How to check O/T health?-Mesa ReverbRocket

[UniPlank84's]

Hello,

I am seeking info on diagnosing the O/T function and health on a double-ended el84 Mesa.

I just registered so I can post here. I have read many informative threads here in the past as my interest in tube amps and electronics has grown to facsination and interest beyond just the plug-in how-to's.

As an intro to the Forum Folks regarding my electrical and music experience=I have played music since 3rd grade and have scratch built several guitars in the past 4 years from plans and some self-designs, a woodworker 30 years by trade, and am well versed in tool and machinery use and have worked with high voltage electrical for many years and am comfortable and competent around it.

My electronics experience to date however is only in guitar p/u and circuit installs and tweaks. But I have been absorbing more and more info's at music related sites and am now very interested in building my amp and accesories experience and skills. Have read Dave Hunters 'The Guitar Amp Handbook', have a good DMT, and have a good Electronics Course text also I am studying to learn components and functions etc.

So now I have a repair situation I hope to use as a springboard-

Amp situation- I have a Mesa ReverbRocket which is now about 9 years old, in mint cosmetic. Time on electronics unknown-bought used at ebay. The seller told me it was rarely used, but I don't assume that is valid. I installed new Mesa el84's right away and it has been error free for 8 months. 3 channels all worked perfect-clean-much gain. I play at low-mid vol and clean to mid level gain settings, avg 20 hrs a week max. The 12..7's seem fine, but are a mishmash of Sovtek & Mesa-obviously some are replacements.

Then 2 weeks ago it went dead. I have since got info from Mesa and local tech via phone, and schematic from Mesa. I slowly and carefully delved into the chassis. To date I have confirmed accurate voltages at standby sw, fil's, power tubes supply. Nothing is obviously amiss inside, speaker(s) confirmed good via a working amp. I bought/tried new TAD el84's and 1x 12ax7. No go.

It does emit a very tiny sound at the speaker which is musical in the 3 channels-no hiss etc, but the vol/gain controls must be max'ed to get that- just not an amplified signal it seems-need to get right up to the spkr to hear anything.

So does it sound like a OT problem, and how does one confirm its function or lack there of?
Any other ideas and advice much appreciated also.


Thanks, John

[Chuck H]

Ok...First let me say that working inside a tube amp is NOT like working on any standard "electrical" wiring. This is an electronics situation. It can be confusing for the novice. Tube amps have power supplies that store charges of many hundreds of volts that can discharge at many amps if the wrong contacts are made. There are potentially LETHAL voltage and current situations inside most tube amps. So whatever you consider when inside a tube amp, when in doubt, don't do it.

Also, It is important to discharge the amp before performing any actual work on it, ie: soldering in or removing components, etc. This can be done by turning the amp off, standby in the play position, connect the plate of any tube to the chassis of the amp. An alligator to alligator clip lead is handy for this. Leave the amp this way for at least a minute and then test for voltage at the "+" end of any filter cap (the large "can" capacitors). You want to see 0 volts. Then unplug the amp and your ready to work on it. Do this every time you perform work on an amp.

Unfortunately this cannot be done while taking voltage readings. The amp must be operating. So the more electronics (not electrical) understanding you have, the safer you'll be.

What kind of test equipement do you have?

That said... I once had a Mesa amp that was not operational for a while before I discovered that I had accidentally fliped the "mute" switch that hides on the bottom of the chassis. It's a well hidden switch on some amps so it's worth checking first.

Also, the effects loop jacks can become dirty in these amps and cause the problems you describe. A little cleaner and quickly plugging and unplugging a male jack can often fix a problem.

Also, Mesa amps seem particularly prone to dirty tube socket pin holes. Again, a little cleaner and softly rotating a tube in the socket can cure this.

If it's not the mute switch or dirty contacts then let me know what kind of test equipement you have and we can proceed.

It's probably NOT the OT.

Chuck

[Chuck H]

Oh yea, The question...

Unplug the tubes and speaker. Then unsolder the OT center tap from it's voltage source.

