adjust your 6/20 to similar specs since I’m pretty sure the Fenders didn’t have redplating issues, despite the overspec grid resistance.

[StrummersOfThunder]

Thanks for that RA. Yes, this Jansen 6/20 has 6L6 power valves.

Looking at the 6V6 datasheets, it also specifically states that the grid leak resistor should not exceed 100kOhm for fixed biasing. A very popular vintage amp, let's call it Princeton Reverb, uses 6V6 power valves with 220 kOhm grid leak resistors too. Tremolo circuit very similar and injecting into fixed grid bias circuit. So this design flaw issue wouldn't be limited to Jansen only. (Some would say that Jansen just copied the Fender designs)
I guess these circuits/amps were factory biased relatively cold and very forgiving valves must have been used

This makes me wonder how many 'fixed grid bias wiggle amps' are out there running way too hot

For this amp, I am going to lower the plate voltage a bit first and then try 150 kOhm grid leak resistors. Then set the fixed bias to 50% MPD and hopefully that will still provide a good tremolo function without the latter causing it to go into red plate territory.

6V6s (and EL84s) probably don’t suffer this issue as much since they draw less grid current than a 6L6 (or EL34). I’d be curious to see if Fender used bias wiggle on any of their 6L6 amps and, if they did, how they implemented it… will check some schematics.

Why are you lowering the plate voltage? The max plate voltage is 500v for a 6L6GC so I doubt it’d be contributing to the redplating. It might not be worth the effort.

Forgot about the Tremolux and Vibrolux - both used 6L6GCs with bias wiggle trem. You could adjust your 6/20 to similar specs since I’m pretty sure the Fenders didn’t have redplating issues, despite the overspec grid resistance.

Just so happens I have both of these amps and yes they both use 6l6 and have bias wiggle trem , lovely trem at that too
Happy to put them on the bench and take measurements

[robthemac]

6V6s (and EL84s) probably don’t suffer this issue as much since they draw less grid current than a 6L6 (or EL34). I’d be curious to see if Fender used bias wiggle on any of their 6L6 amps and, if they did, how they implemented it… will check some schematics.

Why are you lowering the plate voltage? The max plate voltage is 500v for a 6L6GC so I doubt it’d be contributing to the redplating. It might not be worth the effort.

Forgot about the Tremolux and Vibrolux - both used 6L6GCs with bias wiggle trem. You could adjust your 6/20 to similar specs since I’m pretty sure the Fenders didn’t have redplating issues, despite the overspec grid resistance.

Just so happens I have both of these amps and yes they both use 6l6 and have bias wiggle trem , lovely trem at that too
Happy to put them on the bench and take measurements

Gotta catch 'em all!

[Jay]

Thanks for that RA. Yes, this Jansen 6/20 has 6L6 power valves.

Looking at the 6V6 datasheets, it also specifically states that the grid leak resistor should not exceed 100kOhm for fixed biasing. A very popular vintage amp, let's call it Princeton Reverb, uses 6V6 power valves with 220 kOhm grid leak resistors too. Tremolo circuit very similar and injecting into fixed grid bias circuit. So this design flaw issue wouldn't be limited to Jansen only. (Some would say that Jansen just copied the Fender designs)
I guess these circuits/amps were factory biased relatively cold and very forgiving valves must have been used

This makes me wonder how many 'fixed grid bias wiggle amps' are out there running way too hot

For this amp, I am going to lower the plate voltage a bit first and then try 150 kOhm grid leak resistors. Then set the fixed bias to 50% MPD and hopefully that will still provide a good tremolo function without the latter causing it to go into red plate territory.

6V6s (and EL84s) probably don’t suffer this issue as much since they draw less grid current than a 6L6 (or EL34). I’d be curious to see if Fender used bias wiggle on any of their 6L6 amps and, if they did, how they implemented it… will check some schematics.

Why are you lowering the plate voltage? The max plate voltage is 500v for a 6L6GC so I doubt it’d be contributing to the redplating. It might not be worth the effort.

Forgot about the Tremolux and Vibrolux - both used 6L6GCs with bias wiggle trem. You could adjust your 6/20 to similar specs since I’m pretty sure the Fenders didn’t have redplating issues, despite the overspec grid resistance.

Let me check out those circuits and report back.

Lowering the plate voltage would be reducing MPD as well. But first trying lower grid leak resistors.

