Welcome to Solid State Guitar Amp Forum | DIY Guitar Amplifiers. Please login or sign up.

March 28, 2024, 05:09:53 AM

Login with username, password and session length

Recent Posts

 

Lab Series L5 preamp in a box: power handling & biasing

Started by kvandekrol, December 20, 2015, 10:19:00 PM

Previous topic - Next topic

kvandekrol

I am knee deep in a project to put the Lab Series L5 preamp in a 1790NS enclosure (EHX "XO" series, or around the size of a Fulldrive)

It'll be pretty well identical to the original, except it'll be channel-switchable, with the initial gain stage shared between both channels and a footswitch to switch it from there. I am also replacing the CA3094 with a CA3080 + discrete Darlington per the original schematic in the 3904 datasheet.

Anyway: I have a couple of questions regarding power handling and biasing.

First: the supply. I don't think there is any good reason to stick with the +/-15V from the original, when compared with the difficulty of making such a supply (not to mention the danger of "rolling your own" from AC power). I am planning on using a LT1054 voltage inverter that should work with either 9V or 12V.

The LT1054 can handle up to 100mA. So, I crunched the numbers on the L5 and came up with the following power requirements:

CA3080: 1.2mA each (x2) = 2.5mA
LM741: 3mA each max (datasheet says 1.7-2.8mA) (x2) = 6mA
JRC4558D: 6mA each (3.5 - 5.7 datasheet) x12 = 72mA
LF356N: 10mA each (5-10 datasheet) x1 = 10mA (max)

Total: 90.5mA if everything was at the very top of its specified range. This is also based on their power requirements assuming +/-15v. The draw should be around 20% less at +/-12V and 40% less at +/-9V, so the full circuit should be significantly less than the LT1054's capacity.

So my first question is: Should this theoretically work OK? Or have I made a mistake in my calculations someplace?

Second question is much easier. The trimming/biasing/tuning methods for the L5 are well documented in their service manuals, and the schematic itself is loaded with test voltages at major junctions and test points. If I am using a supply lower than the standard +/-15V, should I aim for the voltages and test results they specify (since we are nowhere near the rails no matter if it's running at +/-9v, 12v or 15v) - or should I plan on everything being proportionally smaller?

J M Fahey

Please post the L5 schematic here as a refresher.

And try to get the +/- 15V , you are already adding many variables by chopping it and putting it inside a pedal, avoid at least a few of the unknowns which will pop up.

The idea is to happily build and enjoy it  :cheesy:  not lose hair and sleep chasing problems ... well, not that many  ;)

Even if it's meant to fit in a pedaalboard, it's easy to get roughly +/- 15V from a simple 12VAC wall wart, a few pedals work that way (think MB V-Twin) so it's not an unknown .

You might use +/-12V, no big deal, but I wouldn't go lower.

Enzo

I find it odd that you are building a Lab Series amplifier circuit, yet you feel a simple +/-15v supply is "difficult".  certainly the amp circuits are a lot more complex and difficult to build than the power supply.

On that topic, if you need to plan for 90ma, using a 100ma part is really cutting it close.  A 78M15/79M15 is rated for 500ma, plenty of headroom there.

kvandekrol

Schematic:
http://www.rru.com/~meo/Guitar/Amps/Lab/Images/Lab_Series_L5_L7_L9_L11_scheme.jpeg

The supply itself would be pretty easy to implement, something like this:

http://www.musicfromouterspace.com/index.php?MAINTAB=SYNTHDIY&VPW=1910&VPH=813

...with a 15vAC adapter and regulators for +15 and -15. But the difficult part would be making space for it in the box, with either the 3300uF caps (MfOS project) or 1000uF like the V-Twin. Much more attractive to use an IC and a couple of caps if at all possible. Especially since at 12v bipolar it should only draw an absolute max of 70mA...

phatt

If the MFOS supply takes up too much real estate then this will likely take up less space and deliver enough power. 10 VAC in will become  +/-25~28VDC at regulator inputs.

I'm not sure about the current ability of this idea (better minds may wish to verify) but it is often used in many of the Alesis micro series.
My Alesis Micro parametric EQ runs from a 1300mA 9VAC plug pak.
Phil.

phatt

Just a bit sideways but there are other options which will simplify the compression circuit.
I know the L5 compressor might seem pretty slick and I've never played through the L5 but having built a lot of compressor circuits I'd hazza bet it may suffer from the dreaded pump effect which is why I've always disliked compressor circuits.

