Marshall JVM Bias Drift PCB Surgery
Vložit
- čas přidán 23. 11. 2018
- A brief video demonstrating how to perform a Marshall JVM 410 PCB modification to deal with unstable bias and PCB conductivity causing blown tubes and high tension fuses.
As with many modern PCB-based multi-channel Marshalls, this JVM 410 was blowing the high tension B+ fuse. As recommended by Dr. Tube (drtube.com/en/modifications/j...) the power tube grid leads (pin 5) were isolated from the PCB to ensure (-)DC voltage reaches the tubes without any possible leakage from the other tube socket leads.
Hack to safely bias this amp using a Molex 3-pin connector is shown, that Molex connector is: www.ebay.com/itm/Molex-3-Pin-...
Additionally, a filter choke was added (see the link below on how to install). The filter choke helps smooth out the B+ which feeds the preamp tubes and helps reduce a lot of the noise from the OD channels.
Also if you like my channel and this video on amps, I suggest subscribing to the following awesome channels:
-Uncle Doug: / @uncledoug
Uncle Doug has tons of other videos, probably the best teacher on CZcams regarding amps
-The Guitologist: / @theguitologist
Brad runs an all-around great channel, loves to work on wonky stuff as well
-D-lab: / @d-labelectronics
Terry is likely the best electronics technician working to repair tube amps on YT
Related links:
-Safety! Rob's word on capacitor charge: robrobinette.com/Tube_Amp_Saf...
Rob's site is likely the most comprehensive resource regarding classic Fender amps, and now amps work.
-Brad the Guitologist working on a similar Marshall: • Mother of All Marshall...
-JVM Choke Install: • How to add a choke to ...
-Why install a choke: • PI filter, Resistor Ch...
-drill bits: www.amazon.com/gp/product/B01...
-Marshall Bias Drift info www.lynx.net/~jc/TSL122.html
-Uncle Doug on Biasing: easiest way to really understand how biasing works
• Biasing Single-Ended A...
• Biasing Double-Ended A...
• Biasing Double-Ended A... - Jak na to + styl
sheesh, that’s pretty involved :o I just picked up a used JVM 205 and really like it, but now I’ve got a case of the worries! Appreciate you posting this - great, great info. Guess I’ll just play it till I have an issue. I play at home and never have my masters up past 10:00, I’m hoping that’ll help keep things cool :-)
Ha, yeah not the easiest of procedures, but if it's to be done, this is one of the ways. Hopefully, you have one of the JVMs where Marshall has corrected this issue.
If it were my own JVM and I ran into these issues, I would likely pull the board and convert to a JCM800 or another amp, as the transformers are great for that type of conversion.
@@jloewenheim - my JVM is only about 5 years old, so… I’ve been all over the JVM mod forum, some are real simple, others not so much. Isolating pin 5 is really far out and that glass drill you found is GENIUS. My skill level is bright circuits and FX loop improvements. I’ll definitely pull my chassi and give it a good look. Thanks again \m/
Wow!! This is the best video on the JVM power section ever!! Please do more videos 🤘
Thanks working on it
Excellent video Jake, thanks for sharing your knowledge.
Thanks Rob
Note that I lifted the 5.6K grid stops at pin 5. With no tubes in the amp I still get -10 t0 -12 VDC bias voltage. Way too positive. I suspect one of two of both issues. Board conductivity between the traces from the bias voltage traces to a HV trace. The other thing is the filter caps on the bias voltage, C22 and C23. Both exhibit an ESR of 5.0 ohm Leaking to ground will in fact make the bias voltage more positive. I will pull the board and replace the caps Report findings, I will also isolate pin 5 of the EL34 tubes.
Nore all tubes are good but have run hot, 1 ohm resistors are OK, ESR measured with Atlas ESR70 . Bias rectifier Diode tested good
Hopefully, you can get it solved.
Amazing video! What is the size of the Molex connector?
Great vid, for sure! Thanks a lot, I really appreciate it! Got a little question, is it totally safe to measure negative voltage on pin5 without any tubes installed in amp? As far as I concerned with no power tubes all voltages to the tubes will increase their nominals. And do I still need to connect a cab to my amo when powering it on without tubes installed (preamp tubes and power tubes)? I'm a bit hesitating on that one so that I decided to double check the info and ask you for an explanation here. Thanks in advance!
