London Power ad

[-]
Kits & Books

[-]
Search the Forum








(Advanced Search)

Grid leak biasing
#1
Hello, 

I was wondering about grid-leak biased input stages we sometimes see in older designs. What are the con's and pro's (if any) of such arrangement? I have this Supro Coronado schematic and channel one tube is biased like this: https://el34world.com/charts/Schematics/..._1690t.pdf

First of all it looks as if the channel was maybe designed for a microphone originally , but apparently the amp was not marketed for it. On the other hand a rather high level of gain is called for here and this way of biasing the tube saves the cost of  one resistor and one electrolytic cathode bypass cap which otherwise would have had to be put in there. 

I thought it was built this way to introduce some grid distortion as the whole amp seems intended for distorted sound and output signals from either channel easily drive the splitter grid to conductivity. I scoped the input stage expecting to see half wave clipping but there is none. I used a 1kHz 128mV (peak to peak) sine and got a clean output wave 50 times the input signal (6,4V peak to peak). With the 6,8M grid leak the bias voltage is -650mV with the plate sitting at 77V. I had to increase the signal voltage to almost 2 volts to clip the grid. 
Looks like the triode could've been easily biased using a cathode resistor with a bypass cap and do exactly the same job. 

Was there any particular reason why the first input stage was built this way that I'm unable to see. I'm aware this is most likely another case of bad amp design, I'm just interested in hearing some background about it. 

Tomislaw
Reply
#2
(11-05-2018, 04:06 PM)Tomislaw Wrote: Hello, 

I was wondering about grid-leak biased input stages we sometimes see in older designs. What are the con's and pro's (if any) of such arrangement? I have this Supro Coronado schematic and channel one tube is biased like this: https://el34world.com/charts/Schematics/..._1690t.pdf

First of all it looks as if the channel was maybe designed for a microphone originally , but apparently the amp was not marketed for it. On the other hand a rather high level of gain is called for here and this way of biasing the tube saves the cost of  one resistor and one electrolytic cathode bypass cap which otherwise would have had to be put in there. 

I thought it was built this way to introduce some grid distortion as the whole amp seems intended for distorted sound and output signals from either channel easily drive the splitter grid to conductivity. I scoped the input stage expecting to see half wave clipping but there is none. I used a 1kHz 128mV (peak to peak) sine and got a clean output wave 50 times the input signal (6,4V peak to peak). With the 6,8M grid leak the bias voltage is -650mV with the plate sitting at 77V. I had to increase the signal voltage to almost 2 volts to clip the grid. 
Looks like the triode could've been easily biased using a cathode resistor with a bypass cap and do exactly the same job. 

Was there any particular reason why the first input stage was built this way that I'm unable to see. I'm aware this is most likely another case of bad amp design, I'm just interested in hearing some background about it. 

Tomislaw

Grid-leak bias was used on lots of early amps and it can sound really good for certain kinds of playing.  If you want gritty 50's tone it's great. Particularly for players who don't use pedals for gain sounds.    

That said it has some major limitations.  Grid leak-bais was developed at time when input signals where not expected to be high so the amps with were not designed for distortion.   You'll see grid-leak bias all the time in 40's and early 50's Hi-fi amps. Most early instrument amps are simply adaptations of those same designs.    These amps were developed prior humbuckers, hot pickups and fx pedals so they were far more acceptable sounding with low output single coils.  Grid-leak biased stages will distort easily and cannot handle large input signals because so don't hookup a Fuzzface a grid-leak bias amp.  Since you cannot voice the cathode with a bypass cap the voicing is limited to the input and output caps which are typically big values reflecting their hi-fi roots.

One thing further. The Supro you reference has number of other design choices that limit the headroom of the amp. Whether you like it is all up to your ears but the mix of relatively low plate voltages, cathode bias, a see-saw inverter and light magnet speakers all contribute to the sound.
Reply
#3
(11-05-2018, 04:57 PM)makinrose Wrote: Grid-leak bias was used on lots of early amps and it can sound really good for certain kinds of playing.  If you want gritty 50's tone it's great. Particularly for players who don't use pedals for gain sounds.    

That said it has some major limitations.  Grid leak-bais was developed at time when input signals where not expected to be high so the amps with were not designed for distortion.   You'll see grid-leak bias all the time in 40's and early 50's Hi-fi amps. Most early instrument amps are simply adaptations of those same designs.    These amps were developed prior humbuckers, hot pickups and fx pedals so they were far more acceptable sounding with low output single coils.  Grid-leak biased stages will distort easily and cannot handle large input signals because so don't hookup a Fuzzface a grid-leak bias amp.  Since you cannot voice the cathode with a bypass cap the voicing is limited to the input and output caps which are typically big values reflecting their hi-fi roots.