After some really wacky experiments (some I've posted here) I did eventually stumble upon this dead simple circuit and having now built this into a small pedal unit I'm blown away at how well this works.
Finally a compressor that actually distorts as well as compress,, like a real valve amp. Thank you Mic <3) <3) <3)

Original posting here;
Really Cheap Compressor, by mictester from FSB.
http://www.freestompboxes.org/viewtopic.php?f=13&t=8581

Now if you want pure studio super clean compression DON'T bother,, but if you are wanting something that comes remarkably close to soft clipped Valve power stage compression then this would be very hard to beat.  8)

Nearly all the popular (non optical) comp pedals are way too clean and suffer from the pumping effect which is unnatural. They also introduce a fair share of circuit noise. Some later units use the highly tricked up THAT chip (Boss CS3) which is way less noisy but still do not add the missing rattle.
BTW *Mictester* worked in the factory's that made some of the early pedals (think Maxon Tube screamer days) so a fair bet he knows what works for guitar players. 8|
Phil.

kvandekrol

Quote from: phatt on December 21, 2015, 08:18:01 PM
If the MFOS supply takes up too much real estate then this will likely take up less space and deliver enough power. 10 VAC in will become  +/-25~28VDC at regulator inputs.

I'm not sure about the current ability of this idea (better minds may wish to verify) but it is often used in many of the Alesis micro series.
My Alesis Micro parametric EQ runs from a 1300mA 9VAC plug pak.
Phil.

I like that one a lot. I am guessing that for 15v, I'd need to go up to 12VAC wall wart and then use LDO regulators (LM2940 / 2990) since 12V * 1.414 is just a hair under 17V and the 7815/7915's need at least 2V of wiggle room.

Are the 220uF capacitors enough for a ~100mA current draw? Most of the schematics I've seen for these type of supplies call for 1000uFs' or 2200uF's, but it would be fantastic to be able to get away with this.

phatt

No no no,,, that is a voltage doubler circuit which means with a 10 VAC input the DC voltage at the regulator will be 28VDC.

This kind of PSU maths is not x 1.4,,, it's 2.8 times the ACV.
So keep the ACV input below 12VAC other wise the raw DCV will get too big and may over dissipate the regulators. 8|
Phil.

Adding; Oh caps,,anywhere around that value ,, 330u,, 470u,, as long as they are rated above the raw DCV. I've built this supply and it's dead quite I doubt the need for larger values.

kvandekrol

#8
So, that Alesis doubler schematic led me down a three-day rabbit hole of studying these bipolar PSU schematics, reading through dozens of forum threads discussing it, and I ended up drawing this out as a candidate: (click to see the full thing)



Couple of notes.

1) The parallel caps primarily are for space concerns... the enclosure won't have enough height for bigger caps. For the 100uF booster caps, it also helps share the current load so they don't get too hot - but hopefully using low-ESR caps here will also help.

2) The series diodes after the regulator are for stability - I saw a few references to the 79XX and 79XX's latching up under certain conditions and this supposedly fixes that. I don't know for sure that they are necessary, but the original L5 actually runs on +/-14.4V, so the 0.6v diode drop gets it spot-on with the real thing.

3) The diodes to ground after the regulators are for protection - I saw them in a couple of similar PSU's and in some of the application notes for 78/79's so I am assuming they are good practice, or at the very least won't hurt anything.

Everything look good? This is my first time venturing this far from a 1:1 copy of an existing supply, and while I think I understand what everything does, I could really use a second set of experienced eyes.

We're only looking at a maximum of 200mA current needed... around 100 for all the ICs, and maybe 20mA per LED (there will be three of them). So I am hoping that this is plenty of filtering... I can't spare too much more space on the board!

phatt

I have a hunch you might be over thinking this supply.
As I'm not sure of the correct equations to use for caps I can only go on what has already been built.
Seems your current draw is only going to be 100~200mA max which is in the range of units I've worked on.

I would just breadboard the first part (up to regulators) and read the output voltage.
Read Raw DCV with no load, see how fast the Voltage drops as you add a load.
You can use another meter to read output current at the same time.
Increasing Series Cap values will enable more current but AC transformer would limit that.

Keep in mind the the DC current is effectively half Transformer AC current capacity.
(you double the voltage but current is halved)
You will soon get a real world idea of current ability if something gets hot.