Thanks for the note. The short answer is yes, you can measure the voltages with no tubes in, and it is the safest way to do it because you can insert your test probe in the socket where the tube pin would go instead of having your hands near all the other circuitry from behind. Also, with no power tubes in, you do not need to have a speaker load connected.
All that being said it is still dangerous, so be careful. You are correct that voltages will go up since they are not loaded down (but not a significant amount). The most important thing is that you have stable negative DC voltage in pin5, it will be somewhere in the neighborhood of (-)40-55 volts DC.
Please feel free to email me directly if you want to discuss this further: username@gmail
@@jloewenheim Thank you very much on clarifying it and for giving precautions! Yes, I am aware of a potential danger and do know how to handle high powered equimpent for sure.
Thoroughly informative! Thanks for posting and sharing your knowledge! I have the JVM 410 combo. I turned it on, after about 5 minutes I went to tune my guitar. While nothing was plugged into the amp’s input there was suddenly a loud pop. An audible hum has existed ever since. Prior to the pop I did notice some scratchy noises and light crackle sounds that I had never heard before. The amp is 3 years old and only been played a low levels. The tubes have never been changed... any suggestions are very much appreciated! By the way what level did you bias each side to? Any tubes perform/ last longer than others? Thanks again!
Thanks for the nice words. The first thing that I suggest is that you make a current limiter. Uncle doug has a great video if you need some hints: czcams.com/video/wRFRwOnLsZI/video.html. This will help you find out if there is a current spike from the tubes, and will likely protect the amp. Second, I suggest a bias in the neighborhood of 40ma per tube, this will show up as 80 on the bias test points since you are dealing with 2 parallel tubes per side. You can go deeper on this being that each tube has a bias shunt resistor, allowing you to easily measure what you are getting on each tube, if one is out of spec, you will need to replace. I would start with these tactics first.
Finally, I would also go check out rob Robinette's pages. So much theory in there explained in plain simple terms, having a real understanding of the possible culprits will hopefully help you save hours of frustration. After that, feel free to reach out to me (username @gmail.com). Hopefully just a tube. el34's can be a pain in general. If not, we can discuss the other possibilities.
@@jloewenheim Thank you for getting back so quickly! I will follow your advice. In which state are you located?
@@tabodee CA
Yep - my tech called me up today and sent me a picture of the same exact resistor that had the same hole, in the same place- lol. He has ordered new resistors. And had to change out all of the power tubes. The bias was like 68 on one side and 24 on the other.
@@tabodee Sounds good, you still need to keep track of it that board has become conductive. waivering voltage bias will again kill the tubes, so hopefully the new tubes will cure it. if not he may have to do what i did
Top video, my tsl 601 has little bit of bias drift, ie set it at 80mv and it gradually goes up to 85 -90 Plus ish during warm up and then goes down to about 75mv ish after about 1hour, never absolutely stable,(twin jj EL34 fairly new)
Amp sounds ok, no other issues. Do you reckon this is within limits with your experience? resistors seem ok. I am not sure whether my model amp is susceptible to this pcb problem.
All the best from yorkshire UK. Bill
Thanks, Bill, I don't have experience with your TSL specifically (please let me know the year), but if bias is jumping around, things can start to get questionable. Although that Bias measure may work, with the limits being in the neighborhood of 45 per tube, and showing 90 when you are adjusting; the bias jumping is not a good sign. I would suggest keeping an eye on the tubes, and if you see any red plating or the HT fuse goes, you may have the same problem. JCM2000's (DSL and TSL) are notorious for having this problem. Let me know if you want to discuss further.
@@jloewenheimThanks Jake? I presume that is your name , My amp is 2008 and from further info the problems with pcb's were earlier models ie up to 2003 I understand, My mA come down after initially rising and so I do not experience red plating and sound is ok as far as my playing goes Lol!. Your mod seems to attend to pin5/pcb only whereas others seem to include a pcb bias supply mod as well, is that really necessary?.
Nice to speak with you cheers Bill.
Jake is the name they gave me. If you want to email me, you can do so by adding gmail.com to the end of the username.
Some things to consider: First is that the JVM I worked on is a 2009 (I think you are seeing Dr. Tube's article which doesn't include the JVM). I say this because the 2003 year number is only the timeframe of which Marshall has admitted to. That being said, it is my understanding that the conductivity issues are still an ongoing issue even in current production (I doubt Marshall expected non professionals to turn it up to full volume). Second, the JVM has a very different bias circuit layout. As a result, a new bias board did not make sense on this amp. If it turns out to be an issue, I will have to go a different customized route.