One thing further.  The Supro you reference has number of other design choices that limit the headroom of the amp.  Whether you like it is all up to your ears but the mix of relatively low plate voltages, cathode bias, a see-saw inverter and light magnet speakers all contribute to the sound.

Thank you, makinrose. 

I'm experimenting with Supros as I have a specific sound in mind and some of those old Valco designs suit my taste quite well. I'm all up for a snarly, aggresive tone with limited headroom. 

I have built copies of two Supro models the Coronado 1690T and the 6424T and have modified them in many ways ever since. On the 6424 I got rid of the second channel, converting the spare triode into a tapped gain stage. I'm under impression that both models work better with humbuckers as the input impedance is very low and eats up a lot of signal from single coils. I actually used a typical 68K/1M input on the Coronado where I also added another preamp tube. The original tremolo circuit had the B+ voltage on the intensity pot but only used one triode. I had a bit of a problem getting it to work properly and decided to replace it with the complete tremolo ciruit on two triodes from the 6424. 

The main difference between those two amps if obviously the power amp - the 6424 uses two 6973's whereas the Coronado runs on 6L6's. I went for Heyboer transformers on the 6424, but decided to use the easily available Hammonds for the Coronado. The ones offered by the Classic Tone are quite pricey and they're dedicated for other Supro models, so I didn't want to make an expensive mistake. 

Unlike the Model 24, the Coronado has two gain stages in the tremolo channel and it's quite a hot rodded piece of an amp. I had an unused triode in there too so I stacked it with the first channel stage and added a three way switch to choose A/B or both channels. This way I can have 4 gain stages voiced in different ways, but on the other hand the amp has deviated quite a bit from the original circuit. I'm actually using it for trying out different ideas and experiments with methods of voicing, having temporarily put aside the main goal of approximating the peculiar tone I had been looking for. Originally I was a bit disappointed with the amount of gain I was getting from the original circuits, although the overall tone was promising. Kevin pointed out to me that there are quite a few aspects of the Supro tone and those can be achieved with different methods, which encouraged me for a bit of experimentation on my own. 

Tomislaw
Reply
#4
Hello Citizens!

As they used to say, "All roads lead to Cheops".

Al'likaa' (Goodbye)
Reply
#5
(11-06-2018, 06:50 PM)Tomislaw Wrote:
(11-05-2018, 04:57 PM)makinrose Wrote: Sure thing!!  Those old Supros can be really cool. The ones you build sound like really neat versions.  It's smart to experiment like you did because some of the Supro this sound and work great and other need some help.  One thing I've always found found with Supros it that they are really really speaker specific so I found that using lighter speaker helps get that classic Supro tone.   I'd love see pictures sometime!  

Grid-leak bias was used on lots of early amps and it can sound really good for certain kinds of playing.  If you want gritty 50's tone it's great. Particularly for players who don't use pedals for gain sounds.    

That said it has some major limitations.  Grid leak-bais was developed at time when input signals where not expected to be high so the amps with were not designed for distortion.   You'll see grid-leak bias all the time in 40's and early 50's Hi-fi amps. Most early instrument amps are simply adaptations of those same designs.    These amps were developed prior humbuckers, hot pickups and fx pedals so they were far more acceptable sounding with low output single coils.  Grid-leak biased stages will distort easily and cannot handle large input signals because so don't hookup a Fuzzface a grid-leak bias amp.  Since you cannot voice the cathode with a bypass cap the voicing is limited to the input and output caps which are typically big values reflecting their hi-fi roots.

One thing further.  The Supro you reference has number of other design choices that limit the headroom of the amp.  Whether you like it is all up to your ears but the mix of relatively low plate voltages, cathode bias, a see-saw inverter and light magnet speakers all contribute to the sound.

Thank you, makinrose. 

I'm experimenting with Supros as I have a specific sound in mind and some of those old Valco designs suit my taste quite well. I'm all up for a snarly, aggresive tone with limited headroom. 

I have built copies of two Supro models the Coronado 1690T and the 6424T and have modified them in many ways ever since. On the 6424 I got rid of the second channel, converting the spare triode into a tapped gain stage. I'm under impression that both models work better with humbuckers as the input impedance is very low and eats up a lot of signal from single coils. I actually used a typical 68K/1M input on the Coronado where I also added another preamp tube. The original tremolo circuit had the B+ voltage on the intensity pot but only used one triode. I had a bit of a problem getting it to work properly and decided to replace it with the complete tremolo ciruit on two triodes from the 6424. 

