Help with Peavey Hybrid/Ultraliner output

[reubster]

Howdy-All from down under.
Been lurking for a while, this fantastic site has inspiried me to heat up my iron again.

I've acquired a "nearly" working Peavey Mx VTX....see schematic attached.
Its currently blowing OP Tube V1 and I believe I have found the problem with faulty Q6 & Q7 transistors which form part of what I guess is a solid state phase inverter.

Before I put tubes back in I have a question or two regarding the output stage.
It looks like an Ultralinear stage to me but I'm puzzled by the 100 Ohm connections to pin 4 [screen grid???].
I would have thought each tube should have this resistor, but the amp is wired as per the schematic, ie only one tube per pair gets the resistor.

So whats going on?
Should I modify the OP so all 4 tubes each get a resistor?

Also
Does any one have the actual MX VTX schematic?
It is not available on the Peavey site, I have this Heritage schem and it is close but the LV layout is different.

Cheers in Advance.

[Enzo]

You can always get any Peavey schematic from Peavey for the asking, they only post the ones that came in the owners manual to start with, but they are all available if you request them.. However, I just posted this one over on www.ampix.org in the Enzo gallery for you.


Blowing V1? More than one tube has been destroyed in that socket? Or we have just a dead tube there?

You have four power tubes? They are in parallel pairs, so if the cathode drive transistor for V1 was faulty then V1 AND V2 would fail because they are wired in parallel. Unless that screen resistor was open. V1 is not the end tube, they run V2-1-3-4. So if it is the end tube - V2 - your theory might be correct on the transistor and the other tube - V1 - is not melting because it has an open screen resistor.

Pull all four tubes and measuer voltages at the sockets - EVERY socket. Is there B+ of 500 volts or whatever at all pins 3? Is there a similar voltage at all pins 4? Is there about a +15VDC on all pins 5? Yes positive 15. That much is important.

This IS an ultralinear power stage, the 100 ohm screen resistors are only to half the tubes, they separate the screens on each pair of tubes so the screens can act independently in their own tube. They are a sort of decoupler I guess.

U5 is the phase inverter. SIgnal comes through one half for one phase and that signal in inverted for the other phase by the second half of the IC. Q5,Q7 are the cathode drivers. Q4,6 are just limiters.

There are DC voltage points on this schematic, check them. Make sure reference diodes CR28,29 are not open. And make the base voltage for those drivers right. The test volts across the 3 ohm emitter resistors of Q5,7 will only read with tubes in the sockets and runing. There is a note for proper readings. Check those two 3 ohm 5 watt resistors, they are next to the transistors. Measuring voltage across them checks tube current just as you would across a 1 ohm cathode resistor in apure tube amp. same idea.

We are looking for about +80v on the cathodes I think, and with +15 on the grid, that represents about 65v of bias. These things are running close to class B.

They are just transistors, so you can test them like any other. They should always be replaced in pairs. 6465 is a PV house number, they are really MJ4247, which are 8 amp 120v parts .

[reubster]

Many Thanks Enzo,

Just a quick update for those interested.
[Refering to the MX VTX schematic]

C44 -15V supply cap had leaked, both 15V supply caps replaced.
Transistors Q4 [BC547] and Q5 [2N3773] were found to be kaput, so these and their respective pairs Q6 & Q7 also replaced.

Very nice amp to work on, well laid out, easy to access and remove boards and good quality PCB.

So
DC reference voltages checked and found OK.
Sans tubes:,the socket voltages all check out ok.
With tubes [an unmatched, old & tired junk box selection] the Vi(Idle) is about 10mV [should be 20 - 40mv] and the Vk (Idle) is about 95Volts.

So new matched tubes and some tweaking should bring it all together.
Will also replace HT filter caps and swap all the rectifier diodes for Fast recoverys.

The power stage [signal via power in] actually sounds pretty good and probably great with new matched tubes.
However the preamp is not great.
The disapointment is the "normal" channel as it is just not very clean [yes I have removed CR3 & CR4].
I dont have a CRO but was wondering if this is how the circuit is intended and whether there is a fix.
I really want to push the OP with a clean signal.
Any suggestions greatly appreciated.