First thing would be to check for shorts between the primary and secondary. No contact between the primary and secondary? Good.

Now check for opened windings. Is there continuity between the bottom of the OT secondary and all the impedance taps? Now the primary. Check for continuity between the plate taps and the center tap. Any high resistance readings are bad.

So far so good.

Now you need to generate an AC signal. A keyboard or even a low voltage AC wall wart can be used. Place a signal on the secondary of the output transformer. Measure it's AC volts with your VOM while it's hooked up and operating. Now measure the AC on the primary from plate to plate. It should be substantially higher than the volts into the secondary. Now measure the volts from each plate tap to the center tap. Are they equal (or damn close)? good.

If all checks out your OT is fine.

Chuck

[UniPlank84's]

Thanks Chuck for the detailed reply. I will check the loop jacks and socket pins, then proceed to sequence through the O/T as neccesary. The check will be good learning.

I appreciate your endeavor to impress the hazards involved-deadly serious game inside the box indeed. I have been lucky to have gotten the message well impressed right away when I first began to be curious- it was at Weber and elsewhere that I found the info to hazards etc. when I searched for kit info. The amp tech/building community is very good at impressing this, and I can relate to it in a different serious venue where I have been the guard to others' ignorence. Thank you.

I did come across the 'drain the filter caps' awhile back and how to do that using the standby-left-on method, which I think TedWeber mentioned somewhere on his site, along with the caviat that it is no guarantee they will be at '0'v so better to not rely on the method. I will get or make a discharge stick descibed at Weber, or use the jumpers to the plate method *each and every time after power-up running of the amp during work,mod's* Also keep arms and hands clear of the chassis when probing leads seems neccesary practice and don't work barefoot be the pool on hot summer afternoons etc.

I did do the Standby-on method to confirm '0' volts at the 30uF filter caps in the Mesa and it worked fine- I left it overnight to drain, then checked-meter to 600vdc, blk to grnd lug, carefully/accurately-red lead to +++lead of the F-Caps. I do need to add shrink tube covers to the probes I have to reduce the exposure of the tips to touching something in error-seems a good idea to me.

Mesa Schematic states 330v p/t out, 400v at the standby sw. So I powered on to check those- values confirmed good. Went to the el84's and both confirmed good very close to spec. I have not yet done the v1-v4's points. There is copper foil sheilding covering the V1 and V2 socket pin access so I left those for now.

The reason I am curious to check the O/T- I came across a mention that an undersized speaker cable/wire to a extension can damage the O/T. I have been playing this with a 12" ext cab along with the 10" via the 2 4ohm jacks for the past 5 months, inadvertently and ignorently with a guitar cable which I believe is 20ga. So that has me wondering since 16ga is recommended.

My test equipment is Ideal#61-361 true rms,phase rotation,capacitence,etc dmm. It is not Fluke-which I pondered purchasing, but I found out in my searching this past spring that it is a good competent 2nd tier meter so I bought that and have the manual at hand...a hobby at this point so I have to balance quality/cost.

Pretty long post here. I'll proceed on the checks. Also, this is a very complex unit with many integrations and layers, as well as the pcb tiny-ness of scale. So it may well be a case of a very good study subject for me but leave the surgery to the pro's at Mesa or others. The other possibility is repair may be more $$ than the value, and I may persue a new simple build using the chassis and cab. ie one of the ax84's. The Matchless Spitfire also is intriquing-I have found two versions of that and it looks not too complex...

Anyway maybe cleaning or a simple replacement part is all it needs and not a new pcb/components r/r.


Tnxs, John

[Alex R]

...by the way, in case it does turn out to be a bad OT... Mesa charge a lot for their spares. But a wide variety of OTs will do for that amp. Mojo do what they call a 'Blues Junior upgrade' OT that might be ok, and it would cost you a lot less than the Mesa part. I have used these as replacements in various 2 x el84 amps and they work well.

