Rebuilding the Traynor Mono Block II

[infinia]

The "Leach amp" in its first iterations is not correctly designed.

It would be far better and easier to improve the existing circuit
of your amp.
Implementing a complementary output stage is quite easy,
not counting that there s some minor tweaks that can be made
to greatly improve the performances.

All in all, there wouldn t be more components count.

I agree with improving the original w/ judicious mods and some rewiring. Rather than just chucking the old, without understanding or discussing reasons for the perceived failure modes. Just replacing the amp with another design maybe doomed to repeat the same or adding a new set of mistakes. Sticking w/ quasi outputs and bootstrapped VAS is not a bad thing in itself, after all, this is not hi-fi.
The first step would be understanding the OPs use of the thing and what could cause the potential failures. For example using 2 bass woofers could be closer to 3.x ohms load with up to 3 mH beyond resonance. Is the stock heat sink insufficient? Considering the mass and total radiating surface area of the inside / outside of the chassis is used here.
Adding another set of modern NPN output devices would be high on the list. Improving the dreaded severe gain droop of 2N3773 and increased SOA s/b number one priority. An Idea - using 6 x MJ21196 on the existing cast heat sink and adding MJE150xx drivers. This would relieve the VAS current drive and gain. Maybe simply changing the speaker interconnects to 'speak-ons' can keep outputs from shorting and users from wonky cables or spare microphone cables from being used. Also little things like improving input HF filters and output zobels could make a difference here.

 

[wahab]

Agree with most of your propositions.. As proposed elsewher in this thread, keeping the amp in class B, largely underbiased is a very bad idea, as at low volume, since the crossover is high frequency, it would be more than audible, since the instrument produce few high frequencies that would cover the said distorsion. MJL2119X are cheap and enough for the purpose, although i would switch to complementary OPS... As VBE control , it would be judicious to use five low voltage drop diodes , the kind of BYW80F200 which are insulated TO220 case. A 50R trimpot should be paraleled with one diode to provide an adjustment. This combination is slightly overcompensating , a welcomed feature for an amp that surely will be used at high volume. This , of course in case of complementary output stage... As drivers, the MJE1530x are rugged and suit the work... Compensation cap should be reduced to about 68pf.

 

[infinia]

yes similar, as I said earlier on here, I agree with lightly class AB. for slightly different reasons than it sounds better. discontinuous currents can excite oddball resonances you can get with longer wiring and bottom half of quasi driver / outputs with remote RE. I see adding a Bandaxall diode. Bias with a string of 3 diodes (1N4148 ) on a lower current VAS. Depending on bench testing this may need to be 3X 1N4004 and a 20-50 ohm resistor string. You favor HiFi Bias levels here?

 

[wahab]

I ve no i idea on the optimal bias level for a quasi comp.. 3 diodes should do it , not forgetting a trimpot in paralele with one. I will just point that the output stage is unbalanced in the sense that a pair will inevitably have a higher bias current than the other pair... The VAS current is about 13mA . It was settled as high because the amp was (mis) designed to handle 2R loads... I think that it can be left as it is. I made some sims of this amp, and reducing this current doesn t bring a clear improvement. Anyway, as designed by Traynor, it s a disaster, at least according to my simulator.

 

[PB2]

Note that the top amp shown in Figure 3.23 here does not work: http://www.thatraymond.com/downloads/solidstate_guitar_amplifiers_teemu_kyttala_v1.0.pdf
Self makes the same blunder as noted here: What's wrong with this picture?
Slone also makes this mistake as noted here:

Unstable VAS current in amp from Slone book - diyAudio I actually prototyped it when I was a teenager to find that it would not bias correctly. There are long discussions on here concerning that configuration. This amplifier has an unusual output stage with voltage gain very similar to the old Tiger series that is legendary for burning out.

There is a discussion here covering the design and the Protection and SOAR sections might shed some light on what is required: ETI-466 Amplifier Technical Review Anyone know of a spread sheet covering VI limiter design?