Cheers

[Enzo]

Well, I think that CR3,4 are in the dirt channel. Isn;t the clean channel through U1A?

WHen in the clean channel, do ANY of the lead control have any affect on the sound - they shouldn;t. If they do, I suspect Q1 or its control circuit.

U2A and all of U3 are common to both channels. Are they not in sockets? Check their output pins for unwanted DC or just sub them.

The clean side should be pretty darn clean

[reubster]

There is no bleed between the clean and dirt channel and the dirt's Pull-Bright, Lead Pre and Satu controls have no audible effect when the clean channel is switched in.

Yes CR3 & CR4 are in the dirt channel.....my mistake [typo].
Before my CRO died [had'nt been fired up for 20 years] I saw clipped waveforms at pin 3 of U3A. In my version of the the amp, I only had diodes: CR 13 & 14 [not 35 & 36]. In any case these were the ones I pulled and things improved.
At this point I lost my CRO trace, but I could still hear some mild distortion [like a tube screamer pedal set with a low distortion] at the Pre Amp out. Certainly not "darn clean"....... could it possibly be a leaky electro in the signal path?.

I do intend to replace all the OP-Amps with something better, and may also add filter caps to earth on the +/- rails for each IC.
Will also bypass signal path electros with polystyrene caps.
Quite tempted to replace the +/-15VDC zener regulator circuit with three terminal regulators too.
By the way what do CR9&10 do?

[Enzo]

CR13,14 are limiters. If the signal gets over about half a volt, they conduct, clipping it off. On later board versions they added CR35,36 to make that a full volt of signal before clipping. SO to update your amp, lift one end of CR13,14 and add two more cross connected parallel diodes in series with them as CR35,36.

The plain old vanilla 4558 op amps in there should sound pretty good. Fancier parts might add some subtle nuance or possibly lower noise floor, but I wouldn;t expect night and day myself.

No scope? Use a voltmeter and a signal tracer.

Can you hear the distortion on a sine wave signal? Measure the signal level with an AC voltmeter at the CR13 point.. How large is it? Make some other amplifier into a signal tracer. Make a test lead: run a cord from the input of said amp. Add a cap in series with the tip to block DC. Add a 100k-1meg resistor to ground on the amp side of the cap. Now use the free end of the cap to couple into the MX circuit and see how the signal sounds at each point.

CR9,10? The +27v rectifiers for the +15 rail? Are you sure those are the parts you question? That doesn;t fit the rest of your discussion somehow.


You can rebuild the LV power supplies if you want a project, but to me they are sufficient. There are only 5 ICs on this board, and the load doesn't vary much, so the zeners work well enough, and there are bypass caps on the 15v rails, look at U1A. More than that, I'd consider overkill.

You can try swapping in just any other good general purpose op amps for test, You might swap out C10, C11, but I'd suspect Q2 first. Maybe C27. But I would say this is a guitar amp, not a hifi amp. SO I expect pretty darn clean here.... as a guitar amp. Guitar amps are not designed as PA amps, not designed to be flat, so if you are expecting to have hifi clean ultimately, then the amp will need work.

[reubster]

Thanks again Enzo

I was refering to the "other" CR9 & CR10's on the schemetic: these are located in the feedback loop of U1A.

[Enzo]

They are clamps. The voltage across them cannot go over their junction potential. Same thing as CR3,4. You could lift them and see what you get. I don't see any necessity for them in DC world. or you could add another pair in series as they did at CR35.36, so the threshold is doubled.

[icefloe01]

My son just got one of these amps tonight with a similar problem. One tube would get a nice bright blue glow and before I could turn it off pop went the fuse. For his it's Q6 and Q7 that died. I saw in reubster's post that he said Q4 and Q5 were BC547 and 2N3773. Neither one of those numbers match what are in this one. Q4 and 6 are MPS6531 while Q5 and 7 are Motorola 6465's. I can find the MPS's all day long but the 6465's completely elude me. The MPS6531 seems to have some better ratings on it except the Vceo is 40 instead of 45 so I think I'll stick with those. Does anyone have any info on the 6465's to see how they compare to the BC547's? If I can use those in place of the 6465's that would make things a little easier and he's chomping at the bit for me get this amp revived... LOL Any help would be appreciated.

