Markbass Little Mark II

[Alex R]

Guy came through the door today with one of these Italian amps. First one I've seen. I said, oh, right, it's a preamp, thought you said 500 watts. He said, no it's specced at 500 watts into 4 ohms. And it is. Tiny and light (and silent - the fan runs but the protection relay is in).

I took a look inside - no big mains transformer. The mains seems to go to two switching mosfets (which have both shorted), and the supply goes after filtering to the output stage from there. Then there is a little transformer for the smaller devices.

Well I will change the mosfets and cross my fingers, as most of the rest of it is surface mount. The output devices are not shorted.

Any comments on this design? Any experience with repairs? If it works it is likely to catch on, at least in Europe, as this is an astonishingly light amp! But can it be safe to do away with the isolating effect of a mains transformer?

[Robert M. Martinelli]

Hi Alex,
I currently have one under repair, and, having inspected it pretty thoroughly, I guess one of the reasons they're so light is the neodymium speaker (er...magnet), the other reason is the absence of a mains transformer.

The one I have here is just fine, only the speaker's coil has "gone south", seems like the output devices are rated to withstand the voltage rise due to an open load....

Though they seem well built, and they have an extremely favorable power vs. weight ratio, I don't like the absence of a transformer and the subsequent lack of insulation from mains. Looks like they revived some of the concepts used by Gallien/Krueger back in the 80s....those amps were extremely compact, but their reliability was rather poor.

This said, I don't want to judge them prematurely, I think they've been on the market for too short a time to have a dependable reliability record.....I guess time will tell ( My 44 years old AC30 is still going strong, but that's really from another era ), let's wait and see....
Regards
Bob

[52 Bill]

Welcome to the wonderful world of switching power supplies!

I have not seen one of those models, but have seen my share of switching power supplies. They are being used everywhere now-a-days. I've got a Peavey CS-600 on the bench right now that has one. The first time I ever saw one was nearly twenty years ago, in a Walter Woods head. Now every company is using them, especially in high power bass and PA heads.

[Steve Conner]

All of the audio stuff, input jack, speaker jack, power amp, whatever, is isolated from the mains. It's only part of the power supply (the two big MOSFETs) that is live.

Basically how it works is, the mains gets rectified to about 300 volts DC, and smoothed by a few decent-sized electrolytics, this then gets chopped up at high frequency by the two big FETs, and the resulting AC is stepped down though a small ferrite-cored transformer. This is then rectified and smoothed again, to make the DC rails for the power amp. Just like a PC power supply, the only difference is the choice of output voltages.

If you can't see any decent-sized heatsinks, the power amp is more than likely a Class-D, which works by chopping the DC rails up again (using yet more MOSFETs) into a high-frequency pulse train that approximates the audio signal. This gets reconstituted by a low-pass filter before feeding out the speaker jack.

The ferrite-cored transformer provides isolation from the line in just the same way as the iron-cored transformer in an old-fashioned power amp. It might only be about 1.5" cubed, but it can still pass those 500+ watts no problem.

I hate to sound cynical beyond my years, but the MOSFETs don't often blow by themselves. More commonly, something else goes wrong first and takes them out, or they take a bunch of other stuff with them when they go. I think out of all the SMPS I've mended, there was maybe one that had a blown MOSFET and putting a new one in was all it took to make it work again.

steve

[Robert M. Martinelli]

Well,
As I told already I've seen switching power supplies in (bass) amps before, back in the 80s I remember a friend of mine having more than a headache with his GK ( seems to me it was an ML250 but I'm not sure ). Those amps surely had some design flaw, i.e. they were designed with size reduction in mind, so they were prone to thermal breakdowns due to insufficient heat removal.

I don't want to look romantic, or old-fashioned ( and well, maybe I am ) but, working as an automation engineer, I often deal with Switching PS and PWM inverters and drives, and yes, switching PS are more efficient, but the idea of seeing a switching PS inside an amp makes me sad nonetheless.....Oh, well, I guess I can get used to the idea as long as we're talking 'bout bass amps.....and I know we can't stop progress, but I fear a day will come we'll see a switching PS inside an AC30, a JCM3000 or a Bassman
Regards
(sad) Bob

[Alex R]

No, there are big heatsinks in there, I think the power amp isn't too strange.

The power supply mosfets are on a slab of aluminium.

Thanks for your explanation about that ferrite core transfomer, I just didn't think all that power could be supplied through that little thing. I've never know how SMPSs work and i'm now a little nearer... a couple of questions to try your patience:

Is the mains usually rectified by a normal-type bridge?

The switching chops up the resulting 300vDC so this makes the transformer work - what's the aim of this? How come so small a transformer can be used?

The weight difference from a standard power supply is quite dramatic (this is the head, not the combo).