A complementary Vbe mult so that you can put one on each heat sink: The Kleinschmidt 25A - diyAudio SOA discussion: http://www.diyaudio.com/forums/soli...s-paper-1992-i-v-locus-issue.html#post2151057
Output stage wiring discussion: http://www.diyaudio.com/forums/soli...output-stage-used-open-loop-2.html#post386059
Interesting output protection here, he used 2A devices, LOL but there was no VAS protection so they had to survive the max VAS current: www.diyparadiso.com/datasheets/anpoweramp.pdf I would say that there is a lot to learn from the app note but that the design is not the best, especially the protection. This is the part that I suggested for an improved VAS transistor. It is worth taking a look at the last SOA chart to see how important it is to keep it cool. It would idles at about .84W in this circuit. Keep in mind that only the DC curve applies since the pulsed values are non-repetitive, also note that the chart values are for a case temp of 25 deg C and must be derated for elevated temp: http://www.toshiba.com/taec/components2/Datasheet_Sync//66/7870.pdf

 

[infinia]

Consider halving Iq for the VAS circuit, this reduces Pd to 0.4 Watt. The previous design ran Iq so much higher than normal because of poor driver and OPT gain at full power (gain droop). Iq using 2x 3.3K 1/2W
With new OPT and drivers, suggest using 2SD66a or the modern Fairchild KSC2690A ~ $0.40 . Note TO-126 so EB is reversed from TO-220. flip 180 to small heatsink using fiber washer for stock mounting.

 

[PB2]

Interesting thread on SS Guitar Amps:
Solid State Guitar amps worthwhile??

Another interesting example, not necessarily good, the Acoustic Control 370 bass amp:
GuitarGearHeads - Gear Up Your Sound - User Reviews

Schematic:
Free Pages Personnelles: Erreur 403 - Refus de traitement de la requête (Interdit - Forbidden)

 

[cloudhopper]

Hey Pete, Interesting thread about SS guitar amps. I've heard amazing musicians use both tube and SS and a lot of it has to do with what the guitarist is looking for. BB King plays an old Lab L5 SS amp and with him playing it, it sounds incredible. His style doesn't include a lot of distorted overdrive.

What I like best about tube guitar amps is the touch sensitivity right at the point of distortion. Pick attack can determine if the amp stays clean or digs into a very musical distortion. That type of response is difficult to get with a SS amp. But then again, the metalheads want distortion all the time no matter if they are playing soft or really digging, so a SS distortion all the time is what a lot of them want.

With bass guitar, most generally a clean crisp amp is desired with very little if any distortion. Bass takes a good amount of power and while I've built tube amps for bass, I don't see the benefit for most styles of bass guitar play. I do like to use a tube pre sometimes for bass when I want to add some color to the signal or get a little fuzz in the signal.

Interesting schematic posted too, I've been researching more amps trying to learn more about various topologies and protection approaches. I've ordered Bob Cordell's book too. Hoping it will give me even more insight.

 

[PB2]

Here is a schematic for the LAB 5 and some other versions. Seems to be rated 100/200W into 8 ohms I think depending on the model - will probably do much more into 4 ohms. Note that 8 output devices are used:
http://www.netads.com/~meo/Guitar/Amps/Lab/Images/Lab_Series_L5_L7_L9_L11_scheme.pdf

Do you know what speakers BB King favors in his LAB 5?

I've also ordered Bob's book, looking forward to it!

Here is a guitar amp that I started for my older son:
http://baselaudiolab.com/TWINPOWERREVERBGUITARAMPLIFIER.htm

Here is a thread about it, but the first link is old so use the above link - can a mod fix that?
http://www.diyaudio.com/forums/musical-instruments/81449-our-guitar-amp-project.html

My son's thoughts on what he wants for an amp has been evolving - it would be easiest to
do a solid state design, but I want something really good for him and we were mainly looking
at the MESA Dual Rectifier but something like an updated LAB 5 would be easier.