[Jay]

6V6s (and EL84s) probably don’t suffer this issue as much since they draw less grid current than a 6L6 (or EL34). I’d be curious to see if Fender used bias wiggle on any of their 6L6 amps and, if they did, how they implemented it… will check some schematics.

Why are you lowering the plate voltage? The max plate voltage is 500v for a 6L6GC so I doubt it’d be contributing to the redplating. It might not be worth the effort.

Forgot about the Tremolux and Vibrolux - both used 6L6GCs with bias wiggle trem. You could adjust your 6/20 to similar specs since I’m pretty sure the Fenders didn’t have redplating issues, despite the overspec grid resistance.

Just so happens I have both of these amps and yes they both use 6l6 and have bias wiggle trem , lovely trem at that too
Happy to put them on the bench and take measurements

Might take you up on that. Let me check out the circuits first

[MikeC]

6V6s (and EL84s) probably don’t suffer this issue as much since they draw less grid current than a 6L6 (or EL34). I’d be curious to see if Fender used bias wiggle on any of their 6L6 amps and, if they did, how they implemented it… will check some schematics.

Why are you lowering the plate voltage? The max plate voltage is 500v for a 6L6GC so I doubt it’d be contributing to the redplating. It might not be worth the effort.

Forgot about the Tremolux and Vibrolux - both used 6L6GCs with bias wiggle trem. You could adjust your 6/20 to similar specs since I’m pretty sure the Fenders didn’t have redplating issues, despite the overspec grid resistance.

Let me check out those circuits and report back.

Lowering the plate voltage would be reducing MPD as well. But first trying lower grid leak resistors.

To be pedantic & assist with anyone trying to understand this, MPD (Max Plate Dissipation) remains the same regardless of plate voltage because MPD is a technical specification of the valve in question. BUT... in circuit, if nothing else changes and you reduce the plate voltage, then actual plate dissipation will reduce.

[Jay]

I don't mind you being pedantic Mike. Totally correct you are.

Remember how Molly shared that CD with all the vintage amps? Well, it has a FENDER PDF with 164 models in so I just scrolled through it and summarised the 6L6 valve circuits with tremolo. Thanks Molly, made it easy! I put some of the Bassman in as well for comparison (no trem circuit).

Interestingly, there are only two Fender 6L6 amps that have the tremolo injected into the power valves bias AND use 220K grid leak resistors.

Here my summary. Next step comparing how those two differ from the Jansen 6/20 and 6/30.

But my first immediate observation is that the 6L6 plate voltages on those two are way lower than on the 6/20 (120VDC to be precise).
The other observation is that Fender seemed to know that 100k for grid leak resistors for a 6L6 are spec correct. Why then go 220k?

Edit - updated spreadsheet

Fender 6L6 valve amps.jpg (118.07 KiB) Viewed 563 times

[MikeC]

I don't mind you being pedantic Mike. Totally correct you are.

Remember how Molly shared that CD with all the vintage amps? Well, it has a FENDER PDF with 164 models in so I just scrolled through it and summarised the 6L6 valve circuits with tremolo. Thanks Molly, made it easy! I put some of the Bassman in as well for comparison (no trem circuit).

Interestingly, there are only two Fender 6L6 amps that have the tremolo injected into the power valves bias AND use 220K grid leak resistors.

Here my summary. Next step comparing how those two differ from the Jansen 6/20 and 6/30.

But my first immediate observation is that the 6L6 plate voltages on those two are way lower than on the 6/20 (120VDC to be precise).
The other observation is that Fender seemed to know that 100k for grid leak resistors for a 6L6 are spec correct. Why then go 220k?



Fender 6G6 valve amps.jpg

A mystery to be solved, perhaps by you Jean!

[RectifiedAmps]

But my first immediate observation is that the 6L6 plate voltages on those two are way lower than on the 6/20 (120VDC to be precise).
The other observation is that Fender seemed to know that 100k for grid leak resistors for a 6L6 are spec correct. Why then go 220k?

I love a good spreadsheet! But if you look at a broader range of Fender schematics you'll see that they (nearly) always used 220k on everything - regardless of whether it was 6L6, EL84, or 6V6 and had tremolo or not. The main run of amps that used 100k (or less) were CBS-era ones that had the bias balance control instead of a bias adjust. Otherwise, the Super 60 is the only non-CBS I've seen with 100k resistors - I'm sure there are other oddballs out there though.