All the diodes and transistors around the mosfets gave me a beep ok so I'm crossing my fingers really hard... but that's a reason why I kept the customer here till I got the mosfets off the board, confirmed they were shorted, and sniffed around for anything else gone, so he understood that he might have to pay out to get this work done and still have no amp, if anything surface-mount has gone too. Because I don't have the gear for SMDs - it was good to be able to show him some so he could understand the difficulty. Is anyone taking SMD work on yet? I guess we will eventually have to...

I suppose the transformer has to be a suspect?

[Robert M. Martinelli]

The switching chops up the resulting 300vDC so this makes the transformer work - what's the aim of this? How come so small a transformer can be used?

In a nutshell ( and I'm sure Steve can explain this way better and more thoroughly than me):
Because of the very high switching (carrier) frequency, the ferrite core and the higher input voltage (less current at the primary for a given power); all these factors together improve transfer efficiency, the losses in the "iron" ( er....ferrite) and in the windings are close to zero, all these factors combined together allow for a smaller transformer if compared to conventional (linear) PSs.
Regards
Bob

[Steve Conner]

It's just a fact that transformers can be made smaller without loss of efficiency, as the frequency of the AC they're transforming gets higher. That's why they use 400Hz on board aircraft.

We use 50Hz over here, and all our transformers are slightly bigger for the same VA rating than American 60Hz ones.

Some parts of Canada used to use 25Hz, and the tube radios sold there had a PT twice the size. I think the radio maker kept the same laminations but used a stack twice as tall, the result was a freakish looking PT.

These switchmode power supplies run at something like 50 or 100kHz. The ferrite core is necessary because regular iron laminations would get red hot from losses at these frequencies.

The transformer has far fewer turns of thicker wire than a line-frequency type, so it's unlikely to have burnt out.

The mains is rectified by a normal 4-diode bridge in 220V countries, and sometimes a 2-diode doubler in 120V land. Some newer designs have an active power factor corrector in there too.

As we used to say in the industry, "Behind every successful SMPS design there's a lot of chicken blood"

[Alex R]

As we used to say in the industry, "Behind every successful SMPS design there's a lot of chicken blood"

- don't have a chicken to hand as I used them all up making bad medicine for Scottish Power the last time they overcharged us... way more effective than ringing the call centre. Do you think it might have the same effect if I waved a cooked turkey wing over the amp before I try to fix it?

So a switch mode power supply is about creating higher-frequency AC with rapid-switching devices so that you can use a smaller transformer? Can it be that simple and I failed to understand it for so long? (story of my life...)

[Robert M. Martinelli]

So a switch mode power supply is about creating higher-frequency AC with rapid-switching devices so that you can use a smaller transformer? Can it be that simple and I failed to understand it for so long? (story of my life...)

More or less, yes, the other part is about how voltage gets regulated, and involves a technology called PWM ( Pulse Width Modulation ) : in ( another ) nutshell, the duty cycle of the waveform is made variable, e.g. you can get minimum voltage with a 10% - 90% duty cycle ( 10 % of the period ON - 90% of the period OFF ) and maximum voltage with a 90% - 10% duty cycle ( 90% of the period ON and 10% of the period OFF ). The filtered voltage is an average, then a voltage error signal is fed back to control the duty cycle to have the output voltage match the desired one.
Regards
Bob

[bma1984]

Sorry for digging up an old post, but this seems like an appropriate thread to ask this question. I'm attempting to repair the two switching mosfets in a Markbass Little Mark Tube 800. The mosfet part # is IRFP27N60KPBF-ND. I tried to order two from digikey but the supplier has recently discontinued the chip. Just wondering if anyone could suggest a direct replacement? Here is the digikey page for the chip if that helps: MOSFET N-CH 600V 27A TO-247AC - IRFP27N60KPBF

Thanks

[J M Fahey]

That part is still certainly being made, but MosFet naming conventions changed -.
In the old times "27"N"60" meant "27 (Ampere)" N "60(0V)" and so on, while that part name will now be IRFP "somenumber" as in IRFP450 or whatever.
Search International Rectifier site to find a TO247 case, 27 ampere 600V part, whatever they call it now, and that's it.

[bma1984]

Thank you so much for the advice. When I search for that part, I can't seem to find TO247 with Vdss of 600V. I'll keep looking since you've got me on the right track. If you find something, please let me know. I'm so excited to get this amp going.

Thanks,

[bma1984]

This is pretty close: STW28NK60Z STMicroelectronics MOSFET Power
Would you recommend this, even though max power dissipation is less and Rds is less? Sorry for my lack of knowledge in this area. Thanks again!

[J M Fahey]

It will surely do.
350W dissipation is monster, and, besides, nominal. It will certainly dissipate around 30W at most, and that because of switching losses.
Voltage and current match. Lower Rds is a "better spec".
Looks good.
Buy 4 , cost is nil.
For what it's worth, *I* would use those parts.
EDIT: I forgot, check everything, *something* killed the earlier MosFets.
See if you can power that supply without load (the amplifier disconnected) ; another problem is that switchers many times don't like variacs or bulb limiters, so you will have to take your chances by connecting it straight to the power line.