After much thought and research, it seemed the only logical starting point was to eliminate the chance of (DC+) from pin 4 jumping onto the grid at pin 5. Of course, the other option was to buy a new PCB ($650 US). If I screwed up the isolation, I would have to do that anyway, so I gave it a shot. If you start with the isolation, you can then troubleshoot after should issues still persist.
It has now been about 6 months, and the amp gets a lot of use at Spinal Tap volume levels. Things have held steady so far, and with the choke, it is better sounding.
what choke "henry" did you go with? great vid....you could have desolder pin and slip heatshrink though hole?...unless...lol...they decide to make tight fit...OY
ps done similar iso for adding 1 ohm to gnd from pin 8...hence a possible suggestion.
Thanks for the note. I believe it was a 4 or 5H choke that is stock for Marshall
@@jloewenheim thanks
Great video. Do you know if the Satriani version has similar problems? I know the board layout is different.
THanks for that. Hard to say, as understand it, they are quite similar except Joe had them add a noise gate and tamed things a bit. If the PCB is the same as the 410 and some of the others, there is a good chance of the conductivity and bias issues, so if you have issues, consider this as a possibility
@@jloewenheim They did a lot more as well. For one thing the bias pots are way over on the right side of the board near the filter caps.
@@SaintFredrocks yeah since you posted I did some research. As far as I can tell, they are still using the same PCB mounted tube sockets, which tend to be the real culprit for that Bias Drift. Hopefully, it won't be an issue for you. One thing they added and was very much needed is that internal noise gate. Anyway, if you run into problems, feel free to reach out. You can get me at my username @gmail.com
Jake, I have a JVM 50 watt, I'm getting V3 (only V2 and V3 both EL34s) that is running hot, I have the bias pot turned all the way down and the tube is still running quite hot 50+ma, also when measuring the voltage across the 1 ohm resistor, it is varying wildly from 40mv to 85mv. I've measured all the nearby and relative components, everything seems to be in order, does what I describe sound like the voltage leak issue?
that is definitely a symptom. That being said, one of your tubes could simply be bad, to eliminate that, switch tubes into the other socket. If this continues on the same socket its not the tube. I suggest you measure the bias voltage (will be negative DC) on pin 5 for each socket both with and without tubes in. If that voltage is not holding stable or is showing positive voltage, you likely are having the leak issue described. it is probably more likely to occur with the tubes in. Sounds like it is only on one socket, but if you plan to do the repair, you might as well do both. Also, feel free to email me if you like. username @gmail.com
Would you share any schematic you may have for the amp and or the 215c? PLEASE
dave do you mind emailing my gmail account (username at gmail) and then I can send you whatever you need?
I just had the same problem with my JVM 410H! One of my tubes blew, and after inspecting the Circuit Board I noticed I had the same Resistor and the same Fuse blown as depicted in this video. I was just going to replace the fuse and resistor and try to bias the new tubes, but now I"m thinking maybe I should Isolate pin 5 as he does in this video.... what do you guys think? Also my EL34 Tubes that I purchased only have 7 pins instead of 8. I went on some website that tells you what tubes you can use for replacement and JJ EL34CZ were among the recommendations.
thanks for the note. 7 pins on your EL34? is one broken off in the socket? All EL34 tubes should have 8 pins.
Anyway, if you have questions, feel free to email me : username@gmail
@@jloewenheim This fix worked beautifully! I think my Bias was at around 40 MV, I bumped it up to 70 and absolutely love the tone! This video saved my amp, Thank you sir.
glad to be able to help you find a solution!@@dansittler4619
I've had one of these back in the shop 5x with blown HT fuse, about 3 months between events. I'm wondering if this is the issue. I've fixed dozens of TSL and DSL with this issue but never heard of it on JVM410. I asked another v experienced amp tech and he hasn't come across it either.
Hi Stuart. This amp is still going strong after the repair, so there is definitely a possibility you are seeing that same issue. When I spoke with Dr. Tube, he said there clearly are enough similarities and that the PCB stock is very similar. Anyway, good luck with it, it is not exactly fun to deal with these kinds of hard to diagnose problems. Let me know if you have questions, username @ Gmail.com is the best way to reach me.