Having built many projects from scratch I have learned to test everything on the bench *Before* you commit to a build.
My Micro EQ has 4 LM347n chips and likely draw less than 100mA all up but AFAIK ALL micro units use the same supply including the Microverb unit.

Links;
http://www.play-hookey.com/ac_theory/power_supply/ps_v_multipliers.html

The complete schematic for Micro verb here;
http://bmamps.com/Schematics/Alesis/alesis%20microverb%204_1.pdf
Very last page has the last updated circuit,I note some extra protection on output. :-\
Regards the extra diode protection; AFAIK, If you read the fine print on data sheets for lower voltage regulation it is not needed and obviously pcb real estate is already a bit tight so you may wish to research that a bit more.
Phil.

teemuk

QuoteNow if you want pure studio super clean compression DON'T bother,, but if you are wanting something that comes remarkably close to soft clipped Valve power stage compression then this would be very hard to beat.

There's already the A110 chip and involved circuitry, for "tube like" clipping. It employs "soft clipping" characteristics of the CA3094 OTA and drives Iabc terminal with input signal to achieve dynamically varying degree of asymmetric clipping (and related effects to harmonic distortion).

Compressor part of L5 is actually incredibly simpe. There's a JFET source follower working as a "unity gain" amplifier and a CA3080E OTA gain stage essentially working as its variable "feedback loop". In order to properly feed one of OTA inputs the signal must be heavily attenuated (input range is few millivolts). Varying gain of this stage also varies the gain of the entire follower circuit. The stage can be considered as plain "open loop" amp (with lowish gain), but gain can also be controlled by applying signal to Iabc terminal.

Signal from JFET source advances to three places: aforementioned OTA input, power amp input and circuit that generates Iabc control signal for the OTA.
 
So other circuitry is there pretty much just to generate that control signal: First there's a plain resistive attenuator for the "input" signal, then signal is amplified. Two diodes and an inverting amp full wave rectify the signal, C139 and R194 make up the RC circuit for "filtering" the rectified signal. This circuit primarily establishes attack, sustain and decay characteristics of the compression. Very "crude" circuit for anything remarkable on that regard, BTW. The result signal drives transistor Q102 and it's collector circuit in turn drives both the OTA Iabc input and another transistor, which simply drives a LED indicator lamp.

If an amp has an OTA-based compressor then the circuit most likely follows pretty much this architecture. Follower, OTA, rectifier + RC filter, BJT drive for OTA. What you have in L5, IMO, is about the most minimalistic example of an OTA based compressor circuit. Really.

kvandekrol

Quote from: phatt on December 25, 2015, 01:12:21 AM
I have a hunch you might be over thinking this supply.

Wouldn't be the first time :) (or the fifth, or the tenth...) I tend to overengineer when I don't understand things as well. Then it gradually gets more sensible the longer I spend on it.

Quote from: phatt on December 25, 2015, 01:12:21 AM
I would just breadboard the first part (up to regulators) and read the output voltage.
Read Raw DCV with no load, see how fast the Voltage drops as you add a load.

Will it regulate without a load? I saw a reference (here) to putting a 2.4k load resistor to ground after each regulator to get it to start regulating, so you could test out the supply before hooking it up to something live. Is this good practice? It would provide an extremely minimal load from what I understand, just enough to test, so it should have no real impact when the circuit is live.

Quote from: phatt on December 25, 2015, 01:12:21 AM
Regards the extra diode protection; AFAIK, If you read the fine print on data sheets for lower voltage regulation it is not needed and obviously pcb real estate is already a bit tight so you may wish to research that a bit more.

Yeah, the two sets of diodes after the regulator are actually both doing the same thing with regard to polarity protection, which I realized after thinking through it a bit more. I took out the ones that go to ground and left the ones in series so I could get the aforementioned voltage drop. I don't think I am at risk of having reverse-voltage damage to the regulators from the output caps discharging, but I suppose it doesn't hurt.

Thanks for your help. This has been incredibly valuable!

phatt

Sorry I meant just test the *Rectifier* section to see how much it sags under load but you may as well just breadboard the whole supply with regulators and start with 2k loads.
Once it's up and running measure the voltages at output of both regulators.
Should be very close to stated output.

Now also read voltage at reg inputs, this is the *unregulated rectified supply* entering regulator inputs,
Write those readings down. 8|

Now implement a very simple load test,, swap the 2k load for 1k load and take another reading.
This will demand more current from the supply and hence your unregulated DC Voltage will drop a bit.