I do come across quite a few failed Mesa OTs, and I note that they are often quite small units. Maybe they are going for OT saturation or something. I wouldn't draw the conclusion that they are unreliable - I mean , if you advertise for folks to bring their broken amps to you and then you see some amps that are broken maybe that's not such a surprise - but I can vouch for the fact that Mesa OTs are not bulletproof, and I can certainly aver that Mesa charge a lot for replacements.

[Chuck H]

Alex, With all due respect and recognition of the good will intended I'd like to say that I would stick with the Mesa part. I own a Mesa Subway amp (at least it WAS a Subway amp befor it got to me) and the OT is a very high quality, good sounding unit. The amp sounds so good with the mods I did that if I had to replace the OT I wouldn't want to change a thing. By the tube compliment and funky impedance arrangement described I think John probably has the same OT. It's a good one.

Chuck

[UniPlank84's]

I am pondering the possibilies Alex mentions if Mesa replacement is very high$$, which I suspect it will be since they operate here in Cal- the 'Golden' state However I think Mesa is super and there integrity to their clientele and product exemplery, and I am a long time fan. Wish they had more bluesy lower watt stuff in production-oh well the market moved and they followed. Of course I wouldn't mind a spin myself on a vintage Santana Stage kit...that would be interesting...

Looked at specs and dims for a possible non-Mesa replacement- compared with the actual Mesa part Webers#W022913 small ot is same dim's scale but spec's different w/ 8ohm spkr out only- a possibility if Mesa wants dumb$$ for a replacement. The Hammond 125dse or such..???- that is another thought. Weber's SmokinJoeII is 2 x el84 and uses the smallish ot also - 8,000ohm -so that gives me an idea what is req'd generally for the 84's in that regard-I assume the SJ2 is a-b like the Sub. The Hammond would allow the existing values and spkr outs.

I like the idea of building up something simpler in this that I can maintain if this is too complicated a repair. Like the SmokinJoe1 or 2 which appears doable for a newb. The various Marshall 18ers with Vib appeal more personally but may be more than I should attempt. If I did a change-out I would do it so it could be returned to original config for posterity sake.

I agree Chuck that it makes sense Mesa would intentionally utilize a smallish unit to perform at the outer limits in two wheels off the road mode-so lovely-and never 'boring-k'. Doesn't hurt the profit line either I suppose. Like pcb's etc. Still sounds good though if it can stay on-line enough not to alienate clients, or lead to hackers like me building in easier to work on Vox,Marshall,Matchless networks.

Also I had a closer look at the ot leads etc just before writing this. Lo and behold! there is a well made in-line splice in the brown lead near the noval @ pcb point. With the information you both have shared regards Mesa's o/t's I suspect this may be a second unit and may be ready for a third since my cabling error. So now I'll proceed the check out and see what data I can mine-new to me this will be so hopefully I can get useful info.

Hmmmm...Marshall18Vib...hmmm.

John

[Chuck H]

FWIW if you want to save $$$ on an OT over the boutique and oem stuff, I would reccommend the Hammond 1608 for a pair of EL84s. It's rated at 10 watts, but you can ignore that. I promise. It's a very good sounding OT for 2xEL84 / 15 to 20 watt guitar amps. I've even had custom proprietary Heyboers wound for the same projects and the Hammonds sound just as good if not better. (but Hammond doesn't do deals with MFGs, just distributors. So I can still get the very good Heyboers cheaper). But for DIY stuff the Hammonds are really affordable.

Radio Daze sells the 1608 for US $43.60

http://www.radiodaze.com/

Judging by the size it should fit with a little modification into your Mesa amp.

Chuck

[UniPlank84's]

Here is what I found, hope my results and terms are correct-still learning-

16.6volt signal gave: Primary-97mV plate to plate, 196mV both sides of CT.
Resistnc=.394 p-p, .196 & .196 balanced, @2K scale

signal to Secondary-blk/grn-1710, blk/ylw-1693mV
Resistnc- *Beep* no readings value,

My manual only states the continuity check 'Beep' as a feature, but does not explain it's funtion in any detail, and I have never needed to use it before. Soooo... I assume this means 'short' and indicates a no good ot.