Do you know where to get the replacement resistors for R26 and R27. Will any 1 ohm work 1 watt work? I was really thrown by the bands at first.
Hi Mike, thanks for the comment. Although it likely looks like the 180 you are mentioning, they are using what is the 3 Digit resistor code, typically used for labeling SMD resistors. The code is "1R0", this link will explain how that works: www.hobby-hour.com/electronics/smdcalc.php
Now to explain why they are 1ohm. the reason for that value is so that the voltage measured on the test points will be equal to the bias current (using the formula V=IR, if R=1, then V= I). Hopefully, that clarifies.
@@jloewenheim you've probably noticed that I figured that out by my comment edit. Thank you. What did you replace yours with? I figure I need something that can take the voltage and current, the schematic says 1 watt, and should probably be something that can take a lot heat. There are a million different 1 ohm resistors on digikey and mouser. Any suggestions?
@@rustyzipperband Hi Mike, glad you were able to figure it out. I actually prefer to use 2 or 3-watt 1% resistors metal (3w are what I have at the moment), in particular when it comes to EL34 loaded amps. As you mentioned, you can use the general supply companies like Digikey and Mouser, you can also go to the amp supply companies like Mojotone.
@@jloewenheim awesome! Thanks for that info. There is nothing more frustrating to me than shopping for parts on digikey and mouser. If you have any other alternatives please do share. I've been a hobbyist for over 10 years building effects. I also built the guitar in my profile pic along with many others. I used to go though Tayda, and ebay. I've avoided digikey and mouser till about a week ago.
@@rustyzipperband any chance you can send me an email? my username at gmail.
Hi Mate, great Vid , just wanted to ask, did the pcb surgery do the trick? is the bias now stable? also does the choke mod help with the bias drift? ive just purchased a 205h used, pretty happy with it, once i set the bias best i could to 31mv.
i was thinking about mounting the power tubes to the chassis and wiring back to pcb, should help with heat dissipation.
Also, i'm guessing a jcm800 choke would do it? 6h i think.
What are your thoughts on just adding a cap either side of the resistor , instead of replacing with choke?
Since you have been modding this circuit, i thought you may have the answers !
Thanks again for the vid and cheers from Australia.. Mick.
Thanks. For now The PCB surgery has solved the problem. We are going on more than 2 years now, and it is still going strong. Since the PCB work, the owner of the amp had been playing quite often before our bar shutdown here in CA with no problems. The choke makes the amp quieter, so Bias drift is not going to be tackled by that, but much of the noise floor is gone.. My suggestion is that you play yours stock if you are currently happy with it (unless you are a professional touring musician), if you are lucky you won't encounter these problems. That being said, if you do, go ahead and do the surgery. Let me know if you need any advice. I can be reached at my username@gmail.com
Ive had 4 Marshall amps and all have had this bloody bias drift:
WTF Marshall - This has been an issue for at least 20 years and they haven't fixed it.
By The Way - My Orange Rocker Verb, has had none of the issues Marshall have - Maybe Orange are better amps and they know how to build amps properly..
where can i buy an R27 bias resistor?
there are a lot of places, but what you want is 1ohm, 3 watt, 1% tolerance. Try tubes and more, tube depot, antique electronics, Digikey, Mouser, or arrow.
@@jloewenheim Thank You sooooo much, really appreciated! .... I almost used a 1 watt 1ohm 1%
@@mikeliotta7530 1Watt will likely work as well, but I like the 3 watt in particular if you are using an EL34
Hey thanx again man! Ordered afew 3watt.... Goin in my jvm410. Have a great & safe 4th of July, Bud!
"Suspected" board conductivity LOL. The pcb is made of epoxy bonded glass fibre. It will not conduct. If there's a build up of dirt on the board, that might, but then why not clean it instead of butchering it? I also don't see any moisture build up possible with all the heat for the tubes. There must be another problem.
Thanks for the comment. Although, the ideal application-designed FR-4/G-10 should not conduct, unfortunately, this is not always the case, as there several factors (usually thermal) that can bring about this phenomenon. I work in the semiconductor processing and fabrication world, and this is quite a common challenge.
There are several academic and online articles you can look up regarding the thermal properties of this material, and the potential for expansion/contraction, which can lead to traces that would otherwise be isolated from each other, losing the ability to be insulated from one another electrically.