The main difference between those two amps if obviously the power amp - the 6424 uses two 6973's whereas the Coronado runs on 6L6's. I went for Heyboer transformers on the 6424, but decided to use the easily available Hammonds for the Coronado. The ones offered by the Classic Tone are quite pricey and they're dedicated for other Supro models, so I didn't want to make an expensive mistake. 

Unlike the Model 24, the Coronado has two gain stages in the tremolo channel and it's quite a hot rodded piece of an amp. I had an unused triode in there too so I stacked it with the first channel stage and added a three way switch to choose A/B or both channels. This way I can have 4 gain stages voiced in different ways, but on the other hand the amp has deviated quite a bit from the original circuit. I'm actually using it for trying out different ideas and experiments with methods of voicing, having temporarily put aside the main goal of approximating the peculiar tone I had been looking for. Originally I was a bit disappointed with the amount of gain I was getting from the original circuits, although the overall tone was promising. Kevin pointed out to me that there are quite a few aspects of the Supro tone and those can be achieved with different methods, which encouraged me for a bit of experimentation on my own. 

Tomislaw
Reply
#6
(11-06-2018, 10:55 PM)King TUT Wrote: Hello Citizens!

As they used to say, "All roads lead to Cheops".

Al'likaa' (Goodbye)

Some say "All leads lead to a Marshall" Wink
Reply
#7
(11-07-2018, 01:53 AM)makinrose Wrote: Sure thing!!  Those old Supros can be really cool. The ones you build sound like really neat versions.  It's smart to experiment like you did because some of the Supro this sound and work great and other need some help.  One thing I've always found found with Supros it that they are really really speaker specific so I found that using lighter speaker helps get that classic Supro tone.   I'd love see pictures sometime!  

I will post some pictures later. 

I'm very curious how the new Supro's are built, but there are no schematics available I'm afraid. I watched some demos on the YT and judging by the fat tone I'd guess they are way different from the original circuits and use at least two gain stages per channel. The Black Magick combo has 4 preamp tubes but The Coronado advertised as a reissue has 4 12AX7's as well (just like my experimental build). 

Still to be determined is what speaker configuration would give the best "Supro" sound, but I've not had the chance to crank it up on my 1x12 cabinet. I added master volume pots to both amps and also hooked them up to the attenuator. More about it later Smile
Reply
#8
(11-07-2018, 05:39 AM)Tomislaw Wrote:
(11-06-2018, 10:55 PM)King TUT Wrote: Hello Citizens!

As they used to say, "All roads lead to Cheops".

Al'likaa' (Goodbye)

Some say "All leads lead to a Marshall" Wink

I imagine those would be people who are already deaf from playing Marshalls or who have never actually played one?
Or in this modern era, are so conditioned by ipod sound that Marshall sounds acceptable?

Waaaay back, a friend and I rented a Marshall head and cabinet to try out. We opened it up and traced the circuit and played it with different guitars, basses and effects. It sounded really dreadful. So much IM and harshness! I can't say I've ever heard a stock Marshall that sounded good.

Regarding the Supro: Those old amps like Supro and Gibson et al often had multiple inputs for different instruments, microphones, accordians, phono players with crystal cartridges, intending for the amp to be a portable PA system for the whole band. So, lots of tubes. Then add tremolo or reverb and the tube count rises. 6973 you can deal with more or less like a 6BQ5-EL84 as far as OTs go. Speakers were often quite generic low-watt "full-range" types.

Remember that the old speakers are well broken-in by now and a new driver will sound comparatively hard and lifeless.

Have fun
Reply
#9
(11-07-2018, 01:53 AM)makinrose Wrote: I'd love see pictures sometime!  
OK, so here they go: 
6424T:
[Image: 20181107212950_5be3531e6db3d.jpg]
[Image: 20181107213035_5be3531e6df4b.jpg]


And The Coronado 1690T (formerly) Wink 

[Image: 20181107213109_5be3531e6e28a.jpg]
[Image: 20181107213206_5be3531e6e5de.jpg]
Reply
#10
(11-08-2018, 12:51 AM)makinrose Wrote:
(11-07-2018, 03:10 PM)K O\Connor Wrote:
(11-07-2018, 05:39 AM)Tomislaw Wrote:
(11-06-2018, 10:55 PM)King TUT Wrote: Hello Citizens!

As they used to say, "All roads lead to Cheops".

Al'likaa' (Goodbye)

Some say "All leads lead to a Marshall" Wink

I imagine those would be people who are already deaf from playing Marshalls or who have never actually played one?
Or in this modern era, are so conditioned by ipod sound that Marshall sounds acceptable?