All that matters here is to keep the regulator input DCV at least 3volts above required output Voltage so even a sag down to 20Volts is plenty of headroom to keep regulator working correctly.
If your AC transformer can output say around 1Amp then I very much doubt anything will sag enough to cause a problem.

The reg units might get warm and the load resistors might warm up a bit but leave it on for a while and as long as nothing gets burn da finger hot then it's likely all good to go.
If you want to know for certain just how much current you are passing then do the current test I mentioned before. :tu:

Regards exact voltage for the L5 circuit to work correctly, 14VDC or 15VDC will make no difference to performance.
Phil.

phatt

Quote from: teemuk on December 25, 2015, 04:28:11 AM
QuoteNow if you want pure studio super clean compression DON'T bother,, but if you are wanting something that comes remarkably close to soft clipped Valve power stage compression then this would be very hard to beat.

There's already the A110 chip and involved circuitry, for "tube like" clipping. It employs "soft clipping" characteristics of the CA3094 OTA and drives Iabc terminal with input signal to achieve dynamically varying degree of asymmetric clipping (and related effects to harmonic distortion).

Compressor part of L5 is actually incredibly simpe. There's a JFET source follower working as a "unity gain" amplifier and a CA3080E OTA gain stage essentially working as its variable "feedback loop". In order to properly feed one of OTA inputs the signal must be heavily attenuated (input range is few millivolts). Varying gain of this stage also varies the gain of the entire follower circuit. The stage can be considered as plain "open loop" amp (with lowish gain), but gain can also be controlled by applying signal to Iabc terminal.

Signal from JFET source advances to three places: aforementioned OTA input, power amp input and circuit that generates Iabc control signal for the OTA.
 
So other circuitry is there pretty much just to generate that control signal: First there's a plain resistive attenuator for the "input" signal, then signal is amplified. Two diodes and an inverting amp full wave rectify the signal, C139 and R194 make up the RC circuit for "filtering" the rectified signal. This circuit primarily establishes attack, sustain and decay characteristics of the compression. Very "crude" circuit for anything remarkable on that regard, BTW. The result signal drives transistor Q102 and it's collector circuit in turn drives both the OTA Iabc input and another transistor, which simply drives a LED indicator lamp.

If an amp has an OTA-based compressor then the circuit most likely follows pretty much this architecture. Follower, OTA, rectifier + RC filter, BJT drive for OTA. What you have in L5, IMO, is about the most minimalistic example of an OTA based compressor circuit. Really.

Hi Teemu, thanks for the explanation it all helps, :dbtu: Sadly I'll likely never get to hear/play these classic amplifiers to make a real world assessment. I've read BBKing once used this amplifier but that may not mean much as there is a lot of hype in this industry,, And I don't play all blues songs anyway.

All I know is a lot of compressor circuits for guitar are just plain useless and often a noise problem.
Boss CS2 and a lot of others that use the ca3080 chip are a waste of time, IMExperience xP

Some later units use the THAT chip which seems to be less noise prone but still no hair on the edge of the note.
I had access to a Boss GT100 digi pedal recently and it was a pile of trash, the compression was cold and clinical, to it's credit some of the cab sims sounded ok but you just can't get that classic rock sound ,, That little hair on the edge of clean chords / notes that are part of the magic of early rock guitar. Maybe the L5 can do it but I'll never know. :'(

Anyway,, I'll post my new little circuit soon which uses the compressor I mentioned and basic clipper and a cab sim. Just need to find time to record a snippet of audio to keep Mr Fahey happy. ;)
Phil.

J M Fahey

Yes, the L5 saturation stage does exactly that, at the same time rounds, compresses, gives edge and generates even harmonics ... all traits of "tube sound" .
In this case there is no hype because Lab folded many years ago, and tubeheads *hate*  "sand" so if BB used that amp it's because it did what it promised.
It was justly dubbed "Twin killer" or "Twin on steroids" , both labels apply, but 99% of the Guitar World dropped Twins in any incarnation and embraced Marshall, so it was doomed anyway.

For the very sad story of Lab Amps and Dan Pearce, the designer, read:
https://www.gearslutz.com/board/attachments/so-many-guitars-so-little-time/311163d1348888612-please-recommend-solid-state-guitar-amp-pearce_lab-amps-history.pdf

feel free to comment :(