Dr. Tube has a practical write-up on this related to Marshall PCB's www.drtube.com/modification/jcm2000-stable-bias-mod/
I worked in electronics, including pcb design for over 20 years, and have never even heard of this. How do you make glass conduct by "thermal" means? How does expansion/contraction decrease insulation? I know all about arcing and clearances for high voltage PAT testing, but that's not the case here. The cynic in me says Dr. Tube want to sell their kits. They certainly don't have any creditable description or evidence for FR4 becoming leaky on their website.
@@ricobass0253 I'm certain you are very qualified, and always made sure to take all necessary precautions in your designs, and your point about the epoxy itself is correct. If you have a better method to solve these issues, there is a large community of Marshall owners that were love to learn about it.
The issues, as I understand them have to do with the expansions of the dissimilar materials; those being the traces vs the epoxy. PCBs are not only made of epoxy, as there are several other steps that go into making sure they are able to contain electrical circuits.
From my experience (on modern fabrication which does not involve tubes) we encounter several issues on the more modern boards, where cost savings are looking to be achieved. Examples tend to be related to humidity in the substrate, deflection during layer curing, and general instability of the related traces and pads prior to advanced packaging. These and other factors lead to shorts and other problems, ultimately leading to PCB failure (think about the number of flat-screen TV's that fail because they need a board replacement).
Nonetheless, I had to work on this amp for many hours before resorting to the brute force of isolating the components and having success. There are several tests, component replacements, and other methods that I did not go into for time's sake in the video.
As a final note, I easily lost money on the time spent working on this amp. I also spoke several times with Dr. Tube, and he had no issues discussing how one would go about this on their own. Based on my communications with him, it is hard for me (having no affiliation with him) to see how this is a ploy to sell kits.
Orrrr.... You can use a Bias probe instead of chancing damaging other components.. I've worked on hundreds of Marshall amps.
it's ridiculous how these amps suffer from the exact same problems as the previous jcm2000 and jcm900 series. the engineers who design these don't seem to care to fix issues that are designed into these amps for 30years now.
No excuse Marshall shame on you they knew the TSL and DSL both suffered from conductive boards, I have the TSL 100H
Which lasted for 14 years before this happened and I had to replace the revised mother board , I gigged that amp for six years solid with no problems, and it had external bias points and fuse holders, unlike the cheapskate JVM which I also own and have hardly used as I use a boss katana
Head now, (awaiting derisory comments..)
I'm not sure I have any derisive statements to follow your comment, but I do know that this kind of thing is frustrating (as a player and doing the repairs). I put the video out to help all of us who have encountered this. Unfortunately, there are some mismatches when it comes to trying to infuse legacy (tube) and modern technologies. If you look at my HRD videos, you will see that my real recommendation is to dump all of the modern components, saving only the transformers, and then converting to the original Fender and Marshall build technics (eyelet and turret boards). The only issue is that you lose the convenience of the IC's that enable switching and effects programming.
Nonetheless, thanks for watching and your input.
Derisory comments for using solid state Boss amps from tube snobs is what I was getting at lol!!! I labelled every connector to the board with little tape flags with the legend ID from the PCB, and double checked them before I disconnected them when I changed my TSL 100 mother board, I really didn't expect to see this on the JVM I'm very dissapointed with marshall over this especially being from the UK
I'll probably sell mine now
Steve Grindrod, who designed the two-input Master Volume 2203/2204 in 1976, also designed the JCM 2000 DSL100 in 1995 to compete with the Mese Dual Rectifier... the amps were rushed into production without the normal R&D working out any bugs, etc. and the PCB used in the early DSL/TSL/JVM had a problem with some sort of organic material in the motherboard... which caused all these issues with the board becoming conductive and all sorts of problems.
pots too high man...why you think your burning the resistor dude....id rarther save the resistor and burn shitty trimmer lol.
Just trying to work with what marshall gave in the amp. But I totally agree, the amps I build use Alpa and CTS pots to bias like Fender always did in the BlackFace amps
get that carbon film outta there..actually I don't give a shit..probably alright...but man burnt pot trimmer shit..gigginlingggg me lol.
well your badd with pott mann you burnt pott man...you can like go to ground..and split chain resistors man...no but its noiseyy tho...oh that's anywhere from 10-100k 2-5watters split off of it...its not a good idea dude and jerry rigging...