If so that is good news as that would be simple repair and allow a better grade of part to be used.

Thanks for the link to RadioDaze and the 1608. That is a perfect fit so I can keep the 4r and 8r outs. It may be different mount holes but I can make a mount to keep chassis holes as-is.

Thanks again for your help and advice,
John

[Chuck H]

That beep happens because the resistance is so low that your meter reads a dead short...

BUT, that does not mean that your OT is bad. The resistance on the windings in an OT is so low that it will read as a dead short with all bu the most sensitive (and expensive) equipment...

Did you run all the checks EXACTLY as I posted? I'm having trouble making a conclusion from your post.

A typical input signal on the OT secondary would be .5 volts. That should render a reading on several volts on the primary. Where in the world did you get a 16.6 volt signal? And, did you apply that to the primary? Or the secondary? You must do the tests just as I described to get a usable result.

Chuck

[UniPlank84's]

OK. All I have is a 'wall wart' as you recommended which is labeled 12v from an old sony phone answering machine-maybe this is too high v's, I can see if I have a smaller old from a portable radio which may be 9v. Keyboard voltage is lower(?)- I am reluctant to cut a plug/wire from a good spare keyboard I have at hand. RadShack may sell the plug-don't know-trying to keep this as simple as possible.

The 16.6v is what reads on the meter. The numbers I posted came from two rounds through the sequence you described as best I can. I used this signal on both Pri and 2nd. I did cycle through the scales to get readings and the audible beep is on the 2k of course, but I also tried the lower scale and got nothing, which fits what you say regarding not having proper test equipment.
If I can't get any useful information to you I may just get the new transformer and try it, then if still broken I'll decide wether to keep the Subway until later repair by Mesa or sell it to fund a simple first build. From what I have read it seems I could use a p/p 10 watt in a se build like the ax84 p1 or the el34 based eXtreme.

Ho-hum, thought I had it, and was feeling good again about the Mesa's prospects for being up and playing again. Back to the salt mines.

Thanks for the help Chuck,
John

[Chuck H]

Is that Sony wall wart AC or DC? It needs to be an AC wall wart.

Chuck

[UniPlank84's]

Hello- It's a DC tx'fr Chuck. Sorry for my ignorance. I called a local electronics store to ask about this transformer before proceeding and got impatient advice from the clerk, who said it didn't matter, so I went ahead as I wanted to get on with it, and it isn't an expensive part. I just was hoping to get confirmation one way or the other so I know what direction to gowith this amp.

After my last post I called an electronics tech/assembler and he talked with me in a helpful informative manner and said the secondary needs a very low v ac signal - .5v - to test the secondary, what you said after I tried without enough info. I should have asked for more info before I proceeded.
Mostly I just would like to find out if it IS the OT. If not it seems likely the problem is buried in the PCB with many many tiny caps,transistors, etc and obviously beyond any home-based service. If so it may just be a loss as shipping and Mesa's service cost/time may not be worth it. There is a very friendly and reputable TV/Vid repair here who may be willing to check the ot and confirm its state.

FBOW, If it is a serious and costly repair, I am comtemplating scavenging what I can from it for a el84 based build. Since it IS a Mesa that is not an easy call however since they have pedigree. If it was a China made cheap amp or such then salvaging chassis etc. would be logical.
Anyway I will see what my local TV folks say about testing this. AES is still having their sale so buying a 1608 may be an OK way to see what happens with a new ot. If the Mesa is still bad and sells as a repair project, the 1608 is good for homebrew 6v6/el84 type amps from what I have read so far so I could use that elsewhere.

Thanks again, John

[metalangst]

I'm not sure why you have singled out the transformer as the culprit. Did you check the jack switches? If there is an effects loop you can run signal into the return to test the output section. If there is no effects loop you can inject signal into the phase invertor to check the output. You can also take the signal fron the effects send to check the pre-amp. Divide and conquer.

[Chuck H]

We covered the posibility of dirty switching jacks earlier on. But I think using the effects loop to help isolate the problem is a stellar idea.