 

[cloudhopper]

Do you know what speakers BB King favors in his LAB 5?
My son's thoughts on what he wants for an amp has been evolving - it would be easiest to
do a solid state design, but I want something really good for him and we were mainly looking
at the MESA Dual Rectifier but something like an updated LAB 5 would be easier.

I'm not sure what speakers BB uses. My current favorite is the MOJO clone of the Celestion G12H30. It works really well in both open back and closed back cabs. I have real Celestion G12H30's and like the Mojo clone slightly better. The Mojo Greyhound is a very good speaker too and my dual 12 cab has one Greyhound and one Warehouse Veteran 30 in it.

Mojo Speakers British Vintage Series 12'' BV-30H Speaker 8 OHM 30W 1.5'' Voice Coil 50 oz. Magnet

I like the Warehouse Guitar Speakers too. I've used several of the Veteran 30 but like them best in a closed back cab.

Watch your Tone! | Warehouse Guitar Speaker

I just picked up an Eminence Wizard but haven't given it a good evaluation yet. It's supposed to be similar to the Celestion G12H30.

Again, just like amps, speaker selection has a lot to do with personal tastes and style of music played.

If your son wants awesome touch sensitivity, go tube. It's really not difficult and it doesn't take much power because guitar speakers are so efficient now. Most are around 100db at 1W, 1M. My 18 watt can put out 112db at 1M and that's enough to hurt unless he's playing large venues. In the band I play in, our lead guitar player plays through my Snot Watt design which only puts out about 4W. It's amazing how much volume a 4W amp puts out with an efficient cabinet. I always put a mic (usually a SM47) on it and run into the PA. Makes for lower stage volume and a much better mix out front.

 

[cloudhopper]

Dave at Baron Amps helped refine the Snot Watt and he builds them also.
Here's a clip of the latest Snot Watt.

YouTube - SNOTT WATT

 

[PB2]

Hi Carl,

Thanks for the link and the info. We ended up with a pair of 12" Weber speakers that my son really likes.

Here's Doug Self's Blameless design that is often talked about here, it is essentially a power amp version of the Jensen 990:
TalkTalk Webspace is closing soon!!

The Jensen 990:
Simulation of the JE-990 OP Amp By Deane Jensen - diyAudio

This Traynor design has poor rejection of power supply noise and you can read more about it here:
Doug Self article in EETimes - diyAudio

Changes could be made but might be best to just get it working first.

 

[Traynor Mono Block]

Hi I'm looking for the on/off rocker switch for the Traynor mono-block head. Does anybody have one laying around they want to sell?
r.maddeaux@gmail.com
Thanks

 

[maestroaudio]

Hi, has anyone successfully replaced the output transistors in this mono block?

 

[thoglette]

Did you actually read the thread you've dug up?

The first page points out that these amps regularly blew output devices which were regularly replaced. Before they blew up again.

Heck, this was true for most solid state amps in the '70s.

 

[maestroaudio]

Yes, I'm aware of the problems with this amp, I would just like to know if anyone has come up with a solution.

 

[thoglette]

To which problem? More specifically, what's the problem with your amp? You can still buy 2N3773s from Mouser.

 

[maestroaudio]

This amp needs 1 x 2N5630 and 1 x SJ5480. The rest of the transistors test fine. I've already replaced the leaky filter caps.

 

[Mark Tillotson]

2N5630 are rated at 120V, which is one reason it blows (the rails are +/-56V, ie 112V which is far too close to the explode voltage of the output transistors) A more robust alternative is required.


2SJ5480? No such beast as far as I or google can tell. SJ5480 might be an OEM code I suppose.

 

[thoglette]

I was thinking about that last night: the design came out when 110VAC was more of a target than an actual voltage, these days many people are reporting nearly 120VAC in the USoA. That takes those rails to +/-62V.