Thank you,

Eric.

[reubster]

reply

Hi Eric,
I cant specifically give you that transister data, hopefully someone in the MEF brains trust will be able to assist.

Quoting from Enzo's first reply: 6465 is a PV house number, they are really MJ4247

When I got mine, Q5 & Q7 were not 6465/4247's [as per schema] but some other common mosfet [I cannot recall what].
I replaced what was there with the same type and I struggled to get the right biasing voltages around the valves.

I stuffed around for ages, eventually, a friend gave me some 4247's, I dropped these in and voilla all was good.

So the "beta gain" [I think this is the right term] of this transistor is important...you need to go with MJ4247

BTW I can recommend removing 2 of the OP tubes and putting the speaker onto the 2 ohm tap....................its still very loud, but at least you get some OP tube breakup.

Good luck


,

[icefloe01]

Hi Eric,
Quoting from Enzo's first reply: 6465 is a PV house number, they are really MJ4247

DOH! That's what I get for quick skimming... I grabbed the schematic from the page where Enzo uploaded it. Thank you to Enzo for that. I have a Mace 300 and a VTM60/120 that I need to figure out how to scan in (I guess Kinko's) if anyone would like them. I noticed he also had a pdf of Peavey parts with field replacements on the ampix site that listed the TO-3's as obsolete.

I can't find any spec's for the 4247's except for crossing it to NTE (blah). They sub a 12A 140V part with an hfe of 40-140 with a Vce of 5 and an Ic of 2A. A 2n3773 is rated at hfe 15 with 4V and 8A and hfe 5 at 16A. If I had the spec's on the original, there might be a shot at a circuit mod to make something else work.

I did find www.tubesandmore.com that lists them on their site for $8.70 ea. I hope they actually have some stock on them...

Another problem I had was extracting the control board. Two of the pull knobs had been glued onto their shafts. The "high/pull thick" came loose with a little gentle work but with the "normal gain/pre-pull bright" the shaft pulled out of the pot. It takes a little bit of pressure to get it fully snapped back in so that it still works the pot with the switch pulled, but I think it will be okay. I found it at http://www.vibroworld.com/parts/tech17.html but they don't have the 10K and say to use the 50K version instead. It's been a few years (a decade or two) since I've had the luxury of keeping my head wrapped around audio electronics, so my knowledge of feedback circuits has gotten a bit rusty. That being said, if I need to replace that pot, won't putting in a 50K instead of a 10K cause Q2 to be overdriven more easily or will it not matter as much since it's not a change of a complete order of magnitude? Like I said, I'm a bit rusty here...

Thank you,

Eric.

[icefloe01]

Fortunately, tubesandmore had severeal of the originals left in stock so I ordered four of each.

[Enzo]

MJ4247 was a Motorola part - now On Semiconductor - a TO3 rated 8A 120V 90W hfe40min @3A.

Search this forum for a Peavey Heritage we recently have discussed. It had drive transistor problems too. We discussed subs.

This amp is a Peavey product. WHile you can look around to other sources, the best bet in my mind is to just call the factory parts department. They have all the schematics, and they will have the exact parts. At least if anyone does.

[icefloe01]

MJ4247 was a Motorola part - now On Semiconductor - a TO3 rated 8A 120V 90W hfe40min @3A.

Search this forum for a Peavey Heritage we recently have discussed. It had drive transistor problems too. We discussed subs.

This amp is a Peavey product. While you can look around to other sources, the best bet in my mind is to just call the factory parts department. They have all the schematics, and they will have the exact parts. At least if anyone does.

Thank youfor the info and another potential parts source! I got the parts in from tubesandmore and installed them. Unfortunately, due to not having enough space for a dedicated shop and accidents happen, the reverb pot somehow ended broken in the process. Fortunately it doesn't affect anything and he didn't care as he hates reverb. Now he has a nice LOUD head. He also got a Hughs and Kettner stereo/mono 412 cab with it that someone replaced all the 8 Ohm speakers with 16 Ohm but left the wiring the same. So I had to rewire that too so he would have a proper match to his head. He got everything for about $210 including parts. Luckily for him I knew how to fix it. Now I just have to find my can of Lube-A-Trol and tear into my Mace...