Waaaay back, a friend and I rented a Marshall head and cabinet to try out. We opened it up and traced the circuit and played it with different guitars, basses and effects. It sounded really dreadful. So much IM and harshness! I can't say I've ever heard a stock Marshall that sounded good.

Regarding the Supro: Those old amps like Supro and Gibson et al often had multiple inputs for different instruments, microphones, accordians, phono players with crystal cartridges, intending for the amp to be a portable PA system for the whole band. So, lots of tubes. Then add tremolo or reverb and the tube count rises. 6973 you can deal with more or less like a 6BQ5-EL84 as far as OTs go. Speakers were often quite generic low-watt "full-range" types.

Remember that the old speakers are well broken-in by now and a new driver will sound comparatively hard and lifeless.

Have fun

First off nice work!! I'm sure they sound great.

I also wanted to reply to Kevin's comments on Marshalls. I'm not sure what models he is referring to (maybe all of them?) the comments seem strident but only because they fly in face of what players are told. I think players are so nuts about them because their heroes played them and magazines tell them they sound good.  Most modern Marshalls really sound terrible and I dread having to work on them because they  are so poorly built.  They don't really sound like the older Marshall amps either. 

The last Marshall amps I liked were the JCM 800s and they really require some mods to keep them from sounding harsh and  play at reasonable volumes. The non-master Plexi amps can sound fantastic (just my subjective taste) for certain things  but they are so damn loud no one can play them safely without some sort of way to control the volume (power scaling is the best way to go!). 

When building vintage Marshall type amps for folks what I'm really doing is building them an amp that sounds like the Marshall sound they have in their head rather than a amp the is cumbersome to use and ear splitting in person.
Reply
#11
Hi Guys

You know what they say, "Go big or go home".

Plexis don't have a balanced tone except when cranked since that is what they are optimised for. Play them quietly and they are thin and way too bright. You could probably kill people if you played a Tele through them Smile

The first 800s were okay, since they could get the cranked plexi tone at sane volumes, but still, both have the Marshall sheen of IM harshness.

Everything after that seemed to get worse in most respects, especially the VM (Vintage-Modern) with their horrible KT-66s. A friend was modifying a Marshall combo and the customer wanted KT-66s. He did the mod and tried it with the stock EL-34s. It sounded like a Marshall. Then the KTs go in and sounded like mud. It might have sounded best with one of each kind of tube, but there were no individual bias controls. The reversion to KT-66s was probably a mistake for Marshall, but it fits in with the modern fashion of muddy, compressed, tone that players seem to want at the moment, so maybe it's good for Marshall?

The only Marshalls that ever sounded good to me were modified and I think it's safe to say that Marshall is the most modified brand of amp out there.
Reply
#12
(11-07-2018, 03:10 PM)K O\Connor Wrote: Regarding the Supro: Those old amps like Supro and Gibson et al often had multiple inputs for different instruments, microphones, accordians, phono players with crystal cartridges, intending for the amp to be a portable PA system for the whole band. So, lots of tubes. Then add tremolo or reverb and the tube count rises. 6973 you can deal with more or less like a 6BQ5-EL84 as far as OTs go. Speakers were often quite generic low-watt "full-range" types.

I was referring to The Coronado which originally had 3 12AX7's - channel one had just one triode, channel two used two triodes and tremolo was another. The new version is part of the 1964 Reissue Series but the circuit was changed as 4 preamp tubes are listed in the specs.  According to the info, plugging into channel one links it in parallel with channel two, so they probably use another tube to keep the channel one signal in phase with the other one. This not true to the original, despite their producer's claims the amp is a faithful re-creation of the old Coronado. There are also single jacks for each channel which suggest a significant but probably inevitable change in the input voicing compared with the original circuit.

I too came up with a similar topology for the one I have built, adding to channel one a switchable gain stage with an attenuator in front of it, but decided to use just one input and add a three way channel A/A+B/B selector. I really like the dirty tone I'm getting out of it and it's still far from the hard distortion even though each stage shows gain of ~50.
Reply


Forum Jump:

[-]
Come in where it's warm!
A warm welcome to those interested in tube and hi-fi audio! Fans of Kevin O'Connor's The Ultimate Tone (TUT) book series form a part of our population. Kevin O'Connor is the creator of the popular Power Scaling methodology for amplifiers.
Please remember these three principles: respect, sharing, community.
Not familiar with The Ultimate Tone book series? See discussion topics, or click here to visit London Power/Power Press Publishing.

[-]
Forum Hosted by London Power
London Power logo