My guess is that 220k just 'worked' (or was somehow cheaper) so that's what they stuck with over the decades. I also bet they originally had 'standard' 450v+ plate voltages on the Tremolux and Vibrolux and suffered from redplating so they dropped the voltage in response rather than fixing the design. As you've shown those amps have an unusually low plate voltage which is uncharacteristic for Fender.

[Jay]

But my first immediate observation is that the 6L6 plate voltages on those two are way lower than on the 6/20 (120VDC to be precise).
The other observation is that Fender seemed to know that 100k for grid leak resistors for a 6L6 are spec correct. Why then go 220k?

I love a good spreadsheet! But if you look at a broader range of Fender schematics you'll see that they (nearly) always used 220k on everything - regardless of whether it was 6L6, EL84, or 6V6 and had tremolo or not. The main run of amps that used 100k (or less) were CBS-era ones that had the bias balance control instead of a bias adjust. Otherwise, the Super 60 is the only non-CBS I've seen with 100k resistors - I'm sure there are other oddballs out there though.

My guess is that 220k just 'worked' (or was somehow cheaper) so that's what they stuck with over the decades. I also bet they originally had 'standard' 450v+ plate voltages on the Tremolux and Vibrolux and suffered from redplating so they dropped the voltage in response rather than fixing the design. As you've shown those amps have an unusually low plate voltage which is uncharacteristic for Fender.

Thanks for that background info and I think you would win that bet
I recall Fender's first foray into the Vibrolux EL84 design was aborted pretty quickly due to blown power valves. So I entirely believe they can make design errors that make it into production.

Some progress to report....

The Jansen tremolo circuit is identical to the Tremolux Schematic 6G9-A apart from the depth/intensity pot and the tremolo switch resistor.
The Fender pot is 250K while the Jansen is a 1M pot. The 1M switch resistor goes to the Cathode in the Fender circuit (trem quick start) while it goes to ground in the Jansen. I left the latter as it is.

Mods performed today:
1 - I can see that the depth pot has been changed in the past so not sure if the original value was 500K which some circuits suggest. Anyway, 1M is way out of range so I put a 270K resistor over the middle and bottom lug to make an affective 212K pot. I would have liked to use a 330K but didn't have any.

2 - I took the 220K grid leak resistors out and they measured 330K and 300K respectively. Not good. These were changed with 100K resistors.

3 - While at it, I put in grid stop resistors of 1K5 and grid 2 resistors of 470 Ohm/2W.

Testing revealed the following:
A - At idle bias voltage set to -49V. with trem off and intensity turned to full, the bias voltage drops to -45V. This is a major improvement as previously it would drop to -39V which caused a beautiful red plating display.

B - Tremolo function doesn't appear to be affected by the changes. If anything, I think it is nicer now.

C - Idle current is 29mA for V6 and 36mA for V5. I then swapped the 6L6 valves and the current drawn moves with the valve. So after swapping V6 measured 36mA and V5 29mA. This suggest that one valve might have an issue. So plate dissipation is 15W for one and almost 19W for the other (plate voltages unchanged at 480VDC). So one valve is sitting at maximum allowable dissipation.

D - The valve drawing the higher current still red plates in a small section of the plate. I can probably fix this by lowering the plate voltage but it might just be the valve itself. Clarry doesn't have a valve tester so not sure what I am going to do yet. Maybe put a trimpot in for the two 100K grid leak resistors so I can 'steer' less current to the offending valve. Suggestions welcome.

A couple of photos so you can see it did happen

Red Plate Fixes 20250429_172715.jpg (2.5 MiB) Viewed 544 times

V6 still red plating a bit 20250429_173815.jpg (2.95 MiB) Viewed 544 times

[MikeC]

Hey Jean, I've got an Avo C.T. 160 valve tester if it helps.

[crowbgood1]

Yes, I finished the cab restoration. A mixture of speakers, but sounds great.

Before

Jansen Cab pre.jpg (340.13 KiB) Viewed 88 times

After

Jansen Cab.jpg (519.48 KiB) Viewed 88 times

[AiRdAd]

Looks great Chris - if you want another empty 4x12 cabinet, I've got one

[Bg]

Yes, I finished the cab restoration. A mixture of speakers, but sounds great.

Before
Jansen Cab pre.jpg


After
Jansen Cab.jpg

shit, thats my tiny djw marshall, dibs!

[Bg]

I've got fender grill cloth if you want to get rid of the dog piss stain at some point Otherwise AWESOME